heb-hebr-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407/checkpoint-9360/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.989855846235985,
  "eval_steps": 500,
  "global_step": 9360,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.68021821975708,
      "epoch": 0.005339028296849973,
      "grad_norm": 31.625,
      "learning_rate": 2e-06,
      "loss": 13.9711,
      "mean_token_accuracy": 0.009074668399989605,
      "num_tokens": 10658.0,
      "step": 5
    },
    {
      "entropy": 4.704092264175415,
      "epoch": 0.010678056593699947,
      "grad_norm": 32.75,
      "learning_rate": 4.5e-06,
      "loss": 13.9298,
      "mean_token_accuracy": 0.00866024480201304,
      "num_tokens": 21327.0,
      "step": 10
    },
    {
      "entropy": 4.7672216415405275,
      "epoch": 0.01601708489054992,
      "grad_norm": 36.5,
      "learning_rate": 7e-06,
      "loss": 13.7823,
      "mean_token_accuracy": 0.007704685814678669,
      "num_tokens": 32403.0,
      "step": 15
    },
    {
      "entropy": 4.958004379272461,
      "epoch": 0.021356113187399894,
      "grad_norm": 40.5,
      "learning_rate": 9.5e-06,
      "loss": 13.3459,
      "mean_token_accuracy": 0.006649625021964312,
      "num_tokens": 43471.0,
      "step": 20
    },
    {
      "entropy": 5.477138376235962,
      "epoch": 0.026695141484249868,
      "grad_norm": 38.75,
      "learning_rate": 1.2e-05,
      "loss": 12.4632,
      "mean_token_accuracy": 0.0020860590098891406,
      "num_tokens": 53724.0,
      "step": 25
    },
    {
      "entropy": 6.8245813846588135,
      "epoch": 0.03203416978109984,
      "grad_norm": 30.375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.6429,
      "mean_token_accuracy": 0.0007934643130283803,
      "num_tokens": 65115.0,
      "step": 30
    },
    {
      "entropy": 10.22093849182129,
      "epoch": 0.03737319807794982,
      "grad_norm": 5.46875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.8521,
      "mean_token_accuracy": 0.0002344618726056069,
      "num_tokens": 75902.0,
      "step": 35
    },
    {
      "entropy": 10.684853076934814,
      "epoch": 0.04271222637479979,
      "grad_norm": 4.0625,
      "learning_rate": 1.95e-05,
      "loss": 10.7564,
      "mean_token_accuracy": 0.0011873395298607648,
      "num_tokens": 87051.0,
      "step": 40
    },
    {
      "entropy": 10.69392204284668,
      "epoch": 0.04805125467164976,
      "grad_norm": 2.84375,
      "learning_rate": 2.2e-05,
      "loss": 10.5885,
      "mean_token_accuracy": 0.006381610780954361,
      "num_tokens": 98998.0,
      "step": 45
    },
    {
      "entropy": 10.600316715240478,
      "epoch": 0.053390282968499736,
      "grad_norm": 2.34375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.4445,
      "mean_token_accuracy": 0.011505381856113672,
      "num_tokens": 111207.0,
      "step": 50
    },
    {
      "entropy": 10.673304462432862,
      "epoch": 0.05872931126534971,
      "grad_norm": 2.0625,
      "learning_rate": 2.7e-05,
      "loss": 10.2638,
      "mean_token_accuracy": 0.016619019117206334,
      "num_tokens": 121039.0,
      "step": 55
    },
    {
      "entropy": 10.670036125183106,
      "epoch": 0.06406833956219968,
      "grad_norm": 2.03125,
      "learning_rate": 2.95e-05,
      "loss": 10.1533,
      "mean_token_accuracy": 0.015064884908497334,
      "num_tokens": 132604.0,
      "step": 60
    },
    {
      "entropy": 10.671291160583497,
      "epoch": 0.06940736785904965,
      "grad_norm": 1.7734375,
      "learning_rate": 3.2e-05,
      "loss": 10.0181,
      "mean_token_accuracy": 0.03150726184248924,
      "num_tokens": 142825.0,
      "step": 65
    },
    {
      "entropy": 10.688156223297119,
      "epoch": 0.07474639615589963,
      "grad_norm": 1.78125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.9346,
      "mean_token_accuracy": 0.056240572780370715,
      "num_tokens": 153121.0,
      "step": 70
    },
    {
      "entropy": 10.670577144622802,
      "epoch": 0.0800854244527496,
      "grad_norm": 1.7421875,
      "learning_rate": 3.7e-05,
      "loss": 9.8809,
      "mean_token_accuracy": 0.05582823716104031,
      "num_tokens": 164675.0,
      "step": 75
    },
    {
      "entropy": 10.664248943328857,
      "epoch": 0.08542445274959957,
      "grad_norm": 1.5703125,
      "learning_rate": 3.95e-05,
      "loss": 9.8352,
      "mean_token_accuracy": 0.04962811656296253,
      "num_tokens": 175146.0,
      "step": 80
    },
    {
      "entropy": 10.648977661132813,
      "epoch": 0.09076348104644955,
      "grad_norm": 1.4765625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.8124,
      "mean_token_accuracy": 0.04548483826220036,
      "num_tokens": 186177.0,
      "step": 85
    },
    {
      "entropy": 10.642164325714111,
      "epoch": 0.09610250934329952,
      "grad_norm": 1.6640625,
      "learning_rate": 4.45e-05,
      "loss": 9.685,
      "mean_token_accuracy": 0.048678891733288765,
      "num_tokens": 197143.0,
      "step": 90
    },
    {
      "entropy": 10.636518096923828,
      "epoch": 0.10144153764014949,
      "grad_norm": 1.625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.6182,
      "mean_token_accuracy": 0.052110004052519796,
      "num_tokens": 207979.0,
      "step": 95
    },
    {
      "entropy": 10.610146522521973,
      "epoch": 0.10678056593699947,
      "grad_norm": 1.640625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.5135,
      "mean_token_accuracy": 0.0568073820322752,
      "num_tokens": 218733.0,
      "step": 100
    },
    {
      "entropy": 10.569573020935058,
      "epoch": 0.11211959423384944,
      "grad_norm": 1.625,
      "learning_rate": 5.2e-05,
      "loss": 9.4144,
      "mean_token_accuracy": 0.05693732798099518,
      "num_tokens": 228887.0,
      "step": 105
    },
    {
      "entropy": 10.517056083679199,
      "epoch": 0.11745862253069941,
      "grad_norm": 1.5390625,
      "learning_rate": 5.45e-05,
      "loss": 9.3275,
      "mean_token_accuracy": 0.05374570451676845,
      "num_tokens": 240289.0,
      "step": 110
    },
    {
      "entropy": 10.480204105377197,
      "epoch": 0.12279765082754938,
      "grad_norm": 1.4609375,
      "learning_rate": 5.7e-05,
      "loss": 9.2538,
      "mean_token_accuracy": 0.05631314069032669,
      "num_tokens": 250548.0,
      "step": 115
    },
    {
      "entropy": 10.422281837463379,
      "epoch": 0.12813667912439936,
      "grad_norm": 1.4765625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 9.1616,
      "mean_token_accuracy": 0.05645494796335697,
      "num_tokens": 261669.0,
      "step": 120
    },
    {
      "entropy": 10.368197917938232,
      "epoch": 0.13347570742124934,
      "grad_norm": 1.359375,
      "learning_rate": 6.2e-05,
      "loss": 9.0945,
      "mean_token_accuracy": 0.05810702405869961,
      "num_tokens": 272926.0,
      "step": 125
    },
    {
      "entropy": 10.291021633148194,
      "epoch": 0.1388147357180993,
      "grad_norm": 1.2734375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 9.0261,
      "mean_token_accuracy": 0.05110162906348705,
      "num_tokens": 284173.0,
      "step": 130
    },
    {
      "entropy": 10.235691547393799,
      "epoch": 0.14415376401494928,
      "grad_norm": 1.1953125,
      "learning_rate": 6.7e-05,
      "loss": 8.9153,
      "mean_token_accuracy": 0.05686397515237331,
      "num_tokens": 294850.0,
      "step": 135
    },
    {
      "entropy": 10.190592193603516,
      "epoch": 0.14949279231179927,
      "grad_norm": 1.203125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.8582,
      "mean_token_accuracy": 0.05346960201859474,
      "num_tokens": 305727.0,
      "step": 140
    },
    {
      "entropy": 10.10048542022705,
      "epoch": 0.15483182060864922,
      "grad_norm": 1.3125,
      "learning_rate": 7.2e-05,
      "loss": 8.7462,
      "mean_token_accuracy": 0.058417992666363716,
      "num_tokens": 317243.0,
      "step": 145
    },
    {
      "entropy": 9.970914077758788,
      "epoch": 0.1601708489054992,
      "grad_norm": 1.015625,
      "learning_rate": 7.45e-05,
      "loss": 8.7387,
      "mean_token_accuracy": 0.05338601246476173,
      "num_tokens": 328334.0,
      "step": 150
    },
    {
      "entropy": 9.846445274353027,
      "epoch": 0.16550987720234916,
      "grad_norm": 0.96484375,
      "learning_rate": 7.7e-05,
      "loss": 8.6482,
      "mean_token_accuracy": 0.054720787703990935,
      "num_tokens": 339774.0,
      "step": 155
    },
    {
      "entropy": 9.73973569869995,
      "epoch": 0.17084890549919915,
      "grad_norm": 0.96875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.5543,
      "mean_token_accuracy": 0.05573372282087803,
      "num_tokens": 350211.0,
      "step": 160
    },
    {
      "entropy": 9.550614070892333,
      "epoch": 0.1761879337960491,
      "grad_norm": 0.78125,
      "learning_rate": 8.2e-05,
      "loss": 8.4684,
      "mean_token_accuracy": 0.054642071947455406,
      "num_tokens": 360109.0,
      "step": 165
    },
    {
      "entropy": 9.380690956115723,
      "epoch": 0.1815269620928991,
      "grad_norm": 0.8359375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.4131,
      "mean_token_accuracy": 0.054827965795993805,
      "num_tokens": 370911.0,
      "step": 170
    },
    {
      "entropy": 9.282039070129395,
      "epoch": 0.18686599038974908,
      "grad_norm": 0.76953125,
      "learning_rate": 8.7e-05,
      "loss": 8.3991,
      "mean_token_accuracy": 0.060196897760033605,
      "num_tokens": 381808.0,
      "step": 175
    },
    {
      "entropy": 9.13407211303711,
      "epoch": 0.19220501868659903,
      "grad_norm": 0.86328125,
      "learning_rate": 8.95e-05,
      "loss": 8.3091,
      "mean_token_accuracy": 0.06106638610363006,
      "num_tokens": 392175.0,
      "step": 180
    },
    {
      "entropy": 9.000042152404784,
      "epoch": 0.19754404698344902,
      "grad_norm": 0.78125,
      "learning_rate": 9.2e-05,
      "loss": 8.3874,
      "mean_token_accuracy": 0.049684269726276396,
      "num_tokens": 403261.0,
      "step": 185
    },
    {
      "entropy": 8.931573295593262,
      "epoch": 0.20288307528029897,
      "grad_norm": 0.7265625,
      "learning_rate": 9.45e-05,
      "loss": 8.2853,
      "mean_token_accuracy": 0.05462104417383671,
      "num_tokens": 414992.0,
      "step": 190
    },
    {
      "entropy": 8.791546726226807,
      "epoch": 0.20822210357714896,
      "grad_norm": 0.6328125,
      "learning_rate": 9.7e-05,
      "loss": 8.2732,
      "mean_token_accuracy": 0.05595530979335308,
      "num_tokens": 425316.0,
      "step": 195
    },
    {
      "entropy": 8.729821968078614,
      "epoch": 0.21356113187399894,
      "grad_norm": 0.75,
      "learning_rate": 9.95e-05,
      "loss": 8.344,
      "mean_token_accuracy": 0.05437092892825603,
      "num_tokens": 437759.0,
      "step": 200
    },
    {
      "entropy": 8.712938117980958,
      "epoch": 0.2189001601708489,
      "grad_norm": 0.77734375,
      "learning_rate": 0.000102,
      "loss": 8.2844,
      "mean_token_accuracy": 0.05534115061163902,
      "num_tokens": 448066.0,
      "step": 205
    },
    {
      "entropy": 8.651909160614014,
      "epoch": 0.22423918846769889,
      "grad_norm": 1.109375,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.2206,
      "mean_token_accuracy": 0.055615832656621934,
      "num_tokens": 458632.0,
      "step": 210
    },
    {
      "entropy": 8.63439655303955,
      "epoch": 0.22957821676454884,
      "grad_norm": 0.76171875,
      "learning_rate": 0.000107,
      "loss": 8.2606,
      "mean_token_accuracy": 0.054173830151557925,
      "num_tokens": 469224.0,
      "step": 215
    },
    {
      "entropy": 8.596088218688966,
      "epoch": 0.23491724506139883,
      "grad_norm": 0.875,
      "learning_rate": 0.0001095,
      "loss": 8.2823,
      "mean_token_accuracy": 0.05661116987466812,
      "num_tokens": 480492.0,
      "step": 220
    },
    {
      "entropy": 8.625334930419921,
      "epoch": 0.2402562733582488,
      "grad_norm": 0.921875,
      "learning_rate": 0.000112,
      "loss": 8.2398,
      "mean_token_accuracy": 0.057639367505908014,
      "num_tokens": 491325.0,
      "step": 225
    },
    {
      "entropy": 8.569988918304443,
      "epoch": 0.24559530165509877,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001145,
      "loss": 8.2465,
      "mean_token_accuracy": 0.0603976845741272,
      "num_tokens": 502240.0,
      "step": 230
    },
    {
      "entropy": 8.592293834686279,
      "epoch": 0.25093432995194875,
      "grad_norm": 0.84375,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.2035,
      "mean_token_accuracy": 0.06263627484440804,
      "num_tokens": 512710.0,
      "step": 235
    },
    {
      "entropy": 8.58193883895874,
      "epoch": 0.2562733582487987,
      "grad_norm": 0.71484375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.2513,
      "mean_token_accuracy": 0.05789706408977509,
      "num_tokens": 524023.0,
      "step": 240
    },
    {
      "entropy": 8.547989177703858,
      "epoch": 0.26161238654564867,
      "grad_norm": 3.234375,
      "learning_rate": 0.000122,
      "loss": 8.1422,
      "mean_token_accuracy": 0.06462252773344516,
      "num_tokens": 534612.0,
      "step": 245
    },
    {
      "entropy": 8.475037765502929,
      "epoch": 0.2669514148424987,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0001245,
      "loss": 8.2036,
      "mean_token_accuracy": 0.05881951674818993,
      "num_tokens": 544866.0,
      "step": 250
    },
    {
      "entropy": 8.527704906463622,
      "epoch": 0.27229044313934864,
      "grad_norm": 0.84375,
      "learning_rate": 0.000127,
      "loss": 8.2523,
      "mean_token_accuracy": 0.055548783764243125,
      "num_tokens": 556091.0,
      "step": 255
    },
    {
      "entropy": 8.524747467041015,
      "epoch": 0.2776294714361986,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0001295,
      "loss": 8.2536,
      "mean_token_accuracy": 0.05479630120098591,
      "num_tokens": 566946.0,
      "step": 260
    },
    {
      "entropy": 8.54754581451416,
      "epoch": 0.2829684997330486,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000132,
      "loss": 8.1371,
      "mean_token_accuracy": 0.060486416518688205,
      "num_tokens": 577164.0,
      "step": 265
    },
    {
      "entropy": 8.525337314605713,
      "epoch": 0.28830752802989856,
      "grad_norm": 0.671875,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.1768,
      "mean_token_accuracy": 0.05711319632828236,
      "num_tokens": 588744.0,
      "step": 270
    },
    {
      "entropy": 8.505339241027832,
      "epoch": 0.2936465563267485,
      "grad_norm": 0.921875,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.1727,
      "mean_token_accuracy": 0.055232659354805945,
      "num_tokens": 600943.0,
      "step": 275
    },
    {
      "entropy": 8.492932891845703,
      "epoch": 0.29898558462359853,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0001395,
      "loss": 8.1726,
      "mean_token_accuracy": 0.06307518370449543,
      "num_tokens": 611864.0,
      "step": 280
    },
    {
      "entropy": 8.451044940948487,
      "epoch": 0.3043246129204485,
      "grad_norm": 0.75,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.152,
      "mean_token_accuracy": 0.05931071788072586,
      "num_tokens": 623314.0,
      "step": 285
    },
    {
      "entropy": 8.428306388854981,
      "epoch": 0.30966364121729845,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0001445,
      "loss": 8.1763,
      "mean_token_accuracy": 0.06025952883064747,
      "num_tokens": 633509.0,
      "step": 290
    },
    {
      "entropy": 8.451824760437011,
      "epoch": 0.3150026695141484,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000147,
      "loss": 8.0983,
      "mean_token_accuracy": 0.06582859642803669,
      "num_tokens": 644655.0,
      "step": 295
    },
    {
      "entropy": 8.374590492248535,
      "epoch": 0.3203416978109984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001495,
      "loss": 8.135,
      "mean_token_accuracy": 0.05644083172082901,
      "num_tokens": 655940.0,
      "step": 300
    },
    {
      "entropy": 8.408124351501465,
      "epoch": 0.3256807261078484,
      "grad_norm": 0.765625,
      "learning_rate": 0.000152,
      "loss": 8.1809,
      "mean_token_accuracy": 0.056555739417672155,
      "num_tokens": 667074.0,
      "step": 305
    },
    {
      "entropy": 8.498903751373291,
      "epoch": 0.33101975440469833,
      "grad_norm": 0.703125,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.2309,
      "mean_token_accuracy": 0.06025502309203148,
      "num_tokens": 677942.0,
      "step": 310
    },
    {
      "entropy": 8.455632591247559,
      "epoch": 0.33635878270154834,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000157,
      "loss": 8.1642,
      "mean_token_accuracy": 0.06168182455003261,
      "num_tokens": 689226.0,
      "step": 315
    },
    {
      "entropy": 8.403513431549072,
      "epoch": 0.3416978109983983,
      "grad_norm": 0.76171875,
      "learning_rate": 0.0001595,
      "loss": 8.1786,
      "mean_token_accuracy": 0.059801208600401876,
      "num_tokens": 700830.0,
      "step": 320
    },
    {
      "entropy": 8.456379795074463,
      "epoch": 0.34703683929524826,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000162,
      "loss": 8.1377,
      "mean_token_accuracy": 0.06001129299402237,
      "num_tokens": 712677.0,
      "step": 325
    },
    {
      "entropy": 8.363616943359375,
      "epoch": 0.3523758675920982,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.2395,
      "mean_token_accuracy": 0.05579938478767872,
      "num_tokens": 723833.0,
      "step": 330
    },
    {
      "entropy": 8.497436428070069,
      "epoch": 0.3577148958889482,
      "grad_norm": 0.828125,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.1771,
      "mean_token_accuracy": 0.059333324804902075,
      "num_tokens": 735526.0,
      "step": 335
    },
    {
      "entropy": 8.438710403442382,
      "epoch": 0.3630539241857982,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00016950000000000003,
      "loss": 8.1223,
      "mean_token_accuracy": 0.06136825978755951,
      "num_tokens": 745934.0,
      "step": 340
    },
    {
      "entropy": 8.391829586029052,
      "epoch": 0.36839295248264814,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00017199999999999998,
      "loss": 8.0837,
      "mean_token_accuracy": 0.06251248680055141,
      "num_tokens": 756693.0,
      "step": 345
    },
    {
      "entropy": 8.393558979034424,
      "epoch": 0.37373198077949815,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00017449999999999999,
      "loss": 8.1418,
      "mean_token_accuracy": 0.05746800377964974,
      "num_tokens": 767236.0,
      "step": 350
    },
    {
      "entropy": 8.491841030120849,
      "epoch": 0.3790710090763481,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000177,
      "loss": 8.1254,
      "mean_token_accuracy": 0.06065751984715462,
      "num_tokens": 778552.0,
      "step": 355
    },
    {
      "entropy": 8.428674793243408,
      "epoch": 0.38441003737319807,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001795,
      "loss": 8.1788,
      "mean_token_accuracy": 0.05917464569211006,
      "num_tokens": 789615.0,
      "step": 360
    },
    {
      "entropy": 8.48459882736206,
      "epoch": 0.3897490656700481,
      "grad_norm": 0.8125,
      "learning_rate": 0.000182,
      "loss": 8.1355,
      "mean_token_accuracy": 0.06367055512964725,
      "num_tokens": 799998.0,
      "step": 365
    },
    {
      "entropy": 8.322556686401366,
      "epoch": 0.39508809396689804,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001845,
      "loss": 8.0679,
      "mean_token_accuracy": 0.062298642471432686,
      "num_tokens": 810042.0,
      "step": 370
    },
    {
      "entropy": 8.422898483276366,
      "epoch": 0.400427122263748,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000187,
      "loss": 8.0755,
      "mean_token_accuracy": 0.05906784273684025,
      "num_tokens": 821692.0,
      "step": 375
    },
    {
      "entropy": 8.36357364654541,
      "epoch": 0.40576615056059795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001895,
      "loss": 8.1164,
      "mean_token_accuracy": 0.06748287416994572,
      "num_tokens": 832262.0,
      "step": 380
    },
    {
      "entropy": 8.333745193481445,
      "epoch": 0.41110517885744796,
      "grad_norm": 0.8125,
      "learning_rate": 0.000192,
      "loss": 8.04,
      "mean_token_accuracy": 0.06151870898902416,
      "num_tokens": 843006.0,
      "step": 385
    },
    {
      "entropy": 8.353145503997803,
      "epoch": 0.4164442071542979,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0001945,
      "loss": 8.0332,
      "mean_token_accuracy": 0.062286530062556265,
      "num_tokens": 853978.0,
      "step": 390
    },
    {
      "entropy": 8.357436656951904,
      "epoch": 0.4217832354511479,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019700000000000002,
      "loss": 8.0743,
      "mean_token_accuracy": 0.05977092795073986,
      "num_tokens": 864634.0,
      "step": 395
    },
    {
      "entropy": 8.423241233825683,
      "epoch": 0.4271222637479979,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00019950000000000002,
      "loss": 8.0568,
      "mean_token_accuracy": 0.06662314645946026,
      "num_tokens": 875379.0,
      "step": 400
    },
    {
      "entropy": 8.398690319061279,
      "epoch": 0.43246129204484784,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000202,
      "loss": 8.0674,
      "mean_token_accuracy": 0.06020735241472721,
      "num_tokens": 886428.0,
      "step": 405
    },
    {
      "entropy": 8.425256729125977,
      "epoch": 0.4378003203416978,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00020449999999999998,
      "loss": 8.0827,
      "mean_token_accuracy": 0.0645828690379858,
      "num_tokens": 896700.0,
      "step": 410
    },
    {
      "entropy": 8.369815444946289,
      "epoch": 0.44313934863854776,
      "grad_norm": 0.8125,
      "learning_rate": 0.000207,
      "loss": 8.0577,
      "mean_token_accuracy": 0.06499057859182358,
      "num_tokens": 907478.0,
      "step": 415
    },
    {
      "entropy": 8.39303207397461,
      "epoch": 0.44847837693539777,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002095,
      "loss": 7.9707,
      "mean_token_accuracy": 0.07010989226400852,
      "num_tokens": 918170.0,
      "step": 420
    },
    {
      "entropy": 8.28305778503418,
      "epoch": 0.45381740523224773,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000212,
      "loss": 8.0345,
      "mean_token_accuracy": 0.06685064136981964,
      "num_tokens": 928866.0,
      "step": 425
    },
    {
      "entropy": 8.279892349243164,
      "epoch": 0.4591564335290977,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0002145,
      "loss": 8.0205,
      "mean_token_accuracy": 0.06419647410511971,
      "num_tokens": 940395.0,
      "step": 430
    },
    {
      "entropy": 8.475063610076905,
      "epoch": 0.4644954618259477,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00021700000000000002,
      "loss": 8.0365,
      "mean_token_accuracy": 0.062472108751535416,
      "num_tokens": 950814.0,
      "step": 435
    },
    {
      "entropy": 8.192297077178955,
      "epoch": 0.46983449012279765,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002195,
      "loss": 8.0225,
      "mean_token_accuracy": 0.0661535993218422,
      "num_tokens": 961658.0,
      "step": 440
    },
    {
      "entropy": 8.43046703338623,
      "epoch": 0.4751735184196476,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000222,
      "loss": 8.0993,
      "mean_token_accuracy": 0.06110986284911633,
      "num_tokens": 974790.0,
      "step": 445
    },
    {
      "entropy": 8.331879615783691,
      "epoch": 0.4805125467164976,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0002245,
      "loss": 7.9772,
      "mean_token_accuracy": 0.06222324036061764,
      "num_tokens": 986599.0,
      "step": 450
    },
    {
      "entropy": 8.243362045288086,
      "epoch": 0.4858515750133476,
      "grad_norm": 0.84375,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.9787,
      "mean_token_accuracy": 0.06861450895667076,
      "num_tokens": 997731.0,
      "step": 455
    },
    {
      "entropy": 8.290034008026122,
      "epoch": 0.49119060331019754,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00022950000000000002,
      "loss": 8.0374,
      "mean_token_accuracy": 0.06714494079351425,
      "num_tokens": 1009343.0,
      "step": 460
    },
    {
      "entropy": 8.351651859283447,
      "epoch": 0.4965296316070475,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.9541,
      "mean_token_accuracy": 0.0648241501301527,
      "num_tokens": 1019878.0,
      "step": 465
    },
    {
      "entropy": 8.248888206481933,
      "epoch": 0.5018686599038975,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.9553,
      "mean_token_accuracy": 0.07038948647677898,
      "num_tokens": 1030551.0,
      "step": 470
    },
    {
      "entropy": 8.30935525894165,
      "epoch": 0.5072076882007475,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000237,
      "loss": 7.9893,
      "mean_token_accuracy": 0.06879530549049377,
      "num_tokens": 1041432.0,
      "step": 475
    },
    {
      "entropy": 8.283312702178955,
      "epoch": 0.5125467164975974,
      "grad_norm": 0.84375,
      "learning_rate": 0.0002395,
      "loss": 7.9704,
      "mean_token_accuracy": 0.06568947955965995,
      "num_tokens": 1051769.0,
      "step": 480
    },
    {
      "entropy": 8.256877326965332,
      "epoch": 0.5178857447944474,
      "grad_norm": 1.078125,
      "learning_rate": 0.000242,
      "loss": 7.9282,
      "mean_token_accuracy": 0.06400737687945365,
      "num_tokens": 1062334.0,
      "step": 485
    },
    {
      "entropy": 8.27625093460083,
      "epoch": 0.5232247730912973,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002445,
      "loss": 7.921,
      "mean_token_accuracy": 0.06769514791667461,
      "num_tokens": 1072792.0,
      "step": 490
    },
    {
      "entropy": 8.173498821258544,
      "epoch": 0.5285638013881474,
      "grad_norm": 1.203125,
      "learning_rate": 0.000247,
      "loss": 7.8148,
      "mean_token_accuracy": 0.07047795243561268,
      "num_tokens": 1083642.0,
      "step": 495
    },
    {
      "entropy": 8.192062282562256,
      "epoch": 0.5339028296849974,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002495,
      "loss": 7.9111,
      "mean_token_accuracy": 0.07375151067972183,
      "num_tokens": 1094454.0,
      "step": 500
    },
    {
      "epoch": 0.5339028296849974,
      "eval_entropy": 8.031556909734553,
      "eval_loss": 8.007415771484375,
      "eval_mean_token_accuracy": 0.06165650523715205,
      "eval_num_tokens": 1094454.0,
      "eval_runtime": 1.8278,
      "eval_samples_per_second": 1685.073,
      "eval_steps_per_second": 210.634,
      "step": 500
    },
    {
      "entropy": 8.316486740112305,
      "epoch": 0.5392418579818473,
      "grad_norm": 1.046875,
      "learning_rate": 0.000252,
      "loss": 7.9525,
      "mean_token_accuracy": 0.06484691724181176,
      "num_tokens": 1105068.0,
      "step": 505
    },
    {
      "entropy": 8.111512756347656,
      "epoch": 0.5445808862786973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002545,
      "loss": 7.9097,
      "mean_token_accuracy": 0.07020260021090508,
      "num_tokens": 1115668.0,
      "step": 510
    },
    {
      "entropy": 8.238842296600343,
      "epoch": 0.5499199145755472,
      "grad_norm": 1.046875,
      "learning_rate": 0.000257,
      "loss": 7.8867,
      "mean_token_accuracy": 0.07027992233633995,
      "num_tokens": 1126742.0,
      "step": 515
    },
    {
      "entropy": 8.179580593109131,
      "epoch": 0.5552589428723972,
      "grad_norm": 0.875,
      "learning_rate": 0.0002595,
      "loss": 7.9668,
      "mean_token_accuracy": 0.0645244125276804,
      "num_tokens": 1137351.0,
      "step": 520
    },
    {
      "entropy": 8.22506046295166,
      "epoch": 0.5605979711692471,
      "grad_norm": 0.7421875,
      "learning_rate": 0.000262,
      "loss": 7.8967,
      "mean_token_accuracy": 0.0711884018033743,
      "num_tokens": 1148389.0,
      "step": 525
    },
    {
      "entropy": 8.167451763153077,
      "epoch": 0.5659369994660972,
      "grad_norm": 1.0,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.8994,
      "mean_token_accuracy": 0.0703209575265646,
      "num_tokens": 1158798.0,
      "step": 530
    },
    {
      "entropy": 8.239407634735107,
      "epoch": 0.5712760277629472,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.948,
      "mean_token_accuracy": 0.06852544359862804,
      "num_tokens": 1169005.0,
      "step": 535
    },
    {
      "entropy": 8.212612915039063,
      "epoch": 0.5766150560597971,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.8255,
      "mean_token_accuracy": 0.06702273786067962,
      "num_tokens": 1180560.0,
      "step": 540
    },
    {
      "entropy": 8.105984878540038,
      "epoch": 0.5819540843566471,
      "grad_norm": 0.953125,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.8035,
      "mean_token_accuracy": 0.07094659134745598,
      "num_tokens": 1191162.0,
      "step": 545
    },
    {
      "entropy": 8.162831687927246,
      "epoch": 0.587293112653497,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002745,
      "loss": 7.8252,
      "mean_token_accuracy": 0.06873422190546989,
      "num_tokens": 1202017.0,
      "step": 550
    },
    {
      "entropy": 8.132958889007568,
      "epoch": 0.592632140950347,
      "grad_norm": 1.21875,
      "learning_rate": 0.000277,
      "loss": 7.8232,
      "mean_token_accuracy": 0.07796628661453724,
      "num_tokens": 1213356.0,
      "step": 555
    },
    {
      "entropy": 8.137973403930664,
      "epoch": 0.5979711692471971,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002795,
      "loss": 7.8335,
      "mean_token_accuracy": 0.07087706178426742,
      "num_tokens": 1224164.0,
      "step": 560
    },
    {
      "entropy": 8.08298101425171,
      "epoch": 0.603310197544047,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.8463,
      "mean_token_accuracy": 0.0711651612073183,
      "num_tokens": 1235483.0,
      "step": 565
    },
    {
      "entropy": 8.127214622497558,
      "epoch": 0.608649225840897,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0002845,
      "loss": 7.8595,
      "mean_token_accuracy": 0.07112185694277287,
      "num_tokens": 1248013.0,
      "step": 570
    },
    {
      "entropy": 8.093348598480224,
      "epoch": 0.6139882541377469,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000287,
      "loss": 7.7999,
      "mean_token_accuracy": 0.07371774725615979,
      "num_tokens": 1258512.0,
      "step": 575
    },
    {
      "entropy": 8.157558727264405,
      "epoch": 0.6193272824345969,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002895,
      "loss": 7.8256,
      "mean_token_accuracy": 0.06662675254046917,
      "num_tokens": 1269425.0,
      "step": 580
    },
    {
      "entropy": 7.97422046661377,
      "epoch": 0.6246663107314468,
      "grad_norm": 0.7734375,
      "learning_rate": 0.000292,
      "loss": 7.7958,
      "mean_token_accuracy": 0.07273220643401146,
      "num_tokens": 1280595.0,
      "step": 585
    },
    {
      "entropy": 8.247371101379395,
      "epoch": 0.6300053390282968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002945,
      "loss": 7.8485,
      "mean_token_accuracy": 0.06550373584032058,
      "num_tokens": 1290873.0,
      "step": 590
    },
    {
      "entropy": 8.015791416168213,
      "epoch": 0.6353443673251469,
      "grad_norm": 1.0625,
      "learning_rate": 0.000297,
      "loss": 7.852,
      "mean_token_accuracy": 0.066574127972126,
      "num_tokens": 1303150.0,
      "step": 595
    },
    {
      "entropy": 8.053708696365357,
      "epoch": 0.6406833956219968,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002995,
      "loss": 7.768,
      "mean_token_accuracy": 0.07012191005051135,
      "num_tokens": 1314862.0,
      "step": 600
    },
    {
      "entropy": 8.02850046157837,
      "epoch": 0.6460224239188468,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000302,
      "loss": 7.8029,
      "mean_token_accuracy": 0.07391900308430195,
      "num_tokens": 1326052.0,
      "step": 605
    },
    {
      "entropy": 8.128786611557008,
      "epoch": 0.6513614522156967,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003045,
      "loss": 7.7667,
      "mean_token_accuracy": 0.0753765020519495,
      "num_tokens": 1336847.0,
      "step": 610
    },
    {
      "entropy": 8.02261905670166,
      "epoch": 0.6567004805125467,
      "grad_norm": 1.109375,
      "learning_rate": 0.000307,
      "loss": 7.7701,
      "mean_token_accuracy": 0.06919390633702278,
      "num_tokens": 1347611.0,
      "step": 615
    },
    {
      "entropy": 8.006579780578614,
      "epoch": 0.6620395088093967,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003095,
      "loss": 7.7859,
      "mean_token_accuracy": 0.07407167293131352,
      "num_tokens": 1358679.0,
      "step": 620
    },
    {
      "entropy": 8.048807954788208,
      "epoch": 0.6673785371062466,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000312,
      "loss": 7.7849,
      "mean_token_accuracy": 0.07473289892077446,
      "num_tokens": 1369409.0,
      "step": 625
    },
    {
      "entropy": 8.044457149505615,
      "epoch": 0.6727175654030967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003145,
      "loss": 7.7618,
      "mean_token_accuracy": 0.06960076540708542,
      "num_tokens": 1380380.0,
      "step": 630
    },
    {
      "entropy": 8.039245176315308,
      "epoch": 0.6780565936999466,
      "grad_norm": 0.96875,
      "learning_rate": 0.000317,
      "loss": 7.7081,
      "mean_token_accuracy": 0.07184483371675014,
      "num_tokens": 1390885.0,
      "step": 635
    },
    {
      "entropy": 7.9913702487945555,
      "epoch": 0.6833956219967966,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003195,
      "loss": 7.7418,
      "mean_token_accuracy": 0.07358726263046264,
      "num_tokens": 1400960.0,
      "step": 640
    },
    {
      "entropy": 7.986459016799927,
      "epoch": 0.6887346502936466,
      "grad_norm": 1.125,
      "learning_rate": 0.000322,
      "loss": 7.6917,
      "mean_token_accuracy": 0.07739836499094963,
      "num_tokens": 1411484.0,
      "step": 645
    },
    {
      "entropy": 8.129097318649292,
      "epoch": 0.6940736785904965,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.7908,
      "mean_token_accuracy": 0.07160034552216529,
      "num_tokens": 1422931.0,
      "step": 650
    },
    {
      "entropy": 7.91273045539856,
      "epoch": 0.6994127068873465,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.7237,
      "mean_token_accuracy": 0.07407537326216698,
      "num_tokens": 1434347.0,
      "step": 655
    },
    {
      "entropy": 8.050477170944214,
      "epoch": 0.7047517351841964,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.8135,
      "mean_token_accuracy": 0.07437525913119317,
      "num_tokens": 1445687.0,
      "step": 660
    },
    {
      "entropy": 8.082316970825195,
      "epoch": 0.7100907634810465,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.6471,
      "mean_token_accuracy": 0.08003439009189606,
      "num_tokens": 1456335.0,
      "step": 665
    },
    {
      "entropy": 7.9566230297088625,
      "epoch": 0.7154297917778965,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.7071,
      "mean_token_accuracy": 0.07267946116626263,
      "num_tokens": 1467369.0,
      "step": 670
    },
    {
      "entropy": 7.986024713516235,
      "epoch": 0.7207688200747464,
      "grad_norm": 1.171875,
      "learning_rate": 0.000337,
      "loss": 7.7612,
      "mean_token_accuracy": 0.06941658817231655,
      "num_tokens": 1478734.0,
      "step": 675
    },
    {
      "entropy": 7.984552240371704,
      "epoch": 0.7261078483715964,
      "grad_norm": 0.875,
      "learning_rate": 0.0003395,
      "loss": 7.7359,
      "mean_token_accuracy": 0.07321545220911503,
      "num_tokens": 1489807.0,
      "step": 680
    },
    {
      "entropy": 7.967883205413818,
      "epoch": 0.7314468766684463,
      "grad_norm": 1.046875,
      "learning_rate": 0.000342,
      "loss": 7.6794,
      "mean_token_accuracy": 0.07043348215520381,
      "num_tokens": 1501646.0,
      "step": 685
    },
    {
      "entropy": 7.948320388793945,
      "epoch": 0.7367859049652963,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.6987,
      "mean_token_accuracy": 0.07335374914109707,
      "num_tokens": 1511849.0,
      "step": 690
    },
    {
      "entropy": 7.999287462234497,
      "epoch": 0.7421249332621462,
      "grad_norm": 1.3125,
      "learning_rate": 0.000347,
      "loss": 7.7004,
      "mean_token_accuracy": 0.07678697407245635,
      "num_tokens": 1522183.0,
      "step": 695
    },
    {
      "entropy": 7.895766305923462,
      "epoch": 0.7474639615589963,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003495,
      "loss": 7.7063,
      "mean_token_accuracy": 0.07557001523673534,
      "num_tokens": 1532819.0,
      "step": 700
    },
    {
      "entropy": 8.02945156097412,
      "epoch": 0.7528029898558463,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000352,
      "loss": 7.6718,
      "mean_token_accuracy": 0.07557161971926689,
      "num_tokens": 1543909.0,
      "step": 705
    },
    {
      "entropy": 7.9088582515716555,
      "epoch": 0.7581420181526962,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0003545,
      "loss": 7.7087,
      "mean_token_accuracy": 0.07322404161095619,
      "num_tokens": 1554147.0,
      "step": 710
    },
    {
      "entropy": 7.878606176376342,
      "epoch": 0.7634810464495462,
      "grad_norm": 1.125,
      "learning_rate": 0.000357,
      "loss": 7.6136,
      "mean_token_accuracy": 0.07956673577427864,
      "num_tokens": 1565466.0,
      "step": 715
    },
    {
      "entropy": 7.8776044845581055,
      "epoch": 0.7688200747463961,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003595,
      "loss": 7.5815,
      "mean_token_accuracy": 0.07377611920237541,
      "num_tokens": 1577264.0,
      "step": 720
    },
    {
      "entropy": 7.885418844223023,
      "epoch": 0.7741591030432461,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000362,
      "loss": 7.686,
      "mean_token_accuracy": 0.0757551722228527,
      "num_tokens": 1588222.0,
      "step": 725
    },
    {
      "entropy": 7.901127099990845,
      "epoch": 0.7794981313400962,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003645,
      "loss": 7.6224,
      "mean_token_accuracy": 0.07221109680831432,
      "num_tokens": 1600609.0,
      "step": 730
    },
    {
      "entropy": 7.929693412780762,
      "epoch": 0.7848371596369461,
      "grad_norm": 1.125,
      "learning_rate": 0.000367,
      "loss": 7.669,
      "mean_token_accuracy": 0.07948594614863395,
      "num_tokens": 1612032.0,
      "step": 735
    },
    {
      "entropy": 7.876828384399414,
      "epoch": 0.7901761879337961,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003695,
      "loss": 7.6973,
      "mean_token_accuracy": 0.0742558129131794,
      "num_tokens": 1623228.0,
      "step": 740
    },
    {
      "entropy": 7.857063436508179,
      "epoch": 0.795515216230646,
      "grad_norm": 1.15625,
      "learning_rate": 0.000372,
      "loss": 7.6469,
      "mean_token_accuracy": 0.07817403376102447,
      "num_tokens": 1634293.0,
      "step": 745
    },
    {
      "entropy": 7.910915946960449,
      "epoch": 0.800854244527496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003745,
      "loss": 7.6183,
      "mean_token_accuracy": 0.08015101104974746,
      "num_tokens": 1645782.0,
      "step": 750
    },
    {
      "entropy": 7.885407543182373,
      "epoch": 0.8061932728243459,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377,
      "loss": 7.5871,
      "mean_token_accuracy": 0.0775064155459404,
      "num_tokens": 1657060.0,
      "step": 755
    },
    {
      "entropy": 7.78002610206604,
      "epoch": 0.8115323011211959,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003795,
      "loss": 7.568,
      "mean_token_accuracy": 0.08284709006547927,
      "num_tokens": 1668017.0,
      "step": 760
    },
    {
      "entropy": 7.908759117126465,
      "epoch": 0.816871329418046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000382,
      "loss": 7.6398,
      "mean_token_accuracy": 0.07324047535657882,
      "num_tokens": 1678331.0,
      "step": 765
    },
    {
      "entropy": 7.884358358383179,
      "epoch": 0.8222103577148959,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003845,
      "loss": 7.5591,
      "mean_token_accuracy": 0.0813971921801567,
      "num_tokens": 1688719.0,
      "step": 770
    },
    {
      "entropy": 7.755726432800293,
      "epoch": 0.8275493860117459,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.5821,
      "mean_token_accuracy": 0.07562258839607239,
      "num_tokens": 1699028.0,
      "step": 775
    },
    {
      "entropy": 7.950335073471069,
      "epoch": 0.8328884143085958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.5798,
      "mean_token_accuracy": 0.07690282315015792,
      "num_tokens": 1711122.0,
      "step": 780
    },
    {
      "entropy": 7.790361976623535,
      "epoch": 0.8382274426054458,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.5939,
      "mean_token_accuracy": 0.07726500257849693,
      "num_tokens": 1721694.0,
      "step": 785
    },
    {
      "entropy": 7.733095169067383,
      "epoch": 0.8435664709022958,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.5939,
      "mean_token_accuracy": 0.0780683733522892,
      "num_tokens": 1732530.0,
      "step": 790
    },
    {
      "entropy": 7.877895975112915,
      "epoch": 0.8489054991991457,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.5577,
      "mean_token_accuracy": 0.08093490153551101,
      "num_tokens": 1742860.0,
      "step": 795
    },
    {
      "entropy": 7.750740194320679,
      "epoch": 0.8542445274959958,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003995,
      "loss": 7.5315,
      "mean_token_accuracy": 0.0812942422926426,
      "num_tokens": 1752900.0,
      "step": 800
    },
    {
      "entropy": 7.801487588882447,
      "epoch": 0.8595835557928457,
      "grad_norm": 1.0,
      "learning_rate": 0.000402,
      "loss": 7.5465,
      "mean_token_accuracy": 0.07882912531495094,
      "num_tokens": 1763293.0,
      "step": 805
    },
    {
      "entropy": 7.752074146270752,
      "epoch": 0.8649225840896957,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004045,
      "loss": 7.5696,
      "mean_token_accuracy": 0.0819572426378727,
      "num_tokens": 1773893.0,
      "step": 810
    },
    {
      "entropy": 7.815144634246826,
      "epoch": 0.8702616123865456,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.6358,
      "mean_token_accuracy": 0.07636146545410157,
      "num_tokens": 1785568.0,
      "step": 815
    },
    {
      "entropy": 7.8184651851654055,
      "epoch": 0.8756006406833956,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004095,
      "loss": 7.5594,
      "mean_token_accuracy": 0.08072886392474174,
      "num_tokens": 1797387.0,
      "step": 820
    },
    {
      "entropy": 7.813911390304566,
      "epoch": 0.8809396689802456,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000412,
      "loss": 7.5902,
      "mean_token_accuracy": 0.07879998087882996,
      "num_tokens": 1808382.0,
      "step": 825
    },
    {
      "entropy": 7.829354333877563,
      "epoch": 0.8862786972770955,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004145,
      "loss": 7.5745,
      "mean_token_accuracy": 0.0824002705514431,
      "num_tokens": 1818937.0,
      "step": 830
    },
    {
      "entropy": 7.799869441986084,
      "epoch": 0.8916177255739456,
      "grad_norm": 0.96875,
      "learning_rate": 0.000417,
      "loss": 7.5099,
      "mean_token_accuracy": 0.07985455431044101,
      "num_tokens": 1830547.0,
      "step": 835
    },
    {
      "entropy": 7.7153465270996096,
      "epoch": 0.8969567538707955,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004195,
      "loss": 7.5052,
      "mean_token_accuracy": 0.08278301656246186,
      "num_tokens": 1841569.0,
      "step": 840
    },
    {
      "entropy": 7.764185190200806,
      "epoch": 0.9022957821676455,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000422,
      "loss": 7.5154,
      "mean_token_accuracy": 0.08583573549985886,
      "num_tokens": 1852300.0,
      "step": 845
    },
    {
      "entropy": 7.718573999404907,
      "epoch": 0.9076348104644955,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004245,
      "loss": 7.5077,
      "mean_token_accuracy": 0.08417970538139344,
      "num_tokens": 1863664.0,
      "step": 850
    },
    {
      "entropy": 7.6863548278808596,
      "epoch": 0.9129738387613454,
      "grad_norm": 1.0,
      "learning_rate": 0.000427,
      "loss": 7.4426,
      "mean_token_accuracy": 0.08229955136775971,
      "num_tokens": 1873623.0,
      "step": 855
    },
    {
      "entropy": 7.687994432449341,
      "epoch": 0.9183128670581954,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004295,
      "loss": 7.5429,
      "mean_token_accuracy": 0.0791929729282856,
      "num_tokens": 1886092.0,
      "step": 860
    },
    {
      "entropy": 7.751056241989136,
      "epoch": 0.9236518953550453,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000432,
      "loss": 7.5525,
      "mean_token_accuracy": 0.08293391354382038,
      "num_tokens": 1897983.0,
      "step": 865
    },
    {
      "entropy": 7.769723224639892,
      "epoch": 0.9289909236518954,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004345,
      "loss": 7.5134,
      "mean_token_accuracy": 0.08427821174263954,
      "num_tokens": 1908193.0,
      "step": 870
    },
    {
      "entropy": 7.7698462963104244,
      "epoch": 0.9343299519487454,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000437,
      "loss": 7.4921,
      "mean_token_accuracy": 0.08160046115517616,
      "num_tokens": 1918628.0,
      "step": 875
    },
    {
      "entropy": 7.668918085098267,
      "epoch": 0.9396689802455953,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004395,
      "loss": 7.5372,
      "mean_token_accuracy": 0.08014441058039665,
      "num_tokens": 1929071.0,
      "step": 880
    },
    {
      "entropy": 7.654262733459473,
      "epoch": 0.9450080085424453,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000442,
      "loss": 7.5003,
      "mean_token_accuracy": 0.08210862651467324,
      "num_tokens": 1939567.0,
      "step": 885
    },
    {
      "entropy": 7.773276090621948,
      "epoch": 0.9503470368392952,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004445,
      "loss": 7.5402,
      "mean_token_accuracy": 0.08018058091402054,
      "num_tokens": 1951165.0,
      "step": 890
    },
    {
      "entropy": 7.798144149780273,
      "epoch": 0.9556860651361452,
      "grad_norm": 1.0625,
      "learning_rate": 0.000447,
      "loss": 7.4594,
      "mean_token_accuracy": 0.08885749988257885,
      "num_tokens": 1960855.0,
      "step": 895
    },
    {
      "entropy": 7.707589864730835,
      "epoch": 0.9610250934329952,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.4564,
      "mean_token_accuracy": 0.08240535110235214,
      "num_tokens": 1971873.0,
      "step": 900
    },
    {
      "entropy": 7.671180057525635,
      "epoch": 0.9663641217298452,
      "grad_norm": 0.828125,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.4848,
      "mean_token_accuracy": 0.08260536044836045,
      "num_tokens": 1983561.0,
      "step": 905
    },
    {
      "entropy": 7.731744003295899,
      "epoch": 0.9717031500266952,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.4828,
      "mean_token_accuracy": 0.08393343463540077,
      "num_tokens": 1994534.0,
      "step": 910
    },
    {
      "entropy": 7.590392780303955,
      "epoch": 0.9770421783235451,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.432,
      "mean_token_accuracy": 0.08598880358040333,
      "num_tokens": 2005121.0,
      "step": 915
    },
    {
      "entropy": 7.696979951858521,
      "epoch": 0.9823812066203951,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.4211,
      "mean_token_accuracy": 0.08999745771288872,
      "num_tokens": 2016260.0,
      "step": 920
    },
    {
      "entropy": 7.615106248855591,
      "epoch": 0.987720234917245,
      "grad_norm": 0.890625,
      "learning_rate": 0.000462,
      "loss": 7.4074,
      "mean_token_accuracy": 0.08892037123441696,
      "num_tokens": 2025705.0,
      "step": 925
    },
    {
      "entropy": 7.7097069263458256,
      "epoch": 0.993059263214095,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004645,
      "loss": 7.5596,
      "mean_token_accuracy": 0.07989178895950318,
      "num_tokens": 2036675.0,
      "step": 930
    },
    {
      "entropy": 7.544618844985962,
      "epoch": 0.9983982915109451,
      "grad_norm": 0.953125,
      "learning_rate": 0.000467,
      "loss": 7.4114,
      "mean_token_accuracy": 0.08617193698883056,
      "num_tokens": 2047877.0,
      "step": 935
    },
    {
      "entropy": 7.8020609219868975,
      "epoch": 1.0032034169781099,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004695,
      "loss": 7.3332,
      "mean_token_accuracy": 0.08596346527338028,
      "num_tokens": 2057153.0,
      "step": 940
    },
    {
      "entropy": 7.623947048187256,
      "epoch": 1.00854244527496,
      "grad_norm": 1.265625,
      "learning_rate": 0.000472,
      "loss": 7.4568,
      "mean_token_accuracy": 0.0771791972219944,
      "num_tokens": 2068760.0,
      "step": 945
    },
    {
      "entropy": 7.589405059814453,
      "epoch": 1.01388147357181,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004745,
      "loss": 7.3104,
      "mean_token_accuracy": 0.0916103646159172,
      "num_tokens": 2078671.0,
      "step": 950
    },
    {
      "entropy": 7.588209199905395,
      "epoch": 1.01922050186866,
      "grad_norm": 0.953125,
      "learning_rate": 0.000477,
      "loss": 7.3495,
      "mean_token_accuracy": 0.08704460337758065,
      "num_tokens": 2089581.0,
      "step": 955
    },
    {
      "entropy": 7.6823986053466795,
      "epoch": 1.02455953016551,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004795,
      "loss": 7.293,
      "mean_token_accuracy": 0.09483219981193543,
      "num_tokens": 2100026.0,
      "step": 960
    },
    {
      "entropy": 7.51586127281189,
      "epoch": 1.02989855846236,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000482,
      "loss": 7.3062,
      "mean_token_accuracy": 0.09631321728229522,
      "num_tokens": 2111838.0,
      "step": 965
    },
    {
      "entropy": 7.603687572479248,
      "epoch": 1.0352375867592098,
      "grad_norm": 0.765625,
      "learning_rate": 0.0004845,
      "loss": 7.3908,
      "mean_token_accuracy": 0.0855435535311699,
      "num_tokens": 2123063.0,
      "step": 970
    },
    {
      "entropy": 7.658422660827637,
      "epoch": 1.0405766150560598,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487,
      "loss": 7.279,
      "mean_token_accuracy": 0.09217825904488564,
      "num_tokens": 2134321.0,
      "step": 975
    },
    {
      "entropy": 7.596205329895019,
      "epoch": 1.0459156433529098,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004895,
      "loss": 7.3405,
      "mean_token_accuracy": 0.09226287379860879,
      "num_tokens": 2144216.0,
      "step": 980
    },
    {
      "entropy": 7.586561298370361,
      "epoch": 1.0512546716497597,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492,
      "loss": 7.3236,
      "mean_token_accuracy": 0.08801514208316803,
      "num_tokens": 2155757.0,
      "step": 985
    },
    {
      "entropy": 7.509102058410645,
      "epoch": 1.0565936999466097,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004945,
      "loss": 7.2762,
      "mean_token_accuracy": 0.08867826834321021,
      "num_tokens": 2166661.0,
      "step": 990
    },
    {
      "entropy": 7.554547595977783,
      "epoch": 1.0619327282434596,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497,
      "loss": 7.2522,
      "mean_token_accuracy": 0.0905068688094616,
      "num_tokens": 2176796.0,
      "step": 995
    },
    {
      "entropy": 7.565032482147217,
      "epoch": 1.0672717565403096,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995,
      "loss": 7.3151,
      "mean_token_accuracy": 0.09214983209967613,
      "num_tokens": 2186821.0,
      "step": 1000
    },
    {
      "epoch": 1.0672717565403096,
      "eval_entropy": 7.661522679514699,
      "eval_loss": 7.484097957611084,
      "eval_mean_token_accuracy": 0.08444042449163926,
      "eval_num_tokens": 2186821.0,
      "eval_runtime": 1.8185,
      "eval_samples_per_second": 1693.67,
      "eval_steps_per_second": 211.709,
      "step": 1000
    },
    {
      "entropy": 7.488584995269775,
      "epoch": 1.0726107848371595,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999997458093253,
      "loss": 7.2872,
      "mean_token_accuracy": 0.08735908269882202,
      "num_tokens": 2197953.0,
      "step": 1005
    },
    {
      "entropy": 7.587278175354004,
      "epoch": 1.0779498131340097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999987131606938,
      "loss": 7.2946,
      "mean_token_accuracy": 0.09318367019295692,
      "num_tokens": 2208033.0,
      "step": 1010
    },
    {
      "entropy": 7.508203601837158,
      "epoch": 1.0832888414308597,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999968861708311,
      "loss": 7.306,
      "mean_token_accuracy": 0.0833747498691082,
      "num_tokens": 2218010.0,
      "step": 1015
    },
    {
      "entropy": 7.607468748092652,
      "epoch": 1.0886278697277096,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999942648461873,
      "loss": 7.3213,
      "mean_token_accuracy": 0.08709867373108864,
      "num_tokens": 2229079.0,
      "step": 1020
    },
    {
      "entropy": 7.554487895965576,
      "epoch": 1.0939668980245596,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999908491960169,
      "loss": 7.3188,
      "mean_token_accuracy": 0.08554268255829811,
      "num_tokens": 2240333.0,
      "step": 1025
    },
    {
      "entropy": 7.41667366027832,
      "epoch": 1.0993059263214096,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999866392323784,
      "loss": 7.2619,
      "mean_token_accuracy": 0.09279655292630196,
      "num_tokens": 2251385.0,
      "step": 1030
    },
    {
      "entropy": 7.586219072341919,
      "epoch": 1.1046449546182595,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999816349701349,
      "loss": 7.2562,
      "mean_token_accuracy": 0.09476209431886673,
      "num_tokens": 2261890.0,
      "step": 1035
    },
    {
      "entropy": 7.429596185684204,
      "epoch": 1.1099839829151095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999758364269535,
      "loss": 7.287,
      "mean_token_accuracy": 0.0902016930282116,
      "num_tokens": 2272510.0,
      "step": 1040
    },
    {
      "entropy": 7.436788702011109,
      "epoch": 1.1153230112119594,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999692436233055,
      "loss": 7.2791,
      "mean_token_accuracy": 0.08760200515389442,
      "num_tokens": 2283392.0,
      "step": 1045
    },
    {
      "entropy": 7.570240783691406,
      "epoch": 1.1206620395088094,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999618565824665,
      "loss": 7.2279,
      "mean_token_accuracy": 0.09378096982836723,
      "num_tokens": 2295155.0,
      "step": 1050
    },
    {
      "entropy": 7.490195608139038,
      "epoch": 1.1260010678056593,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999536753305157,
      "loss": 7.3268,
      "mean_token_accuracy": 0.08983222171664237,
      "num_tokens": 2307208.0,
      "step": 1055
    },
    {
      "entropy": 7.629471921920777,
      "epoch": 1.1313400961025093,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999446998963364,
      "loss": 7.2481,
      "mean_token_accuracy": 0.09013928584754467,
      "num_tokens": 2317480.0,
      "step": 1060
    },
    {
      "entropy": 7.449492406845093,
      "epoch": 1.1366791243993593,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999349303116159,
      "loss": 7.3373,
      "mean_token_accuracy": 0.08976641297340393,
      "num_tokens": 2328840.0,
      "step": 1065
    },
    {
      "entropy": 7.580901098251343,
      "epoch": 1.1420181526962092,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999243666108448,
      "loss": 7.285,
      "mean_token_accuracy": 0.09045098423957824,
      "num_tokens": 2340450.0,
      "step": 1070
    },
    {
      "entropy": 7.4406897068023685,
      "epoch": 1.1473571809930592,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999130088313176,
      "loss": 7.3033,
      "mean_token_accuracy": 0.09590284526348114,
      "num_tokens": 2351415.0,
      "step": 1075
    },
    {
      "entropy": 7.485285329818725,
      "epoch": 1.1526962092899091,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999008570131319,
      "loss": 7.2965,
      "mean_token_accuracy": 0.09630362093448638,
      "num_tokens": 2362153.0,
      "step": 1080
    },
    {
      "entropy": 7.625531721115112,
      "epoch": 1.1580352375867593,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998879111991889,
      "loss": 7.2759,
      "mean_token_accuracy": 0.089786946028471,
      "num_tokens": 2372352.0,
      "step": 1085
    },
    {
      "entropy": 7.374940299987793,
      "epoch": 1.1633742658836093,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499874171435193,
      "loss": 7.2659,
      "mean_token_accuracy": 0.09284015148878097,
      "num_tokens": 2382294.0,
      "step": 1090
    },
    {
      "entropy": 7.572332715988159,
      "epoch": 1.1687132941804592,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998596377696511,
      "loss": 7.2382,
      "mean_token_accuracy": 0.09612747579813004,
      "num_tokens": 2394054.0,
      "step": 1095
    },
    {
      "entropy": 7.367642450332641,
      "epoch": 1.1740523224773092,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998443102538734,
      "loss": 7.2585,
      "mean_token_accuracy": 0.09028631299734116,
      "num_tokens": 2405408.0,
      "step": 1100
    },
    {
      "entropy": 7.572077369689941,
      "epoch": 1.1793913507741591,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998281889419725,
      "loss": 7.2905,
      "mean_token_accuracy": 0.08611830845475196,
      "num_tokens": 2416378.0,
      "step": 1105
    },
    {
      "entropy": 7.381543397903442,
      "epoch": 1.184730379071009,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998112738908636,
      "loss": 7.183,
      "mean_token_accuracy": 0.09558979794383049,
      "num_tokens": 2427016.0,
      "step": 1110
    },
    {
      "entropy": 7.501509714126587,
      "epoch": 1.190069407367859,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997935651602641,
      "loss": 7.1302,
      "mean_token_accuracy": 0.09728729203343392,
      "num_tokens": 2437824.0,
      "step": 1115
    },
    {
      "entropy": 7.4092631340026855,
      "epoch": 1.195408435664709,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997750628126931,
      "loss": 7.2588,
      "mean_token_accuracy": 0.08902693465352059,
      "num_tokens": 2448829.0,
      "step": 1120
    },
    {
      "entropy": 7.444784927368164,
      "epoch": 1.200747463961559,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997557669134718,
      "loss": 7.2186,
      "mean_token_accuracy": 0.09314063340425491,
      "num_tokens": 2459629.0,
      "step": 1125
    },
    {
      "entropy": 7.362234878540039,
      "epoch": 1.206086492258409,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004997356775307231,
      "loss": 7.2803,
      "mean_token_accuracy": 0.09419681504368782,
      "num_tokens": 2471478.0,
      "step": 1130
    },
    {
      "entropy": 7.532480144500733,
      "epoch": 1.2114255205552589,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499714794735371,
      "loss": 7.1266,
      "mean_token_accuracy": 0.10543641149997711,
      "num_tokens": 2481573.0,
      "step": 1135
    },
    {
      "entropy": 7.350391530990601,
      "epoch": 1.2167645488521088,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996931186011406,
      "loss": 7.256,
      "mean_token_accuracy": 0.09152917265892029,
      "num_tokens": 2493429.0,
      "step": 1140
    },
    {
      "entropy": 7.421112728118897,
      "epoch": 1.222103577148959,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499670649204558,
      "loss": 7.218,
      "mean_token_accuracy": 0.09748484790325165,
      "num_tokens": 2503882.0,
      "step": 1145
    },
    {
      "entropy": 7.419456720352173,
      "epoch": 1.227442605445809,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996473866249496,
      "loss": 7.1971,
      "mean_token_accuracy": 0.09453025832772255,
      "num_tokens": 2513978.0,
      "step": 1150
    },
    {
      "entropy": 7.512584495544433,
      "epoch": 1.232781633742659,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996233309444424,
      "loss": 7.2582,
      "mean_token_accuracy": 0.09497607797384262,
      "num_tokens": 2523806.0,
      "step": 1155
    },
    {
      "entropy": 7.397188282012939,
      "epoch": 1.2381206620395089,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499598482247963,
      "loss": 7.2416,
      "mean_token_accuracy": 0.09581049084663391,
      "num_tokens": 2533695.0,
      "step": 1160
    },
    {
      "entropy": 7.449572372436523,
      "epoch": 1.2434596903363588,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004995728406232378,
      "loss": 7.1616,
      "mean_token_accuracy": 0.10089872553944587,
      "num_tokens": 2544315.0,
      "step": 1165
    },
    {
      "entropy": 7.433754110336304,
      "epoch": 1.2487987186332088,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000499546406160793,
      "loss": 7.1648,
      "mean_token_accuracy": 0.09890817850828171,
      "num_tokens": 2555985.0,
      "step": 1170
    },
    {
      "entropy": 7.336302137374878,
      "epoch": 1.2541377469300587,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995191789539531,
      "loss": 7.2444,
      "mean_token_accuracy": 0.09039352908730507,
      "num_tokens": 2567472.0,
      "step": 1175
    },
    {
      "entropy": 7.473226404190063,
      "epoch": 1.2594767752269087,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004994911590988421,
      "loss": 7.2288,
      "mean_token_accuracy": 0.09728409722447395,
      "num_tokens": 2578030.0,
      "step": 1180
    },
    {
      "entropy": 7.461479711532593,
      "epoch": 1.2648158035237587,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994623466943816,
      "loss": 7.2471,
      "mean_token_accuracy": 0.093011786788702,
      "num_tokens": 2589925.0,
      "step": 1185
    },
    {
      "entropy": 7.40461311340332,
      "epoch": 1.2701548318206086,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994327418422922,
      "loss": 7.1787,
      "mean_token_accuracy": 0.09393652454018593,
      "num_tokens": 2601825.0,
      "step": 1190
    },
    {
      "entropy": 7.403125238418579,
      "epoch": 1.2754938601174586,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994023446470912,
      "loss": 7.1985,
      "mean_token_accuracy": 0.09799938648939133,
      "num_tokens": 2612726.0,
      "step": 1195
    },
    {
      "entropy": 7.348568964004516,
      "epoch": 1.2808328884143085,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993711552160936,
      "loss": 7.1336,
      "mean_token_accuracy": 0.09304829761385917,
      "num_tokens": 2623753.0,
      "step": 1200
    },
    {
      "entropy": 7.404272222518921,
      "epoch": 1.2861719167111585,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993391736594117,
      "loss": 7.1593,
      "mean_token_accuracy": 0.0968549020588398,
      "num_tokens": 2635275.0,
      "step": 1205
    },
    {
      "entropy": 7.480840063095092,
      "epoch": 1.2915109450080084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993064000899538,
      "loss": 7.2019,
      "mean_token_accuracy": 0.09882493540644646,
      "num_tokens": 2646329.0,
      "step": 1210
    },
    {
      "entropy": 7.275875473022461,
      "epoch": 1.2968499733048584,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992728346234246,
      "loss": 7.2084,
      "mean_token_accuracy": 0.09495408535003662,
      "num_tokens": 2657629.0,
      "step": 1215
    },
    {
      "entropy": 7.453315734863281,
      "epoch": 1.3021890016017084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992384773783244,
      "loss": 7.1886,
      "mean_token_accuracy": 0.09616189375519753,
      "num_tokens": 2667746.0,
      "step": 1220
    },
    {
      "entropy": 7.27887864112854,
      "epoch": 1.3075280298985585,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992033284759491,
      "loss": 7.1507,
      "mean_token_accuracy": 0.10439181700348854,
      "num_tokens": 2678157.0,
      "step": 1225
    },
    {
      "entropy": 7.484576034545898,
      "epoch": 1.3128670581954085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991673880403891,
      "loss": 7.2097,
      "mean_token_accuracy": 0.09509778320789337,
      "num_tokens": 2689319.0,
      "step": 1230
    },
    {
      "entropy": 7.382059717178345,
      "epoch": 1.3182060864922585,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991306561985294,
      "loss": 7.1406,
      "mean_token_accuracy": 0.09976356402039528,
      "num_tokens": 2699718.0,
      "step": 1235
    },
    {
      "entropy": 7.359973287582397,
      "epoch": 1.3235451147891084,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990931330800492,
      "loss": 7.19,
      "mean_token_accuracy": 0.09778250828385353,
      "num_tokens": 2710888.0,
      "step": 1240
    },
    {
      "entropy": 7.3708795547485355,
      "epoch": 1.3288841430859584,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990548188174211,
      "loss": 7.1108,
      "mean_token_accuracy": 0.0972976379096508,
      "num_tokens": 2722024.0,
      "step": 1245
    },
    {
      "entropy": 7.322368574142456,
      "epoch": 1.3342231713828083,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990157135459105,
      "loss": 7.0693,
      "mean_token_accuracy": 0.09633990824222564,
      "num_tokens": 2731780.0,
      "step": 1250
    },
    {
      "entropy": 7.395527935028076,
      "epoch": 1.3395621996796583,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989758174035758,
      "loss": 7.1745,
      "mean_token_accuracy": 0.0960126519203186,
      "num_tokens": 2742524.0,
      "step": 1255
    },
    {
      "entropy": 7.409232425689697,
      "epoch": 1.3449012279765082,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004989351305312676,
      "loss": 7.1378,
      "mean_token_accuracy": 0.09832295700907707,
      "num_tokens": 2752077.0,
      "step": 1260
    },
    {
      "entropy": 7.322344446182251,
      "epoch": 1.3502402562733582,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988936530726276,
      "loss": 7.1473,
      "mean_token_accuracy": 0.09714090898633003,
      "num_tokens": 2763625.0,
      "step": 1265
    },
    {
      "entropy": 7.293940925598145,
      "epoch": 1.3555792845702082,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988513851740889,
      "loss": 7.1327,
      "mean_token_accuracy": 0.10281604751944542,
      "num_tokens": 2775262.0,
      "step": 1270
    },
    {
      "entropy": 7.3934331893920895,
      "epoch": 1.3609183128670583,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988083269848753,
      "loss": 7.1964,
      "mean_token_accuracy": 0.09679302722215652,
      "num_tokens": 2786449.0,
      "step": 1275
    },
    {
      "entropy": 7.385905408859253,
      "epoch": 1.3662573411639083,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004987644786570003,
      "loss": 7.1696,
      "mean_token_accuracy": 0.09244814142584801,
      "num_tokens": 2796891.0,
      "step": 1280
    },
    {
      "entropy": 7.33904447555542,
      "epoch": 1.3715963694607582,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004987198403452674,
      "loss": 7.1607,
      "mean_token_accuracy": 0.09559275880455971,
      "num_tokens": 2806926.0,
      "step": 1285
    },
    {
      "entropy": 7.3863444328308105,
      "epoch": 1.3769353977576082,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986744122072688,
      "loss": 7.0899,
      "mean_token_accuracy": 0.09752057045698166,
      "num_tokens": 2817750.0,
      "step": 1290
    },
    {
      "entropy": 7.180061292648316,
      "epoch": 1.3822744260544582,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986281944033851,
      "loss": 7.1001,
      "mean_token_accuracy": 0.10428606942296029,
      "num_tokens": 2827770.0,
      "step": 1295
    },
    {
      "entropy": 7.401736879348755,
      "epoch": 1.3876134543513081,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985811870967848,
      "loss": 7.1361,
      "mean_token_accuracy": 0.10217376500368118,
      "num_tokens": 2839583.0,
      "step": 1300
    },
    {
      "entropy": 7.363368988037109,
      "epoch": 1.392952482648158,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985333904534237,
      "loss": 7.1358,
      "mean_token_accuracy": 0.10174108073115348,
      "num_tokens": 2850621.0,
      "step": 1305
    },
    {
      "entropy": 7.260199737548828,
      "epoch": 1.398291510945008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004984848046420445,
      "loss": 7.1302,
      "mean_token_accuracy": 0.09183874055743217,
      "num_tokens": 2862287.0,
      "step": 1310
    },
    {
      "entropy": 7.392441749572754,
      "epoch": 1.403630539241858,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984354298341756,
      "loss": 7.1278,
      "mean_token_accuracy": 0.10441029965877532,
      "num_tokens": 2871769.0,
      "step": 1315
    },
    {
      "entropy": 7.305068445205689,
      "epoch": 1.408969567538708,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004983852662041312,
      "loss": 7.1009,
      "mean_token_accuracy": 0.0979885146021843,
      "num_tokens": 2882590.0,
      "step": 1320
    },
    {
      "entropy": 7.323603248596191,
      "epoch": 1.414308595835558,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00049833431392901,
      "loss": 7.1618,
      "mean_token_accuracy": 0.09200560078024864,
      "num_tokens": 2893903.0,
      "step": 1325
    },
    {
      "entropy": 7.257247972488403,
      "epoch": 1.4196476241324079,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982825731886956,
      "loss": 7.0838,
      "mean_token_accuracy": 0.09896045774221421,
      "num_tokens": 2904532.0,
      "step": 1330
    },
    {
      "entropy": 7.321243238449097,
      "epoch": 1.4249866524292578,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982300441658546,
      "loss": 7.0602,
      "mean_token_accuracy": 0.09538981467485427,
      "num_tokens": 2916340.0,
      "step": 1335
    },
    {
      "entropy": 7.252778673171997,
      "epoch": 1.4303256807261078,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981767270459367,
      "loss": 7.0686,
      "mean_token_accuracy": 0.10086248964071273,
      "num_tokens": 2926424.0,
      "step": 1340
    },
    {
      "entropy": 7.384427738189697,
      "epoch": 1.4356647090229577,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981226220171744,
      "loss": 7.1487,
      "mean_token_accuracy": 0.09660428166389465,
      "num_tokens": 2937620.0,
      "step": 1345
    },
    {
      "entropy": 7.306267547607422,
      "epoch": 1.4410037373198077,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980677292705811,
      "loss": 7.0283,
      "mean_token_accuracy": 0.10887585952877998,
      "num_tokens": 2948274.0,
      "step": 1350
    },
    {
      "entropy": 7.251709699630737,
      "epoch": 1.4463427656166576,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004980120489999519,
      "loss": 7.1382,
      "mean_token_accuracy": 0.09639003127813339,
      "num_tokens": 2958986.0,
      "step": 1355
    },
    {
      "entropy": 7.4081559658050535,
      "epoch": 1.4516817939135078,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004979555814018617,
      "loss": 7.162,
      "mean_token_accuracy": 0.09467695504426957,
      "num_tokens": 2970933.0,
      "step": 1360
    },
    {
      "entropy": 7.283700609207154,
      "epoch": 1.4570208222103578,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497898326675665,
      "loss": 7.1182,
      "mean_token_accuracy": 0.10104830637574196,
      "num_tokens": 2981197.0,
      "step": 1365
    },
    {
      "entropy": 7.330152797698974,
      "epoch": 1.4623598505072077,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004978402850234956,
      "loss": 7.1262,
      "mean_token_accuracy": 0.10599048286676407,
      "num_tokens": 2993732.0,
      "step": 1370
    },
    {
      "entropy": 7.264380216598511,
      "epoch": 1.4676988788040577,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977814566502651,
      "loss": 7.071,
      "mean_token_accuracy": 0.10518286749720573,
      "num_tokens": 3004617.0,
      "step": 1375
    },
    {
      "entropy": 7.271407747268677,
      "epoch": 1.4730379071009077,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977218417636629,
      "loss": 7.0138,
      "mean_token_accuracy": 0.10099183842539787,
      "num_tokens": 3015058.0,
      "step": 1380
    },
    {
      "entropy": 7.180662345886231,
      "epoch": 1.4783769353977576,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004976614405741546,
      "loss": 7.0199,
      "mean_token_accuracy": 0.10256859883666039,
      "num_tokens": 3025995.0,
      "step": 1385
    },
    {
      "entropy": 7.355836534500122,
      "epoch": 1.4837159636946076,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004976002532949823,
      "loss": 7.0472,
      "mean_token_accuracy": 0.1040610671043396,
      "num_tokens": 3036060.0,
      "step": 1390
    },
    {
      "entropy": 7.186236763000489,
      "epoch": 1.4890549919914575,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004975382801421631,
      "loss": 7.0578,
      "mean_token_accuracy": 0.1070075087249279,
      "num_tokens": 3047012.0,
      "step": 1395
    },
    {
      "entropy": 7.2215704917907715,
      "epoch": 1.4943940202883075,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974755213344886,
      "loss": 6.9844,
      "mean_token_accuracy": 0.1064968764781952,
      "num_tokens": 3058186.0,
      "step": 1400
    },
    {
      "entropy": 7.222527360916137,
      "epoch": 1.4997330485851574,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974119770935242,
      "loss": 6.9946,
      "mean_token_accuracy": 0.10440263673663139,
      "num_tokens": 3068566.0,
      "step": 1405
    },
    {
      "entropy": 7.2857832431793215,
      "epoch": 1.5050720768820076,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973476476436081,
      "loss": 7.0199,
      "mean_token_accuracy": 0.10458874180912972,
      "num_tokens": 3079588.0,
      "step": 1410
    },
    {
      "entropy": 7.08985447883606,
      "epoch": 1.5104111051788576,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004972825332118505,
      "loss": 7.0023,
      "mean_token_accuracy": 0.10629142746329308,
      "num_tokens": 3092376.0,
      "step": 1415
    },
    {
      "entropy": 7.245244026184082,
      "epoch": 1.5157501334757075,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004972166340281331,
      "loss": 7.0461,
      "mean_token_accuracy": 0.10235923826694489,
      "num_tokens": 3103439.0,
      "step": 1420
    },
    {
      "entropy": 7.26457109451294,
      "epoch": 1.5210891617725575,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004971499503251083,
      "loss": 7.1008,
      "mean_token_accuracy": 0.10314854681491852,
      "num_tokens": 3115007.0,
      "step": 1425
    },
    {
      "entropy": 7.262210988998413,
      "epoch": 1.5264281900694074,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004970824823381976,
      "loss": 7.0539,
      "mean_token_accuracy": 0.10062195062637329,
      "num_tokens": 3125632.0,
      "step": 1430
    },
    {
      "entropy": 7.242954254150391,
      "epoch": 1.5317672183662574,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497014230305592,
      "loss": 7.0316,
      "mean_token_accuracy": 0.10418535098433494,
      "num_tokens": 3135825.0,
      "step": 1435
    },
    {
      "entropy": 7.2584309577941895,
      "epoch": 1.5371062466631074,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969451944682501,
      "loss": 7.0369,
      "mean_token_accuracy": 0.10161466002464295,
      "num_tokens": 3147689.0,
      "step": 1440
    },
    {
      "entropy": 7.136695671081543,
      "epoch": 1.5424452749599573,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004968753750698978,
      "loss": 6.993,
      "mean_token_accuracy": 0.10334522351622581,
      "num_tokens": 3158754.0,
      "step": 1445
    },
    {
      "entropy": 7.298619794845581,
      "epoch": 1.5477843032568073,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004968047723570273,
      "loss": 7.0495,
      "mean_token_accuracy": 0.09448943957686425,
      "num_tokens": 3169332.0,
      "step": 1450
    },
    {
      "entropy": 7.218168115615844,
      "epoch": 1.5531233315536572,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967333865788961,
      "loss": 7.124,
      "mean_token_accuracy": 0.10236010178923607,
      "num_tokens": 3180761.0,
      "step": 1455
    },
    {
      "entropy": 7.4003571510314945,
      "epoch": 1.5584623598505072,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004966612179875267,
      "loss": 7.0762,
      "mean_token_accuracy": 0.09902083575725555,
      "num_tokens": 3191635.0,
      "step": 1460
    },
    {
      "entropy": 7.129079866409302,
      "epoch": 1.5638013881473571,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004965882668377048,
      "loss": 7.0424,
      "mean_token_accuracy": 0.10169995054602624,
      "num_tokens": 3202452.0,
      "step": 1465
    },
    {
      "entropy": 7.34759840965271,
      "epoch": 1.569140416444207,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004965145333869789,
      "loss": 6.971,
      "mean_token_accuracy": 0.10686884224414825,
      "num_tokens": 3213298.0,
      "step": 1470
    },
    {
      "entropy": 7.113559675216675,
      "epoch": 1.574479444741057,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964400178956595,
      "loss": 7.0056,
      "mean_token_accuracy": 0.10754135549068451,
      "num_tokens": 3224314.0,
      "step": 1475
    },
    {
      "entropy": 7.2728907585144045,
      "epoch": 1.579818473037907,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004963647206268181,
      "loss": 7.0051,
      "mean_token_accuracy": 0.1074408732354641,
      "num_tokens": 3235660.0,
      "step": 1480
    },
    {
      "entropy": 7.222253370285034,
      "epoch": 1.585157501334757,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496288641846286,
      "loss": 7.0763,
      "mean_token_accuracy": 0.09584841802716255,
      "num_tokens": 3246856.0,
      "step": 1485
    },
    {
      "entropy": 7.304793119430542,
      "epoch": 1.590496529631607,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004962117818226538,
      "loss": 7.1116,
      "mean_token_accuracy": 0.10176649838685989,
      "num_tokens": 3258692.0,
      "step": 1490
    },
    {
      "entropy": 7.17184624671936,
      "epoch": 1.5958355579284569,
      "grad_norm": 0.921875,
      "learning_rate": 0.00049613414082727,
      "loss": 7.0102,
      "mean_token_accuracy": 0.10286101996898651,
      "num_tokens": 3270577.0,
      "step": 1495
    },
    {
      "entropy": 7.200855493545532,
      "epoch": 1.6011745862253068,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960557191342405,
      "loss": 6.9376,
      "mean_token_accuracy": 0.10783654004335404,
      "num_tokens": 3281262.0,
      "step": 1500
    },
    {
      "epoch": 1.6011745862253068,
      "eval_entropy": 7.068539691281009,
      "eval_loss": 7.1312689781188965,
      "eval_mean_token_accuracy": 0.10161492010796225,
      "eval_num_tokens": 3281262.0,
      "eval_runtime": 1.8156,
      "eval_samples_per_second": 1696.444,
      "eval_steps_per_second": 212.055,
      "step": 1500
    },
    {
      "entropy": 7.21118049621582,
      "epoch": 1.606513614522157,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000495976517020427,
      "loss": 6.9955,
      "mean_token_accuracy": 0.10954990237951279,
      "num_tokens": 3291552.0,
      "step": 1505
    },
    {
      "entropy": 7.275783395767212,
      "epoch": 1.611852642819007,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004958965347654469,
      "loss": 7.0386,
      "mean_token_accuracy": 0.10242254436016082,
      "num_tokens": 3301977.0,
      "step": 1510
    },
    {
      "entropy": 7.121581172943115,
      "epoch": 1.617191671115857,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004958157726516715,
      "loss": 6.9699,
      "mean_token_accuracy": 0.10966179370880128,
      "num_tokens": 3312783.0,
      "step": 1515
    },
    {
      "entropy": 7.285579776763916,
      "epoch": 1.622530699412707,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004957342309642256,
      "loss": 6.9964,
      "mean_token_accuracy": 0.10571648105978966,
      "num_tokens": 3323895.0,
      "step": 1520
    },
    {
      "entropy": 7.104770946502685,
      "epoch": 1.6278697277095568,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956519099909857,
      "loss": 6.9295,
      "mean_token_accuracy": 0.10804956257343293,
      "num_tokens": 3334103.0,
      "step": 1525
    },
    {
      "entropy": 7.208183097839355,
      "epoch": 1.6332087560064068,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004955688100225803,
      "loss": 6.9827,
      "mean_token_accuracy": 0.10653971955180168,
      "num_tokens": 3346279.0,
      "step": 1530
    },
    {
      "entropy": 7.092655944824219,
      "epoch": 1.6385477843032568,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954849313523875,
      "loss": 6.9509,
      "mean_token_accuracy": 0.10072449594736099,
      "num_tokens": 3357582.0,
      "step": 1535
    },
    {
      "entropy": 7.243341445922852,
      "epoch": 1.643886812600107,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004954002742765347,
      "loss": 6.9531,
      "mean_token_accuracy": 0.10738401338458062,
      "num_tokens": 3369297.0,
      "step": 1540
    },
    {
      "entropy": 7.092299699783325,
      "epoch": 1.649225840896957,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004953148390938973,
      "loss": 7.0684,
      "mean_token_accuracy": 0.10321348384022713,
      "num_tokens": 3382430.0,
      "step": 1545
    },
    {
      "entropy": 7.257068061828614,
      "epoch": 1.6545648691938069,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000495228626106098,
      "loss": 7.0312,
      "mean_token_accuracy": 0.11251498833298683,
      "num_tokens": 3393809.0,
      "step": 1550
    },
    {
      "entropy": 7.158068895339966,
      "epoch": 1.6599038974906568,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951416356175053,
      "loss": 7.0077,
      "mean_token_accuracy": 0.10660256892442703,
      "num_tokens": 3405164.0,
      "step": 1555
    },
    {
      "entropy": 7.227698040008545,
      "epoch": 1.6652429257875068,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004950538679352327,
      "loss": 7.0212,
      "mean_token_accuracy": 0.09924901276826859,
      "num_tokens": 3416091.0,
      "step": 1560
    },
    {
      "entropy": 7.283638668060303,
      "epoch": 1.6705819540843567,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004949653233691373,
      "loss": 7.0712,
      "mean_token_accuracy": 0.10288623869419097,
      "num_tokens": 3427675.0,
      "step": 1565
    },
    {
      "entropy": 7.135433340072632,
      "epoch": 1.6759209823812067,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004948760022318194,
      "loss": 6.9453,
      "mean_token_accuracy": 0.11070238053798676,
      "num_tokens": 3438119.0,
      "step": 1570
    },
    {
      "entropy": 7.199069452285767,
      "epoch": 1.6812600106780566,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004947859048386205,
      "loss": 6.9806,
      "mean_token_accuracy": 0.10209845751523972,
      "num_tokens": 3449151.0,
      "step": 1575
    },
    {
      "entropy": 7.132083177566528,
      "epoch": 1.6865990389749066,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004946950315076225,
      "loss": 6.8718,
      "mean_token_accuracy": 0.10959059745073318,
      "num_tokens": 3459697.0,
      "step": 1580
    },
    {
      "entropy": 7.093738555908203,
      "epoch": 1.6919380672717566,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004946033825596474,
      "loss": 6.9422,
      "mean_token_accuracy": 0.10823669135570527,
      "num_tokens": 3470409.0,
      "step": 1585
    },
    {
      "entropy": 7.1705444812774655,
      "epoch": 1.6972770955686065,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945109583182548,
      "loss": 6.9651,
      "mean_token_accuracy": 0.1007260449230671,
      "num_tokens": 3481515.0,
      "step": 1590
    },
    {
      "entropy": 7.167929172515869,
      "epoch": 1.7026161238654565,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004944177591097415,
      "loss": 6.9069,
      "mean_token_accuracy": 0.10608908087015152,
      "num_tokens": 3492683.0,
      "step": 1595
    },
    {
      "entropy": 7.120382785797119,
      "epoch": 1.7079551521623064,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004943237852631404,
      "loss": 6.9469,
      "mean_token_accuracy": 0.10530265048146248,
      "num_tokens": 3503067.0,
      "step": 1600
    },
    {
      "entropy": 7.19160099029541,
      "epoch": 1.7132941804591564,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004942290371102193,
      "loss": 6.9651,
      "mean_token_accuracy": 0.10647035390138626,
      "num_tokens": 3513611.0,
      "step": 1605
    },
    {
      "entropy": 7.181960487365723,
      "epoch": 1.7186332087560063,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004941335149854794,
      "loss": 6.9617,
      "mean_token_accuracy": 0.10861775949597359,
      "num_tokens": 3523802.0,
      "step": 1610
    },
    {
      "entropy": 7.190891218185425,
      "epoch": 1.7239722370528563,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004940372192261546,
      "loss": 6.9297,
      "mean_token_accuracy": 0.10318318530917167,
      "num_tokens": 3534764.0,
      "step": 1615
    },
    {
      "entropy": 7.121490812301635,
      "epoch": 1.7293112653497063,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004939401501722098,
      "loss": 6.9932,
      "mean_token_accuracy": 0.10756855309009553,
      "num_tokens": 3544852.0,
      "step": 1620
    },
    {
      "entropy": 7.093603610992432,
      "epoch": 1.7346502936465562,
      "grad_norm": 0.875,
      "learning_rate": 0.0004938423081663401,
      "loss": 6.7933,
      "mean_token_accuracy": 0.11004197001457214,
      "num_tokens": 3555470.0,
      "step": 1625
    },
    {
      "entropy": 7.121168518066407,
      "epoch": 1.7399893219434062,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004937436935539695,
      "loss": 6.9935,
      "mean_token_accuracy": 0.10067067667841911,
      "num_tokens": 3567497.0,
      "step": 1630
    },
    {
      "entropy": 7.194992160797119,
      "epoch": 1.7453283502402561,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004936443066832494,
      "loss": 6.9725,
      "mean_token_accuracy": 0.10623072907328605,
      "num_tokens": 3578354.0,
      "step": 1635
    },
    {
      "entropy": 7.0331216812133786,
      "epoch": 1.7506673785371063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004935441479050579,
      "loss": 6.8995,
      "mean_token_accuracy": 0.11453714296221733,
      "num_tokens": 3589369.0,
      "step": 1640
    },
    {
      "entropy": 7.2113524913787845,
      "epoch": 1.7560064068339563,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004934432175729981,
      "loss": 6.9121,
      "mean_token_accuracy": 0.10899419113993644,
      "num_tokens": 3600396.0,
      "step": 1645
    },
    {
      "entropy": 7.116479682922363,
      "epoch": 1.7613454351308062,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000493341516043397,
      "loss": 6.9588,
      "mean_token_accuracy": 0.11054658442735672,
      "num_tokens": 3610670.0,
      "step": 1650
    },
    {
      "entropy": 7.107414627075196,
      "epoch": 1.7666844634276562,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004932390436753043,
      "loss": 6.9206,
      "mean_token_accuracy": 0.10563154965639114,
      "num_tokens": 3620821.0,
      "step": 1655
    },
    {
      "entropy": 7.1294042587280275,
      "epoch": 1.7720234917245061,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004931358008304911,
      "loss": 6.8989,
      "mean_token_accuracy": 0.11248534172773361,
      "num_tokens": 3631469.0,
      "step": 1660
    },
    {
      "entropy": 7.078049898147583,
      "epoch": 1.777362520021356,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004930317878734486,
      "loss": 6.9114,
      "mean_token_accuracy": 0.1081713691353798,
      "num_tokens": 3641605.0,
      "step": 1665
    },
    {
      "entropy": 7.201143026351929,
      "epoch": 1.782701548318206,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004929270051713866,
      "loss": 6.9417,
      "mean_token_accuracy": 0.1108034186065197,
      "num_tokens": 3652107.0,
      "step": 1670
    },
    {
      "entropy": 7.0569343090057375,
      "epoch": 1.7880405766150562,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004928214530942331,
      "loss": 6.9729,
      "mean_token_accuracy": 0.10586601719260216,
      "num_tokens": 3663399.0,
      "step": 1675
    },
    {
      "entropy": 7.078392410278321,
      "epoch": 1.7933796049119062,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004927151320146313,
      "loss": 6.951,
      "mean_token_accuracy": 0.10743931010365486,
      "num_tokens": 3673638.0,
      "step": 1680
    },
    {
      "entropy": 7.162407350540161,
      "epoch": 1.7987186332087561,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004926080423079404,
      "loss": 6.9121,
      "mean_token_accuracy": 0.10548929050564766,
      "num_tokens": 3684459.0,
      "step": 1685
    },
    {
      "entropy": 7.04658784866333,
      "epoch": 1.804057661505606,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004925001843522323,
      "loss": 6.9414,
      "mean_token_accuracy": 0.10741329342126846,
      "num_tokens": 3696113.0,
      "step": 1690
    },
    {
      "entropy": 7.169419765472412,
      "epoch": 1.809396689802456,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004923915585282916,
      "loss": 6.9062,
      "mean_token_accuracy": 0.10803867951035499,
      "num_tokens": 3707345.0,
      "step": 1695
    },
    {
      "entropy": 7.098963403701783,
      "epoch": 1.814735718099306,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004922821652196136,
      "loss": 6.744,
      "mean_token_accuracy": 0.11550131812691689,
      "num_tokens": 3717573.0,
      "step": 1700
    },
    {
      "entropy": 7.035619926452637,
      "epoch": 1.820074746396156,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921720048124034,
      "loss": 6.9383,
      "mean_token_accuracy": 0.11059200093150139,
      "num_tokens": 3726658.0,
      "step": 1705
    },
    {
      "entropy": 7.158073139190674,
      "epoch": 1.825413774693006,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004920610776955742,
      "loss": 6.9495,
      "mean_token_accuracy": 0.11338067129254341,
      "num_tokens": 3737769.0,
      "step": 1710
    },
    {
      "entropy": 7.071769857406617,
      "epoch": 1.8307528029898559,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004919493842607456,
      "loss": 6.8737,
      "mean_token_accuracy": 0.11217370927333832,
      "num_tokens": 3748608.0,
      "step": 1715
    },
    {
      "entropy": 7.165196466445923,
      "epoch": 1.8360918312867058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918369249022433,
      "loss": 6.9148,
      "mean_token_accuracy": 0.11021506115794182,
      "num_tokens": 3759063.0,
      "step": 1720
    },
    {
      "entropy": 7.131874465942383,
      "epoch": 1.8414308595835558,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004917237000170966,
      "loss": 6.8688,
      "mean_token_accuracy": 0.10861963406205177,
      "num_tokens": 3769622.0,
      "step": 1725
    },
    {
      "entropy": 7.098778629302979,
      "epoch": 1.8467698878804057,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004916097100050375,
      "loss": 6.8727,
      "mean_token_accuracy": 0.11213916838169098,
      "num_tokens": 3780587.0,
      "step": 1730
    },
    {
      "entropy": 7.07066969871521,
      "epoch": 1.8521089161772557,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004914949552684992,
      "loss": 6.9236,
      "mean_token_accuracy": 0.11197843477129936,
      "num_tokens": 3792386.0,
      "step": 1735
    },
    {
      "entropy": 7.145204401016235,
      "epoch": 1.8574479444741057,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000491379436212615,
      "loss": 6.9396,
      "mean_token_accuracy": 0.11039775982499123,
      "num_tokens": 3803820.0,
      "step": 1740
    },
    {
      "entropy": 7.06385440826416,
      "epoch": 1.8627869727709556,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000491263153245216,
      "loss": 6.8982,
      "mean_token_accuracy": 0.1090501606464386,
      "num_tokens": 3815187.0,
      "step": 1745
    },
    {
      "entropy": 7.122003555297852,
      "epoch": 1.8681260010678056,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004911461067768309,
      "loss": 6.9121,
      "mean_token_accuracy": 0.10848403200507165,
      "num_tokens": 3826933.0,
      "step": 1750
    },
    {
      "entropy": 7.051048755645752,
      "epoch": 1.8734650293646555,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004910282972206834,
      "loss": 6.7997,
      "mean_token_accuracy": 0.11381994634866714,
      "num_tokens": 3836934.0,
      "step": 1755
    },
    {
      "entropy": 7.05472617149353,
      "epoch": 1.8788040576615055,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004909097249926913,
      "loss": 6.9048,
      "mean_token_accuracy": 0.10961363092064857,
      "num_tokens": 3848662.0,
      "step": 1760
    },
    {
      "entropy": 7.01745228767395,
      "epoch": 1.8841430859583554,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004907903905114652,
      "loss": 6.7385,
      "mean_token_accuracy": 0.1175423063337803,
      "num_tokens": 3860052.0,
      "step": 1765
    },
    {
      "entropy": 7.047714710235596,
      "epoch": 1.8894821142552054,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004906702941983066,
      "loss": 6.8871,
      "mean_token_accuracy": 0.10923791378736496,
      "num_tokens": 3871867.0,
      "step": 1770
    },
    {
      "entropy": 7.121882438659668,
      "epoch": 1.8948211425520554,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905494364772065,
      "loss": 6.9453,
      "mean_token_accuracy": 0.1086169384419918,
      "num_tokens": 3883612.0,
      "step": 1775
    },
    {
      "entropy": 7.0241187572479244,
      "epoch": 1.9001601708489055,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004904278177748441,
      "loss": 6.868,
      "mean_token_accuracy": 0.10942395329475403,
      "num_tokens": 3895520.0,
      "step": 1780
    },
    {
      "entropy": 7.096190881729126,
      "epoch": 1.9054991991457555,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004903054385205854,
      "loss": 6.8573,
      "mean_token_accuracy": 0.10636932700872422,
      "num_tokens": 3905898.0,
      "step": 1785
    },
    {
      "entropy": 7.115457010269165,
      "epoch": 1.9108382274426055,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004901822991464811,
      "loss": 6.9076,
      "mean_token_accuracy": 0.10664125084877014,
      "num_tokens": 3916339.0,
      "step": 1790
    },
    {
      "entropy": 7.09182596206665,
      "epoch": 1.9161772557394554,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004900584000872655,
      "loss": 6.8633,
      "mean_token_accuracy": 0.11126209571957588,
      "num_tokens": 3927131.0,
      "step": 1795
    },
    {
      "entropy": 6.981594228744507,
      "epoch": 1.9215162840363054,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004899337417803551,
      "loss": 6.8705,
      "mean_token_accuracy": 0.11410257518291474,
      "num_tokens": 3938415.0,
      "step": 1800
    },
    {
      "entropy": 7.157159852981567,
      "epoch": 1.9268553123331553,
      "grad_norm": 0.9375,
      "learning_rate": 0.000489808324665847,
      "loss": 6.8829,
      "mean_token_accuracy": 0.11425913646817207,
      "num_tokens": 3949700.0,
      "step": 1805
    },
    {
      "entropy": 7.0690977573394775,
      "epoch": 1.9321943406300055,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896821491865167,
      "loss": 6.9192,
      "mean_token_accuracy": 0.10939718559384345,
      "num_tokens": 3961879.0,
      "step": 1810
    },
    {
      "entropy": 7.1183741092681885,
      "epoch": 1.9375333689268555,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004895552157878174,
      "loss": 6.8921,
      "mean_token_accuracy": 0.1135055162012577,
      "num_tokens": 3972903.0,
      "step": 1815
    },
    {
      "entropy": 7.02628173828125,
      "epoch": 1.9428723972237054,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004894275249178782,
      "loss": 6.8852,
      "mean_token_accuracy": 0.11946139633655548,
      "num_tokens": 3984432.0,
      "step": 1820
    },
    {
      "entropy": 7.089619779586792,
      "epoch": 1.9482114255205554,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004892990770275021,
      "loss": 6.9504,
      "mean_token_accuracy": 0.10337705612182617,
      "num_tokens": 3996998.0,
      "step": 1825
    },
    {
      "entropy": 7.125215005874634,
      "epoch": 1.9535504538174053,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004891698725701648,
      "loss": 6.8477,
      "mean_token_accuracy": 0.11870014742016792,
      "num_tokens": 4007614.0,
      "step": 1830
    },
    {
      "entropy": 7.0385946273803714,
      "epoch": 1.9588894821142553,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890399120020131,
      "loss": 6.8262,
      "mean_token_accuracy": 0.10775638073682785,
      "num_tokens": 4017645.0,
      "step": 1835
    },
    {
      "entropy": 7.1667829036712645,
      "epoch": 1.9642285104111052,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004889091957818631,
      "loss": 6.9244,
      "mean_token_accuracy": 0.10332238003611564,
      "num_tokens": 4029353.0,
      "step": 1840
    },
    {
      "entropy": 7.053739356994629,
      "epoch": 1.9695675387079552,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004887777243711988,
      "loss": 6.8718,
      "mean_token_accuracy": 0.11049497202038765,
      "num_tokens": 4040421.0,
      "step": 1845
    },
    {
      "entropy": 7.0194093704223635,
      "epoch": 1.9749065670048052,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048864549823417,
      "loss": 6.8325,
      "mean_token_accuracy": 0.1106531098484993,
      "num_tokens": 4051858.0,
      "step": 1850
    },
    {
      "entropy": 7.106148815155029,
      "epoch": 1.9802455953016551,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004885125178375915,
      "loss": 6.8803,
      "mean_token_accuracy": 0.1077659197151661,
      "num_tokens": 4062063.0,
      "step": 1855
    },
    {
      "entropy": 7.0133593559265135,
      "epoch": 1.985584623598505,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004883787836509405,
      "loss": 6.6985,
      "mean_token_accuracy": 0.11615452989935875,
      "num_tokens": 4072479.0,
      "step": 1860
    },
    {
      "entropy": 7.0427937507629395,
      "epoch": 1.990923651895355,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004882442961463558,
      "loss": 6.8677,
      "mean_token_accuracy": 0.11139330714941025,
      "num_tokens": 4082912.0,
      "step": 1865
    },
    {
      "entropy": 7.077240419387818,
      "epoch": 1.996262680192205,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004881090557986353,
      "loss": 6.8434,
      "mean_token_accuracy": 0.11269779428839684,
      "num_tokens": 4093644.0,
      "step": 1870
    },
    {
      "entropy": 6.915128972795275,
      "epoch": 2.00106780565937,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004879730630852353,
      "loss": 6.7595,
      "mean_token_accuracy": 0.11212469720178181,
      "num_tokens": 4104114.0,
      "step": 1875
    },
    {
      "entropy": 7.109417867660523,
      "epoch": 2.0064068339562198,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048783631848626764,
      "loss": 6.6362,
      "mean_token_accuracy": 0.1157489076256752,
      "num_tokens": 4114847.0,
      "step": 1880
    },
    {
      "entropy": 6.914434337615967,
      "epoch": 2.0117458622530697,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004876988224844992,
      "loss": 6.4861,
      "mean_token_accuracy": 0.12341470643877983,
      "num_tokens": 4126078.0,
      "step": 1885
    },
    {
      "entropy": 7.015130233764649,
      "epoch": 2.01708489054992,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048756057556534934,
      "loss": 6.5112,
      "mean_token_accuracy": 0.12700850293040275,
      "num_tokens": 4136678.0,
      "step": 1890
    },
    {
      "entropy": 6.966462326049805,
      "epoch": 2.02242391884677,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004874215782168884,
      "loss": 6.5355,
      "mean_token_accuracy": 0.1265234611928463,
      "num_tokens": 4146794.0,
      "step": 1895
    },
    {
      "entropy": 7.0654607772827145,
      "epoch": 2.02776294714362,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004872818309298365,
      "loss": 6.571,
      "mean_token_accuracy": 0.12405075505375862,
      "num_tokens": 4156825.0,
      "step": 1900
    },
    {
      "entropy": 6.920158243179321,
      "epoch": 2.03310197544047,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048714133419756065,
      "loss": 6.5027,
      "mean_token_accuracy": 0.11929825246334076,
      "num_tokens": 4166765.0,
      "step": 1905
    },
    {
      "entropy": 7.046618700027466,
      "epoch": 2.03844100373732,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048700008851607437,
      "loss": 6.6338,
      "mean_token_accuracy": 0.12172048911452293,
      "num_tokens": 4177404.0,
      "step": 1910
    },
    {
      "entropy": 7.052170658111573,
      "epoch": 2.04378003203417,
      "grad_norm": 0.78125,
      "learning_rate": 0.00048685809438403494,
      "loss": 6.5971,
      "mean_token_accuracy": 0.12037592306733132,
      "num_tokens": 4188804.0,
      "step": 1915
    },
    {
      "entropy": 6.9027196884155275,
      "epoch": 2.04911906033102,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004867153523027422,
      "loss": 6.4989,
      "mean_token_accuracy": 0.1245061419904232,
      "num_tokens": 4200009.0,
      "step": 1920
    },
    {
      "entropy": 7.009029340744019,
      "epoch": 2.05445808862787,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004865718627761363,
      "loss": 6.5549,
      "mean_token_accuracy": 0.12319564968347549,
      "num_tokens": 4210776.0,
      "step": 1925
    },
    {
      "entropy": 6.948365020751953,
      "epoch": 2.05979711692472,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048642762631079646,
      "loss": 6.5094,
      "mean_token_accuracy": 0.13088932484388352,
      "num_tokens": 4221412.0,
      "step": 1930
    },
    {
      "entropy": 6.867213582992553,
      "epoch": 2.0651361452215697,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004862826434159388,
      "loss": 6.5012,
      "mean_token_accuracy": 0.12945465818047525,
      "num_tokens": 4231747.0,
      "step": 1935
    },
    {
      "entropy": 6.948292064666748,
      "epoch": 2.0704751735184197,
      "grad_norm": 1.0,
      "learning_rate": 0.00048613691460341456,
      "loss": 6.5181,
      "mean_token_accuracy": 0.12791630178689956,
      "num_tokens": 4241942.0,
      "step": 1940
    },
    {
      "entropy": 6.955828285217285,
      "epoch": 2.0758142018152697,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00048599044038770856,
      "loss": 6.6039,
      "mean_token_accuracy": 0.12196211889386177,
      "num_tokens": 4253151.0,
      "step": 1945
    },
    {
      "entropy": 6.93486852645874,
      "epoch": 2.0811532301121196,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004858432212859372,
      "loss": 6.5329,
      "mean_token_accuracy": 0.12614453956484795,
      "num_tokens": 4263579.0,
      "step": 1950
    },
    {
      "entropy": 6.9105040550231935,
      "epoch": 2.0864922584089696,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004856952578178464,
      "loss": 6.5513,
      "mean_token_accuracy": 0.12219369858503341,
      "num_tokens": 4274764.0,
      "step": 1955
    },
    {
      "entropy": 7.026956415176391,
      "epoch": 2.0918312867058195,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004855465505058102,
      "loss": 6.6513,
      "mean_token_accuracy": 0.11608931571245193,
      "num_tokens": 4286561.0,
      "step": 1960
    },
    {
      "entropy": 6.919632291793823,
      "epoch": 2.0971703150026695,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004853970998748289,
      "loss": 6.603,
      "mean_token_accuracy": 0.1245616041123867,
      "num_tokens": 4296101.0,
      "step": 1965
    },
    {
      "entropy": 6.937914419174194,
      "epoch": 2.1025093432995194,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048524690645252675,
      "loss": 6.5519,
      "mean_token_accuracy": 0.11856919229030609,
      "num_tokens": 4307613.0,
      "step": 1970
    },
    {
      "entropy": 6.950258350372314,
      "epoch": 2.1078483715963694,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004850959707691503,
      "loss": 6.597,
      "mean_token_accuracy": 0.12681118920445442,
      "num_tokens": 4318042.0,
      "step": 1975
    },
    {
      "entropy": 6.949284172058105,
      "epoch": 2.1131873998932194,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000484944293357567,
      "loss": 6.4691,
      "mean_token_accuracy": 0.12812729254364968,
      "num_tokens": 4328690.0,
      "step": 1980
    },
    {
      "entropy": 6.885101842880249,
      "epoch": 2.1185264281900693,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048479187475326254,
      "loss": 6.5129,
      "mean_token_accuracy": 0.1259954057633877,
      "num_tokens": 4339188.0,
      "step": 1985
    },
    {
      "entropy": 6.827868366241455,
      "epoch": 2.1238654564869193,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048463871549433947,
      "loss": 6.4728,
      "mean_token_accuracy": 0.1304394595324993,
      "num_tokens": 4350242.0,
      "step": 1990
    },
    {
      "entropy": 6.834375524520874,
      "epoch": 2.1292044847837692,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00048448481612151524,
      "loss": 6.466,
      "mean_token_accuracy": 0.13104819133877754,
      "num_tokens": 4361138.0,
      "step": 1995
    },
    {
      "entropy": 6.958746910095215,
      "epoch": 2.134543513080619,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048433017717812015,
      "loss": 6.5523,
      "mean_token_accuracy": 0.1202632024884224,
      "num_tokens": 4372157.0,
      "step": 2000
    },
    {
      "epoch": 2.134543513080619,
      "eval_entropy": 6.773753176107035,
      "eval_loss": 6.934739112854004,
      "eval_mean_token_accuracy": 0.11093968841936681,
      "eval_num_tokens": 4372157.0,
      "eval_runtime": 1.8151,
      "eval_samples_per_second": 1696.873,
      "eval_steps_per_second": 212.109,
      "step": 2000
    },
    {
      "entropy": 6.880448961257935,
      "epoch": 2.139882541377469,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004841747992100954,
      "loss": 6.5359,
      "mean_token_accuracy": 0.12958219572901725,
      "num_tokens": 4382735.0,
      "step": 2005
    },
    {
      "entropy": 6.900627279281617,
      "epoch": 2.145221569674319,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048401868276599144,
      "loss": 6.5906,
      "mean_token_accuracy": 0.11768823489546776,
      "num_tokens": 4393839.0,
      "step": 2010
    },
    {
      "entropy": 6.905943441390991,
      "epoch": 2.150560597971169,
      "grad_norm": 1.0,
      "learning_rate": 0.00048386182839696575,
      "loss": 6.5689,
      "mean_token_accuracy": 0.12359676659107208,
      "num_tokens": 4404612.0,
      "step": 2015
    },
    {
      "entropy": 6.915348386764526,
      "epoch": 2.1558996262680195,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004837042366567809,
      "loss": 6.5583,
      "mean_token_accuracy": 0.12159886509180069,
      "num_tokens": 4415969.0,
      "step": 2020
    },
    {
      "entropy": 6.947312211990356,
      "epoch": 2.161238654564869,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048354590810180306,
      "loss": 6.6298,
      "mean_token_accuracy": 0.12084809169173241,
      "num_tokens": 4427507.0,
      "step": 2025
    },
    {
      "entropy": 6.882000398635864,
      "epoch": 2.1665776828617194,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004833868432909992,
      "loss": 6.5736,
      "mean_token_accuracy": 0.12455990388989449,
      "num_tokens": 4438425.0,
      "step": 2030
    },
    {
      "entropy": 6.8933765411376955,
      "epoch": 2.1719167111585693,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048322704278593595,
      "loss": 6.5373,
      "mean_token_accuracy": 0.1295296534895897,
      "num_tokens": 4448301.0,
      "step": 2035
    },
    {
      "entropy": 6.85424017906189,
      "epoch": 2.1772557394554193,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048306650715077703,
      "loss": 6.5551,
      "mean_token_accuracy": 0.12878405377268792,
      "num_tokens": 4459214.0,
      "step": 2040
    },
    {
      "entropy": 6.8902589797973635,
      "epoch": 2.1825947677522692,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048290523695228163,
      "loss": 6.542,
      "mean_token_accuracy": 0.1292489416897297,
      "num_tokens": 4469696.0,
      "step": 2045
    },
    {
      "entropy": 6.856093168258667,
      "epoch": 2.187933796049119,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004827432327598021,
      "loss": 6.5611,
      "mean_token_accuracy": 0.12441405653953552,
      "num_tokens": 4481020.0,
      "step": 2050
    },
    {
      "entropy": 6.89211859703064,
      "epoch": 2.193272824345969,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004825804951452823,
      "loss": 6.5913,
      "mean_token_accuracy": 0.115212332457304,
      "num_tokens": 4492877.0,
      "step": 2055
    },
    {
      "entropy": 6.875864267349243,
      "epoch": 2.198611852642819,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004824170246832553,
      "loss": 6.5649,
      "mean_token_accuracy": 0.12695193216204642,
      "num_tokens": 4504184.0,
      "step": 2060
    },
    {
      "entropy": 6.8207022190094,
      "epoch": 2.203950880939669,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004822528219508413,
      "loss": 6.531,
      "mean_token_accuracy": 0.12651038020849228,
      "num_tokens": 4514807.0,
      "step": 2065
    },
    {
      "entropy": 6.828022575378418,
      "epoch": 2.209289909236519,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048208788752774587,
      "loss": 6.5443,
      "mean_token_accuracy": 0.12110874876379966,
      "num_tokens": 4526047.0,
      "step": 2070
    },
    {
      "entropy": 6.911036157608033,
      "epoch": 2.214628937533369,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048192222199625783,
      "loss": 6.4733,
      "mean_token_accuracy": 0.13001587241888046,
      "num_tokens": 4536506.0,
      "step": 2075
    },
    {
      "entropy": 6.8622376918792725,
      "epoch": 2.219967965830219,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004817558259412469,
      "loss": 6.6217,
      "mean_token_accuracy": 0.12201437205076218,
      "num_tokens": 4547351.0,
      "step": 2080
    },
    {
      "entropy": 6.931761169433594,
      "epoch": 2.225306994127069,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004815886999501621,
      "loss": 6.5862,
      "mean_token_accuracy": 0.11850906684994697,
      "num_tokens": 4558213.0,
      "step": 2085
    },
    {
      "entropy": 6.9577799320220945,
      "epoch": 2.230646022423919,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004814208446130292,
      "loss": 6.5277,
      "mean_token_accuracy": 0.12626541927456855,
      "num_tokens": 4567943.0,
      "step": 2090
    },
    {
      "entropy": 6.7742188453674315,
      "epoch": 2.235985050720769,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048125226052244904,
      "loss": 6.577,
      "mean_token_accuracy": 0.12711381241679193,
      "num_tokens": 4580000.0,
      "step": 2095
    },
    {
      "entropy": 6.921990203857422,
      "epoch": 2.2413240790176188,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004810829482735952,
      "loss": 6.5967,
      "mean_token_accuracy": 0.12330932468175888,
      "num_tokens": 4592160.0,
      "step": 2100
    },
    {
      "entropy": 6.837577629089355,
      "epoch": 2.2466631073144687,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004809129084642121,
      "loss": 6.5491,
      "mean_token_accuracy": 0.1259201429784298,
      "num_tokens": 4603747.0,
      "step": 2105
    },
    {
      "entropy": 6.867647123336792,
      "epoch": 2.2520021356113187,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004807421416946125,
      "loss": 6.5647,
      "mean_token_accuracy": 0.126456468552351,
      "num_tokens": 4615660.0,
      "step": 2110
    },
    {
      "entropy": 6.81969141960144,
      "epoch": 2.2573411639081686,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004805706485676759,
      "loss": 6.575,
      "mean_token_accuracy": 0.12571234479546547,
      "num_tokens": 4626408.0,
      "step": 2115
    },
    {
      "entropy": 6.942286539077759,
      "epoch": 2.2626801922050186,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004803984296888459,
      "loss": 6.5099,
      "mean_token_accuracy": 0.12510694563388824,
      "num_tokens": 4638236.0,
      "step": 2120
    },
    {
      "entropy": 6.795414113998413,
      "epoch": 2.2680192205018685,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048022548566612855,
      "loss": 6.5967,
      "mean_token_accuracy": 0.12317438274621964,
      "num_tokens": 4649649.0,
      "step": 2125
    },
    {
      "entropy": 6.934534072875977,
      "epoch": 2.2733582487987185,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004800518171100899,
      "loss": 6.5907,
      "mean_token_accuracy": 0.12582072243094444,
      "num_tokens": 4660502.0,
      "step": 2130
    },
    {
      "entropy": 6.865882444381714,
      "epoch": 2.2786972770955685,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004798774246338538,
      "loss": 6.5365,
      "mean_token_accuracy": 0.125477909296751,
      "num_tokens": 4671486.0,
      "step": 2135
    },
    {
      "entropy": 6.848501825332642,
      "epoch": 2.2840363053924184,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004797023088530999,
      "loss": 6.5941,
      "mean_token_accuracy": 0.125645562261343,
      "num_tokens": 4681436.0,
      "step": 2140
    },
    {
      "entropy": 6.98210334777832,
      "epoch": 2.2893753336892684,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047952647038606157,
      "loss": 6.5881,
      "mean_token_accuracy": 0.1284101627767086,
      "num_tokens": 4692065.0,
      "step": 2145
    },
    {
      "entropy": 6.836091232299805,
      "epoch": 2.2947143619861183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004793499098535233,
      "loss": 6.5034,
      "mean_token_accuracy": 0.1249152347445488,
      "num_tokens": 4702107.0,
      "step": 2150
    },
    {
      "entropy": 6.831265258789062,
      "epoch": 2.3000533902829687,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004791726278788191,
      "loss": 6.6077,
      "mean_token_accuracy": 0.12131517902016639,
      "num_tokens": 4712060.0,
      "step": 2155
    },
    {
      "entropy": 6.904486179351807,
      "epoch": 2.3053924185798182,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004789946250878296,
      "loss": 6.5162,
      "mean_token_accuracy": 0.13010889962315558,
      "num_tokens": 4721805.0,
      "step": 2160
    },
    {
      "entropy": 6.740131425857544,
      "epoch": 2.3107314468766686,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004788159021089805,
      "loss": 6.5035,
      "mean_token_accuracy": 0.1257294774055481,
      "num_tokens": 4733630.0,
      "step": 2165
    },
    {
      "entropy": 6.8522107124328615,
      "epoch": 2.3160704751735186,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004786364595732401,
      "loss": 6.5728,
      "mean_token_accuracy": 0.12956593036651612,
      "num_tokens": 4743634.0,
      "step": 2170
    },
    {
      "entropy": 6.872754287719727,
      "epoch": 2.3214095034703686,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004784562981141168,
      "loss": 6.5247,
      "mean_token_accuracy": 0.12224958017468453,
      "num_tokens": 4754815.0,
      "step": 2175
    },
    {
      "entropy": 6.810030364990235,
      "epoch": 2.3267485317672185,
      "grad_norm": 1.0,
      "learning_rate": 0.0004782754183676573,
      "loss": 6.5851,
      "mean_token_accuracy": 0.12206983789801598,
      "num_tokens": 4765657.0,
      "step": 2180
    },
    {
      "entropy": 6.9540458679199215,
      "epoch": 2.3320875600640685,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047809382097244407,
      "loss": 6.6016,
      "mean_token_accuracy": 0.12477592378854752,
      "num_tokens": 4777436.0,
      "step": 2185
    },
    {
      "entropy": 6.874520349502563,
      "epoch": 2.3374265883609184,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047791150656959326,
      "loss": 6.587,
      "mean_token_accuracy": 0.12307127118110657,
      "num_tokens": 4788149.0,
      "step": 2190
    },
    {
      "entropy": 6.784716367721558,
      "epoch": 2.3427656166577684,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047772847580275225,
      "loss": 6.5332,
      "mean_token_accuracy": 0.12545752599835397,
      "num_tokens": 4799512.0,
      "step": 2195
    },
    {
      "entropy": 6.875580453872681,
      "epoch": 2.3481046449546183,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004775447293180976,
      "loss": 6.5423,
      "mean_token_accuracy": 0.12484646216034889,
      "num_tokens": 4810410.0,
      "step": 2200
    },
    {
      "entropy": 6.906468296051026,
      "epoch": 2.3534436732514683,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047736026776433254,
      "loss": 6.6885,
      "mean_token_accuracy": 0.11681632474064826,
      "num_tokens": 4822247.0,
      "step": 2205
    },
    {
      "entropy": 6.934691524505615,
      "epoch": 2.3587827015483183,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047717509179268504,
      "loss": 6.555,
      "mean_token_accuracy": 0.1287574701011181,
      "num_tokens": 4833550.0,
      "step": 2210
    },
    {
      "entropy": 6.831350469589234,
      "epoch": 2.364121729845168,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047698920205690494,
      "loss": 6.5855,
      "mean_token_accuracy": 0.12659201845526696,
      "num_tokens": 4845095.0,
      "step": 2215
    },
    {
      "entropy": 6.821032953262329,
      "epoch": 2.369460758142018,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047680259921326225,
      "loss": 6.504,
      "mean_token_accuracy": 0.12526951134204864,
      "num_tokens": 4856641.0,
      "step": 2220
    },
    {
      "entropy": 6.866345167160034,
      "epoch": 2.374799786438868,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004766152839205443,
      "loss": 6.5542,
      "mean_token_accuracy": 0.12299052476882935,
      "num_tokens": 4868274.0,
      "step": 2225
    },
    {
      "entropy": 6.865297317504883,
      "epoch": 2.380138814735718,
      "grad_norm": 0.984375,
      "learning_rate": 0.000476427256840054,
      "loss": 6.5413,
      "mean_token_accuracy": 0.12318603098392486,
      "num_tokens": 4878308.0,
      "step": 2230
    },
    {
      "entropy": 6.845654821395874,
      "epoch": 2.385477843032568,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004762385186356068,
      "loss": 6.5951,
      "mean_token_accuracy": 0.12697439640760422,
      "num_tokens": 4888811.0,
      "step": 2235
    },
    {
      "entropy": 6.8985631465911865,
      "epoch": 2.390816871329418,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004760490699735289,
      "loss": 6.5085,
      "mean_token_accuracy": 0.13273092210292817,
      "num_tokens": 4898719.0,
      "step": 2240
    },
    {
      "entropy": 6.801689481735229,
      "epoch": 2.396155899626268,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004758589115226549,
      "loss": 6.5474,
      "mean_token_accuracy": 0.12448980137705803,
      "num_tokens": 4909480.0,
      "step": 2245
    },
    {
      "entropy": 6.8546003818511965,
      "epoch": 2.401494927923118,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047566804395432493,
      "loss": 6.5932,
      "mean_token_accuracy": 0.1219810850918293,
      "num_tokens": 4920560.0,
      "step": 2250
    },
    {
      "entropy": 6.816539764404297,
      "epoch": 2.406833956219968,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047547646794238277,
      "loss": 6.4552,
      "mean_token_accuracy": 0.12189477160573006,
      "num_tokens": 4931499.0,
      "step": 2255
    },
    {
      "entropy": 6.754684591293335,
      "epoch": 2.412172984516818,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047528418416317334,
      "loss": 6.5083,
      "mean_token_accuracy": 0.13347715884447098,
      "num_tokens": 4942133.0,
      "step": 2260
    },
    {
      "entropy": 6.8414233207702635,
      "epoch": 2.417512012813668,
      "grad_norm": 1.0,
      "learning_rate": 0.0004750911932955402,
      "loss": 6.5734,
      "mean_token_accuracy": 0.12275219932198525,
      "num_tokens": 4953255.0,
      "step": 2265
    },
    {
      "entropy": 6.86159462928772,
      "epoch": 2.4228510411105177,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047489749602082316,
      "loss": 6.5256,
      "mean_token_accuracy": 0.12575036883354188,
      "num_tokens": 4964766.0,
      "step": 2270
    },
    {
      "entropy": 6.822673892974853,
      "epoch": 2.4281900694073677,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047470309302285617,
      "loss": 6.5921,
      "mean_token_accuracy": 0.12224113792181016,
      "num_tokens": 4976644.0,
      "step": 2275
    },
    {
      "entropy": 6.7894798755645756,
      "epoch": 2.4335290977042177,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047450798498796466,
      "loss": 6.5303,
      "mean_token_accuracy": 0.12349486351013184,
      "num_tokens": 4986976.0,
      "step": 2280
    },
    {
      "entropy": 6.881432390213012,
      "epoch": 2.4388681260010676,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004743121726049628,
      "loss": 6.5324,
      "mean_token_accuracy": 0.12652337923645973,
      "num_tokens": 4998747.0,
      "step": 2285
    },
    {
      "entropy": 6.814818048477173,
      "epoch": 2.444207154297918,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004741156565651518,
      "loss": 6.5198,
      "mean_token_accuracy": 0.13365631699562072,
      "num_tokens": 5009323.0,
      "step": 2290
    },
    {
      "entropy": 6.901062488555908,
      "epoch": 2.4495461825947675,
      "grad_norm": 1.0,
      "learning_rate": 0.00047391843756231693,
      "loss": 6.5853,
      "mean_token_accuracy": 0.12531132400035858,
      "num_tokens": 5019434.0,
      "step": 2295
    },
    {
      "entropy": 6.782677125930786,
      "epoch": 2.454885210891618,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047372051629272516,
      "loss": 6.4909,
      "mean_token_accuracy": 0.1259647473692894,
      "num_tokens": 5029884.0,
      "step": 2300
    },
    {
      "entropy": 6.791875982284546,
      "epoch": 2.460224239188468,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047352189345512284,
      "loss": 6.4898,
      "mean_token_accuracy": 0.13430592641234398,
      "num_tokens": 5040323.0,
      "step": 2305
    },
    {
      "entropy": 6.763852119445801,
      "epoch": 2.465563267485318,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004733225697507332,
      "loss": 6.5126,
      "mean_token_accuracy": 0.12424452975392342,
      "num_tokens": 5051363.0,
      "step": 2310
    },
    {
      "entropy": 6.80372052192688,
      "epoch": 2.470902295782168,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004731225458832537,
      "loss": 6.4255,
      "mean_token_accuracy": 0.13885923177003862,
      "num_tokens": 5061221.0,
      "step": 2315
    },
    {
      "entropy": 6.77863450050354,
      "epoch": 2.4762413240790178,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047292182255885367,
      "loss": 6.4959,
      "mean_token_accuracy": 0.13018832132220268,
      "num_tokens": 5072578.0,
      "step": 2320
    },
    {
      "entropy": 6.769831752777099,
      "epoch": 2.4815803523758677,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004727204004861721,
      "loss": 6.4961,
      "mean_token_accuracy": 0.12298517078161239,
      "num_tokens": 5083244.0,
      "step": 2325
    },
    {
      "entropy": 6.7560343742370605,
      "epoch": 2.4869193806727177,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004725182803763144,
      "loss": 6.5339,
      "mean_token_accuracy": 0.1289779581129551,
      "num_tokens": 5094115.0,
      "step": 2330
    },
    {
      "entropy": 6.829773807525635,
      "epoch": 2.4922584089695676,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004723154629428507,
      "loss": 6.5461,
      "mean_token_accuracy": 0.12436275333166122,
      "num_tokens": 5104838.0,
      "step": 2335
    },
    {
      "entropy": 6.859137296676636,
      "epoch": 2.4975974372664176,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004721119489018128,
      "loss": 6.5327,
      "mean_token_accuracy": 0.12045159563422203,
      "num_tokens": 5115078.0,
      "step": 2340
    },
    {
      "entropy": 6.797474527359009,
      "epoch": 2.5029364655632675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047190773897169203,
      "loss": 6.5572,
      "mean_token_accuracy": 0.12596193179488183,
      "num_tokens": 5126414.0,
      "step": 2345
    },
    {
      "entropy": 6.828856039047241,
      "epoch": 2.5082754938601175,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047170283387343634,
      "loss": 6.562,
      "mean_token_accuracy": 0.12051519602537156,
      "num_tokens": 5136271.0,
      "step": 2350
    },
    {
      "entropy": 6.800413131713867,
      "epoch": 2.5136145221569675,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047149723433044787,
      "loss": 6.567,
      "mean_token_accuracy": 0.12625120356678962,
      "num_tokens": 5146780.0,
      "step": 2355
    },
    {
      "entropy": 6.86309061050415,
      "epoch": 2.5189535504538174,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00047129094106858056,
      "loss": 6.5779,
      "mean_token_accuracy": 0.12552282735705375,
      "num_tokens": 5157941.0,
      "step": 2360
    },
    {
      "entropy": 6.79384446144104,
      "epoch": 2.5242925787506674,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047108395481613736,
      "loss": 6.5004,
      "mean_token_accuracy": 0.1237537682056427,
      "num_tokens": 5168994.0,
      "step": 2365
    },
    {
      "entropy": 6.758298110961914,
      "epoch": 2.5296316070475173,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00047087627630386803,
      "loss": 6.4828,
      "mean_token_accuracy": 0.1223898023366928,
      "num_tokens": 5180780.0,
      "step": 2370
    },
    {
      "entropy": 6.754440212249756,
      "epoch": 2.5349706353443673,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000470667906264966,
      "loss": 6.5634,
      "mean_token_accuracy": 0.12323177009820938,
      "num_tokens": 5192090.0,
      "step": 2375
    },
    {
      "entropy": 6.85933289527893,
      "epoch": 2.5403096636412172,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004704588454350661,
      "loss": 6.4706,
      "mean_token_accuracy": 0.1317416600883007,
      "num_tokens": 5201854.0,
      "step": 2380
    },
    {
      "entropy": 6.773879671096802,
      "epoch": 2.545648691938067,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004702490945522423,
      "loss": 6.4941,
      "mean_token_accuracy": 0.13044386208057404,
      "num_tokens": 5213998.0,
      "step": 2385
    },
    {
      "entropy": 6.708115959167481,
      "epoch": 2.550987720234917,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004700386543570043,
      "loss": 6.4527,
      "mean_token_accuracy": 0.12814613208174705,
      "num_tokens": 5225217.0,
      "step": 2390
    },
    {
      "entropy": 6.704124879837036,
      "epoch": 2.556326748531767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004698275255922956,
      "loss": 6.3849,
      "mean_token_accuracy": 0.1327833689749241,
      "num_tokens": 5235098.0,
      "step": 2395
    },
    {
      "entropy": 6.790454530715943,
      "epoch": 2.561665776828617,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046961570900349073,
      "loss": 6.5778,
      "mean_token_accuracy": 0.12630253434181213,
      "num_tokens": 5246704.0,
      "step": 2400
    },
    {
      "entropy": 6.754661941528321,
      "epoch": 2.567004805125467,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004694032053383922,
      "loss": 6.4673,
      "mean_token_accuracy": 0.13032191917300223,
      "num_tokens": 5257331.0,
      "step": 2405
    },
    {
      "entropy": 6.766871213912964,
      "epoch": 2.572343833422317,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004691900153472286,
      "loss": 6.4521,
      "mean_token_accuracy": 0.13126400411128997,
      "num_tokens": 5267662.0,
      "step": 2410
    },
    {
      "entropy": 6.807767772674561,
      "epoch": 2.5776828617191674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046897613978265124,
      "loss": 6.4927,
      "mean_token_accuracy": 0.12516679093241692,
      "num_tokens": 5277977.0,
      "step": 2415
    },
    {
      "entropy": 6.830004692077637,
      "epoch": 2.583021890016017,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004687615793997319,
      "loss": 6.6179,
      "mean_token_accuracy": 0.11929458528757095,
      "num_tokens": 5288987.0,
      "step": 2420
    },
    {
      "entropy": 6.791280221939087,
      "epoch": 2.5883609183128673,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004685463349559602,
      "loss": 6.4433,
      "mean_token_accuracy": 0.12909359782934188,
      "num_tokens": 5300300.0,
      "step": 2425
    },
    {
      "entropy": 6.760449028015136,
      "epoch": 2.593699946609717,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00046833040721124035,
      "loss": 6.464,
      "mean_token_accuracy": 0.12651569321751593,
      "num_tokens": 5311606.0,
      "step": 2430
    },
    {
      "entropy": 6.779611396789551,
      "epoch": 2.599038974906567,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004681137969278896,
      "loss": 6.608,
      "mean_token_accuracy": 0.11662786602973937,
      "num_tokens": 5323218.0,
      "step": 2435
    },
    {
      "entropy": 6.729658365249634,
      "epoch": 2.6043780032034167,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004678965048706341,
      "loss": 6.459,
      "mean_token_accuracy": 0.13026198297739028,
      "num_tokens": 5334516.0,
      "step": 2440
    },
    {
      "entropy": 6.765009355545044,
      "epoch": 2.609717031500267,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004676785318066075,
      "loss": 6.5194,
      "mean_token_accuracy": 0.12988420128822326,
      "num_tokens": 5345751.0,
      "step": 2445
    },
    {
      "entropy": 6.8113123893737795,
      "epoch": 2.615056059797117,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004674598785053475,
      "loss": 6.5869,
      "mean_token_accuracy": 0.12195999845862389,
      "num_tokens": 5356918.0,
      "step": 2450
    },
    {
      "entropy": 6.801035404205322,
      "epoch": 2.620395088093967,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046724054573879314,
      "loss": 6.4824,
      "mean_token_accuracy": 0.13370697051286698,
      "num_tokens": 5368175.0,
      "step": 2455
    },
    {
      "entropy": 6.681986713409424,
      "epoch": 2.625734116390817,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004670205342812827,
      "loss": 6.4613,
      "mean_token_accuracy": 0.128626449406147,
      "num_tokens": 5379074.0,
      "step": 2460
    },
    {
      "entropy": 6.836401033401489,
      "epoch": 2.631073144687667,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00046679984490955015,
      "loss": 6.5203,
      "mean_token_accuracy": 0.12719035670161247,
      "num_tokens": 5389380.0,
      "step": 2465
    },
    {
      "entropy": 6.695728015899658,
      "epoch": 2.636412172984517,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004665784784027229,
      "loss": 6.5465,
      "mean_token_accuracy": 0.1358778417110443,
      "num_tokens": 5401378.0,
      "step": 2470
    },
    {
      "entropy": 6.82390513420105,
      "epoch": 2.641751201281367,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004663564355423189,
      "loss": 6.4823,
      "mean_token_accuracy": 0.12906705513596534,
      "num_tokens": 5411434.0,
      "step": 2475
    },
    {
      "entropy": 6.819698429107666,
      "epoch": 2.647090229578217,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004661337171122442,
      "loss": 6.4761,
      "mean_token_accuracy": 0.1277168557047844,
      "num_tokens": 5421597.0,
      "step": 2480
    },
    {
      "entropy": 6.689829397201538,
      "epoch": 2.652429257875067,
      "grad_norm": 0.828125,
      "learning_rate": 0.00046591032389878934,
      "loss": 6.485,
      "mean_token_accuracy": 0.126693993806839,
      "num_tokens": 5432166.0,
      "step": 2485
    },
    {
      "entropy": 6.698560667037964,
      "epoch": 2.6577682861719167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004656862566906276,
      "loss": 6.4949,
      "mean_token_accuracy": 0.12689822018146515,
      "num_tokens": 5443156.0,
      "step": 2490
    },
    {
      "entropy": 6.902831506729126,
      "epoch": 2.6631073144687667,
      "grad_norm": 0.921875,
      "learning_rate": 0.00046546151627881166,
      "loss": 6.5727,
      "mean_token_accuracy": 0.11838782355189323,
      "num_tokens": 5453830.0,
      "step": 2495
    },
    {
      "entropy": 6.678612470626831,
      "epoch": 2.6684463427656167,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004652361034567708,
      "loss": 6.4709,
      "mean_token_accuracy": 0.1326916351914406,
      "num_tokens": 5465371.0,
      "step": 2500
    },
    {
      "epoch": 2.6684463427656167,
      "eval_entropy": 6.506537835009686,
      "eval_loss": 6.804722785949707,
      "eval_mean_token_accuracy": 0.11863555662430726,
      "eval_num_tokens": 5465371.0,
      "eval_runtime": 1.8173,
      "eval_samples_per_second": 1694.823,
      "eval_steps_per_second": 211.853,
      "step": 2500
    },
    {
      "entropy": 6.768008804321289,
      "epoch": 2.6737853710624666,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046501001902030814,
      "loss": 6.5267,
      "mean_token_accuracy": 0.1259946919977665,
      "num_tokens": 5476687.0,
      "step": 2505
    },
    {
      "entropy": 6.789381408691407,
      "epoch": 2.6791243993593166,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004647832637675982,
      "loss": 6.4999,
      "mean_token_accuracy": 0.1216110959649086,
      "num_tokens": 5487825.0,
      "step": 2510
    },
    {
      "entropy": 6.750199174880981,
      "epoch": 2.6844634276561665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046455583849918324,
      "loss": 6.4768,
      "mean_token_accuracy": 0.13176634460687636,
      "num_tokens": 5498561.0,
      "step": 2515
    },
    {
      "entropy": 6.7578329086303714,
      "epoch": 2.6898024559530165,
      "grad_norm": 1.0,
      "learning_rate": 0.0004643277440179715,
      "loss": 6.4973,
      "mean_token_accuracy": 0.12797005027532576,
      "num_tokens": 5509624.0,
      "step": 2520
    },
    {
      "entropy": 6.720186614990235,
      "epoch": 2.6951414842498664,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004640989811292335,
      "loss": 6.3996,
      "mean_token_accuracy": 0.13824927657842637,
      "num_tokens": 5519484.0,
      "step": 2525
    },
    {
      "entropy": 6.689453601837158,
      "epoch": 2.7004805125467164,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004638695506405996,
      "loss": 6.3951,
      "mean_token_accuracy": 0.132322496175766,
      "num_tokens": 5531839.0,
      "step": 2530
    },
    {
      "entropy": 6.736231422424316,
      "epoch": 2.7058195408435664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004636394533620571,
      "loss": 6.5275,
      "mean_token_accuracy": 0.1268465995788574,
      "num_tokens": 5541787.0,
      "step": 2535
    },
    {
      "entropy": 6.809881973266601,
      "epoch": 2.7111585691404163,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004634086901059474,
      "loss": 6.582,
      "mean_token_accuracy": 0.1202095814049244,
      "num_tokens": 5553493.0,
      "step": 2540
    },
    {
      "entropy": 6.805474138259887,
      "epoch": 2.7164975974372663,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004631772616869629,
      "loss": 6.4382,
      "mean_token_accuracy": 0.12904566749930382,
      "num_tokens": 5564809.0,
      "step": 2545
    },
    {
      "entropy": 6.642514276504516,
      "epoch": 2.7218366257341167,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046294516892214454,
      "loss": 6.4544,
      "mean_token_accuracy": 0.12669157981872559,
      "num_tokens": 5575575.0,
      "step": 2550
    },
    {
      "entropy": 6.770498991012573,
      "epoch": 2.727175654030966,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00046271241263087846,
      "loss": 6.4719,
      "mean_token_accuracy": 0.12543186768889428,
      "num_tokens": 5587276.0,
      "step": 2555
    },
    {
      "entropy": 6.732878065109253,
      "epoch": 2.7325146823278166,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004624789936348936,
      "loss": 6.4318,
      "mean_token_accuracy": 0.1274913065135479,
      "num_tokens": 5597557.0,
      "step": 2560
    },
    {
      "entropy": 6.679828453063965,
      "epoch": 2.737853710624666,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046224491275825825,
      "loss": 6.4392,
      "mean_token_accuracy": 0.13149973079562188,
      "num_tokens": 5608512.0,
      "step": 2565
    },
    {
      "entropy": 6.733901882171631,
      "epoch": 2.7431927389215165,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004620101708273776,
      "loss": 6.5047,
      "mean_token_accuracy": 0.13109129667282104,
      "num_tokens": 5621788.0,
      "step": 2570
    },
    {
      "entropy": 6.806593799591065,
      "epoch": 2.748531767218366,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046177476867099055,
      "loss": 6.4933,
      "mean_token_accuracy": 0.12798713520169258,
      "num_tokens": 5632493.0,
      "step": 2575
    },
    {
      "entropy": 6.739014434814453,
      "epoch": 2.7538707955152164,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046153870712016694,
      "loss": 6.5157,
      "mean_token_accuracy": 0.12476658746600151,
      "num_tokens": 5643163.0,
      "step": 2580
    },
    {
      "entropy": 6.717135906219482,
      "epoch": 2.7592098238120664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004613019870083045,
      "loss": 6.4407,
      "mean_token_accuracy": 0.1334136761724949,
      "num_tokens": 5653362.0,
      "step": 2585
    },
    {
      "entropy": 6.775650072097778,
      "epoch": 2.7645488521089163,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004610646091711261,
      "loss": 6.5442,
      "mean_token_accuracy": 0.13016111105680467,
      "num_tokens": 5663637.0,
      "step": 2590
    },
    {
      "entropy": 6.6957603931427006,
      "epoch": 2.7698878804057663,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004608265744466763,
      "loss": 6.4834,
      "mean_token_accuracy": 0.1269044943153858,
      "num_tokens": 5674446.0,
      "step": 2595
    },
    {
      "entropy": 6.8000424861907955,
      "epoch": 2.7752269087026162,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046058788367531927,
      "loss": 6.5139,
      "mean_token_accuracy": 0.13062895238399505,
      "num_tokens": 5684768.0,
      "step": 2600
    },
    {
      "entropy": 6.765134954452515,
      "epoch": 2.780565936999466,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004603485376997349,
      "loss": 6.4089,
      "mean_token_accuracy": 0.13543231412768364,
      "num_tokens": 5696474.0,
      "step": 2605
    },
    {
      "entropy": 6.602987575531006,
      "epoch": 2.785904965296316,
      "grad_norm": 1.0,
      "learning_rate": 0.0004601085373649163,
      "loss": 6.5767,
      "mean_token_accuracy": 0.12214590460062028,
      "num_tokens": 5707829.0,
      "step": 2610
    },
    {
      "entropy": 6.819445943832397,
      "epoch": 2.791243993593166,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000459867883518167,
      "loss": 6.5069,
      "mean_token_accuracy": 0.12737508937716485,
      "num_tokens": 5719325.0,
      "step": 2615
    },
    {
      "entropy": 6.790378141403198,
      "epoch": 2.796583021890016,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004596265770090974,
      "loss": 6.4816,
      "mean_token_accuracy": 0.12503273263573647,
      "num_tokens": 5730859.0,
      "step": 2620
    },
    {
      "entropy": 6.709441614151001,
      "epoch": 2.801922050186866,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004593846186896222,
      "loss": 6.4519,
      "mean_token_accuracy": 0.132213544100523,
      "num_tokens": 5741595.0,
      "step": 2625
    },
    {
      "entropy": 6.777496957778931,
      "epoch": 2.807261078483716,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00045914200941395725,
      "loss": 6.4752,
      "mean_token_accuracy": 0.12735849544405936,
      "num_tokens": 5752989.0,
      "step": 2630
    },
    {
      "entropy": 6.731167650222778,
      "epoch": 2.812600106780566,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004588987500386165,
      "loss": 6.4721,
      "mean_token_accuracy": 0.12772611528635025,
      "num_tokens": 5764440.0,
      "step": 2635
    },
    {
      "entropy": 6.696313142776489,
      "epoch": 2.817939135077416,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004586548414224093,
      "loss": 6.3725,
      "mean_token_accuracy": 0.13323593363165856,
      "num_tokens": 5776389.0,
      "step": 2640
    },
    {
      "entropy": 6.725729465484619,
      "epoch": 2.823278163374266,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045841028442643675,
      "loss": 6.4058,
      "mean_token_accuracy": 0.13175813257694244,
      "num_tokens": 5787803.0,
      "step": 2645
    },
    {
      "entropy": 6.698521423339844,
      "epoch": 2.828617191671116,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004581650799140891,
      "loss": 6.4867,
      "mean_token_accuracy": 0.12856226190924644,
      "num_tokens": 5799647.0,
      "step": 2650
    },
    {
      "entropy": 6.737258195877075,
      "epoch": 2.8339562199679658,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004579192287510427,
      "loss": 6.4974,
      "mean_token_accuracy": 0.12667600736021994,
      "num_tokens": 5809968.0,
      "step": 2655
    },
    {
      "entropy": 6.778554439544678,
      "epoch": 2.8392952482648157,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004576727318052568,
      "loss": 6.465,
      "mean_token_accuracy": 0.13686817660927772,
      "num_tokens": 5820075.0,
      "step": 2660
    },
    {
      "entropy": 6.646070718765259,
      "epoch": 2.8446342765616657,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00045742558994697063,
      "loss": 6.4838,
      "mean_token_accuracy": 0.13021821826696395,
      "num_tokens": 5830686.0,
      "step": 2665
    },
    {
      "entropy": 6.814020347595215,
      "epoch": 2.8499733048585156,
      "grad_norm": 1.0,
      "learning_rate": 0.0004571778040487001,
      "loss": 6.458,
      "mean_token_accuracy": 0.13131239861249924,
      "num_tokens": 5841451.0,
      "step": 2670
    },
    {
      "entropy": 6.666039228439331,
      "epoch": 2.8553123331553656,
      "grad_norm": 1.0,
      "learning_rate": 0.0004569293749852347,
      "loss": 6.4986,
      "mean_token_accuracy": 0.1276731461286545,
      "num_tokens": 5852650.0,
      "step": 2675
    },
    {
      "entropy": 6.715946388244629,
      "epoch": 2.8606513614522155,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004566803036336351,
      "loss": 6.4434,
      "mean_token_accuracy": 0.1310188926756382,
      "num_tokens": 5865056.0,
      "step": 2680
    },
    {
      "entropy": 6.740395164489746,
      "epoch": 2.865990389749066,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045643059087322895,
      "loss": 6.4935,
      "mean_token_accuracy": 0.12656164914369583,
      "num_tokens": 5876069.0,
      "step": 2685
    },
    {
      "entropy": 6.742620897293091,
      "epoch": 2.8713294180459155,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004561802375856088,
      "loss": 6.4406,
      "mean_token_accuracy": 0.1361336812376976,
      "num_tokens": 5886366.0,
      "step": 2690
    },
    {
      "entropy": 6.720592260360718,
      "epoch": 2.876668446342766,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004559292446546281,
      "loss": 6.508,
      "mean_token_accuracy": 0.1300479516386986,
      "num_tokens": 5897101.0,
      "step": 2695
    },
    {
      "entropy": 6.787854623794556,
      "epoch": 2.8820074746396154,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00045567761296639887,
      "loss": 6.5176,
      "mean_token_accuracy": 0.1308740459382534,
      "num_tokens": 5908748.0,
      "step": 2700
    },
    {
      "entropy": 6.783561897277832,
      "epoch": 2.8873465029364658,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000455425343409288,
      "loss": 6.545,
      "mean_token_accuracy": 0.12536448761820793,
      "num_tokens": 5920830.0,
      "step": 2705
    },
    {
      "entropy": 6.626713418960572,
      "epoch": 2.8926855312333153,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045517243687391454,
      "loss": 6.3492,
      "mean_token_accuracy": 0.13662005811929703,
      "num_tokens": 5932360.0,
      "step": 2710
    },
    {
      "entropy": 6.75290880203247,
      "epoch": 2.8980245595301657,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004549188942531461,
      "loss": 6.4833,
      "mean_token_accuracy": 0.12693905979394912,
      "num_tokens": 5944067.0,
      "step": 2715
    },
    {
      "entropy": 6.806321859359741,
      "epoch": 2.9033635878270156,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000454664716442096,
      "loss": 6.5369,
      "mean_token_accuracy": 0.1230388343334198,
      "num_tokens": 5955974.0,
      "step": 2720
    },
    {
      "entropy": 6.792594861984253,
      "epoch": 2.9087026161238656,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00045440990433812025,
      "loss": 6.5309,
      "mean_token_accuracy": 0.1249440647661686,
      "num_tokens": 5966542.0,
      "step": 2725
    },
    {
      "entropy": 6.660552597045898,
      "epoch": 2.9140416444207156,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000454154458840814,
      "loss": 6.4113,
      "mean_token_accuracy": 0.1303204596042633,
      "num_tokens": 5977342.0,
      "step": 2730
    },
    {
      "entropy": 6.760577392578125,
      "epoch": 2.9193806727175655,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004538983808520085,
      "loss": 6.5565,
      "mean_token_accuracy": 0.12395976111292839,
      "num_tokens": 5988786.0,
      "step": 2735
    },
    {
      "entropy": 6.788736200332641,
      "epoch": 2.9247197010144155,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00045364167127576817,
      "loss": 6.5493,
      "mean_token_accuracy": 0.11930835843086243,
      "num_tokens": 6000167.0,
      "step": 2740
    },
    {
      "entropy": 6.671005630493164,
      "epoch": 2.9300587293112654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045338433101838705,
      "loss": 6.4336,
      "mean_token_accuracy": 0.12854471430182457,
      "num_tokens": 6012214.0,
      "step": 2745
    },
    {
      "entropy": 6.764946031570434,
      "epoch": 2.9353977576081154,
      "grad_norm": 0.921875,
      "learning_rate": 0.00045312636098838576,
      "loss": 6.524,
      "mean_token_accuracy": 0.13039105236530305,
      "num_tokens": 6023032.0,
      "step": 2750
    },
    {
      "entropy": 6.726640939712524,
      "epoch": 2.9407367859049653,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004528677620965083,
      "loss": 6.4315,
      "mean_token_accuracy": 0.13164386823773383,
      "num_tokens": 6033398.0,
      "step": 2755
    },
    {
      "entropy": 6.7339318752288815,
      "epoch": 2.9460758142018153,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045260853525571894,
      "loss": 6.4657,
      "mean_token_accuracy": 0.12935673370957373,
      "num_tokens": 6044549.0,
      "step": 2760
    },
    {
      "entropy": 6.581913232803345,
      "epoch": 2.9514148424986653,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004523486813811987,
      "loss": 6.2531,
      "mean_token_accuracy": 0.13347729817032813,
      "num_tokens": 6055213.0,
      "step": 2765
    },
    {
      "entropy": 6.656859970092773,
      "epoch": 2.956753870795515,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004520882013903423,
      "loss": 6.3928,
      "mean_token_accuracy": 0.13870861008763313,
      "num_tokens": 6065488.0,
      "step": 2770
    },
    {
      "entropy": 6.702912044525147,
      "epoch": 2.962092899092365,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045182709620275504,
      "loss": 6.4888,
      "mean_token_accuracy": 0.1314636677503586,
      "num_tokens": 6077525.0,
      "step": 2775
    },
    {
      "entropy": 6.749990224838257,
      "epoch": 2.967431927389215,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004515653667402494,
      "loss": 6.4758,
      "mean_token_accuracy": 0.1329290971159935,
      "num_tokens": 6088707.0,
      "step": 2780
    },
    {
      "entropy": 6.70402946472168,
      "epoch": 2.972770955686065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004513030139268418,
      "loss": 6.4342,
      "mean_token_accuracy": 0.13106571063399314,
      "num_tokens": 6099231.0,
      "step": 2785
    },
    {
      "entropy": 6.698221158981323,
      "epoch": 2.978109983982915,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004510400386887493,
      "loss": 6.5101,
      "mean_token_accuracy": 0.12947579026222228,
      "num_tokens": 6110356.0,
      "step": 2790
    },
    {
      "entropy": 6.7977382183074955,
      "epoch": 2.983449012279765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045077644195438645,
      "loss": 6.4675,
      "mean_token_accuracy": 0.1286630518734455,
      "num_tokens": 6120722.0,
      "step": 2795
    },
    {
      "entropy": 6.746077966690064,
      "epoch": 2.988788040576615,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000450512224654362,
      "loss": 6.4564,
      "mean_token_accuracy": 0.13280715346336364,
      "num_tokens": 6131404.0,
      "step": 2800
    },
    {
      "entropy": 6.678772449493408,
      "epoch": 2.994127068873465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045024738772147534,
      "loss": 6.4695,
      "mean_token_accuracy": 0.12211451306939125,
      "num_tokens": 6141402.0,
      "step": 2805
    },
    {
      "entropy": 6.752857732772827,
      "epoch": 2.999466097170315,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00044998193209071365,
      "loss": 6.407,
      "mean_token_accuracy": 0.13422416672110557,
      "num_tokens": 6152082.0,
      "step": 2810
    },
    {
      "entropy": 6.670749982198079,
      "epoch": 3.00427122263748,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00044971585869924826,
      "loss": 6.124,
      "mean_token_accuracy": 0.14834013250139025,
      "num_tokens": 6162282.0,
      "step": 2815
    },
    {
      "entropy": 6.658770990371704,
      "epoch": 3.00961025093433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004494491684864316,
      "loss": 6.1142,
      "mean_token_accuracy": 0.13844860345125198,
      "num_tokens": 6173818.0,
      "step": 2820
    },
    {
      "entropy": 6.700858926773071,
      "epoch": 3.01494927923118,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00044918186239379363,
      "loss": 6.145,
      "mean_token_accuracy": 0.1486111745238304,
      "num_tokens": 6184406.0,
      "step": 2825
    },
    {
      "entropy": 6.702039670944214,
      "epoch": 3.02028830752803,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004489139413650385,
      "loss": 6.1982,
      "mean_token_accuracy": 0.13922229036688805,
      "num_tokens": 6195817.0,
      "step": 2830
    },
    {
      "entropy": 6.690669584274292,
      "epoch": 3.02562733582488,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004486454063460415,
      "loss": 6.0552,
      "mean_token_accuracy": 0.14872528836131096,
      "num_tokens": 6206176.0,
      "step": 2835
    },
    {
      "entropy": 6.633033227920532,
      "epoch": 3.03096636412173,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004483762582848457,
      "loss": 6.1384,
      "mean_token_accuracy": 0.14411605522036552,
      "num_tokens": 6215783.0,
      "step": 2840
    },
    {
      "entropy": 6.706024503707885,
      "epoch": 3.03630539241858,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004481064981316583,
      "loss": 6.1254,
      "mean_token_accuracy": 0.13902800604701043,
      "num_tokens": 6227146.0,
      "step": 2845
    },
    {
      "entropy": 6.64505181312561,
      "epoch": 3.04164442071543,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00044783612683884744,
      "loss": 6.1002,
      "mean_token_accuracy": 0.14859205186367036,
      "num_tokens": 6237417.0,
      "step": 2850
    },
    {
      "entropy": 6.639095830917358,
      "epoch": 3.04698344901228,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004475651453609388,
      "loss": 6.1527,
      "mean_token_accuracy": 0.14703191369771956,
      "num_tokens": 6249136.0,
      "step": 2855
    },
    {
      "entropy": 6.657324695587159,
      "epoch": 3.0523224773091298,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00044729355465461236,
      "loss": 6.1194,
      "mean_token_accuracy": 0.1403007425367832,
      "num_tokens": 6260811.0,
      "step": 2860
    },
    {
      "entropy": 6.6688212871551515,
      "epoch": 3.0576615056059797,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000447021355678699,
      "loss": 6.1916,
      "mean_token_accuracy": 0.13820797055959702,
      "num_tokens": 6272776.0,
      "step": 2865
    },
    {
      "entropy": 6.72321252822876,
      "epoch": 3.0630005339028297,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004467485493941769,
      "loss": 6.1175,
      "mean_token_accuracy": 0.1424892447888851,
      "num_tokens": 6283335.0,
      "step": 2870
    },
    {
      "entropy": 6.535885858535766,
      "epoch": 3.0683395621996796,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044647513676416835,
      "loss": 6.0816,
      "mean_token_accuracy": 0.14990985095500947,
      "num_tokens": 6293383.0,
      "step": 2875
    },
    {
      "entropy": 6.673948097229004,
      "epoch": 3.0736785904965296,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004462011187539364,
      "loss": 6.1289,
      "mean_token_accuracy": 0.1540223926305771,
      "num_tokens": 6303373.0,
      "step": 2880
    },
    {
      "entropy": 6.657982444763183,
      "epoch": 3.0790176187933795,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004459264963308812,
      "loss": 6.1452,
      "mean_token_accuracy": 0.1427639067173004,
      "num_tokens": 6315674.0,
      "step": 2885
    },
    {
      "entropy": 6.5796790599823,
      "epoch": 3.0843566470902295,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044565127046453687,
      "loss": 6.0308,
      "mean_token_accuracy": 0.14728442281484605,
      "num_tokens": 6326183.0,
      "step": 2890
    },
    {
      "entropy": 6.510620546340943,
      "epoch": 3.0896956753870795,
      "grad_norm": 0.90625,
      "learning_rate": 0.00044537544212656775,
      "loss": 6.1323,
      "mean_token_accuracy": 0.14221204370260238,
      "num_tokens": 6337495.0,
      "step": 2895
    },
    {
      "entropy": 6.6099615573883055,
      "epoch": 3.0950347036839294,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00044509901229076537,
      "loss": 6.0958,
      "mean_token_accuracy": 0.14601162523031236,
      "num_tokens": 6348935.0,
      "step": 2900
    },
    {
      "entropy": 6.617335224151612,
      "epoch": 3.1003737319807794,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004448219819330447,
      "loss": 6.2415,
      "mean_token_accuracy": 0.13788999915122985,
      "num_tokens": 6359701.0,
      "step": 2905
    },
    {
      "entropy": 6.688894605636596,
      "epoch": 3.1057127602776293,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004445443520314409,
      "loss": 6.1635,
      "mean_token_accuracy": 0.1362752579152584,
      "num_tokens": 6370134.0,
      "step": 2910
    },
    {
      "entropy": 6.596358346939087,
      "epoch": 3.1110517885744793,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00044426612356610555,
      "loss": 6.1984,
      "mean_token_accuracy": 0.1391242265701294,
      "num_tokens": 6381644.0,
      "step": 2915
    },
    {
      "entropy": 6.677271556854248,
      "epoch": 3.1163908168713292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004439872975193037,
      "loss": 6.1728,
      "mean_token_accuracy": 0.1386138141155243,
      "num_tokens": 6393386.0,
      "step": 2920
    },
    {
      "entropy": 6.570891571044922,
      "epoch": 3.121729845168179,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00044370787487540986,
      "loss": 6.1742,
      "mean_token_accuracy": 0.14069369807839394,
      "num_tokens": 6405487.0,
      "step": 2925
    },
    {
      "entropy": 6.6059774398803714,
      "epoch": 3.1270688734650296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000443427856620905,
      "loss": 6.1768,
      "mean_token_accuracy": 0.14093007072806357,
      "num_tokens": 6416470.0,
      "step": 2930
    },
    {
      "entropy": 6.669240856170655,
      "epoch": 3.132407901761879,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044314724374437274,
      "loss": 6.2054,
      "mean_token_accuracy": 0.14186814278364182,
      "num_tokens": 6428579.0,
      "step": 2935
    },
    {
      "entropy": 6.625475597381592,
      "epoch": 3.1377469300587295,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004428660372364959,
      "loss": 6.2372,
      "mean_token_accuracy": 0.14359409660100936,
      "num_tokens": 6438976.0,
      "step": 2940
    },
    {
      "entropy": 6.640156650543213,
      "epoch": 3.1430859583555795,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004425842380900532,
      "loss": 6.1401,
      "mean_token_accuracy": 0.13635687604546548,
      "num_tokens": 6449919.0,
      "step": 2945
    },
    {
      "entropy": 6.540213251113892,
      "epoch": 3.1484249866524294,
      "grad_norm": 0.9375,
      "learning_rate": 0.00044230184729991565,
      "loss": 6.1379,
      "mean_token_accuracy": 0.14237724021077156,
      "num_tokens": 6460874.0,
      "step": 2950
    },
    {
      "entropy": 6.662953329086304,
      "epoch": 3.1537640149492794,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004420188658630428,
      "loss": 6.1318,
      "mean_token_accuracy": 0.14336148053407669,
      "num_tokens": 6471606.0,
      "step": 2955
    },
    {
      "entropy": 6.579173994064331,
      "epoch": 3.1591030432461293,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044173529477847974,
      "loss": 6.1682,
      "mean_token_accuracy": 0.140760400891304,
      "num_tokens": 6481917.0,
      "step": 2960
    },
    {
      "entropy": 6.60531063079834,
      "epoch": 3.1644420715429793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044145113504735295,
      "loss": 6.152,
      "mean_token_accuracy": 0.14812284484505653,
      "num_tokens": 6493295.0,
      "step": 2965
    },
    {
      "entropy": 6.603976917266846,
      "epoch": 3.1697810998398293,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004411663876728674,
      "loss": 6.1612,
      "mean_token_accuracy": 0.1435246966779232,
      "num_tokens": 6504208.0,
      "step": 2970
    },
    {
      "entropy": 6.5477807998657225,
      "epoch": 3.175120128136679,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00044088105366030234,
      "loss": 6.0951,
      "mean_token_accuracy": 0.1487772673368454,
      "num_tokens": 6515347.0,
      "step": 2975
    },
    {
      "entropy": 6.601194953918457,
      "epoch": 3.180459156433529,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004405951340170084,
      "loss": 6.1623,
      "mean_token_accuracy": 0.13532078862190247,
      "num_tokens": 6527674.0,
      "step": 2980
    },
    {
      "entropy": 6.564567041397095,
      "epoch": 3.185798184730379,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00044030862975240364,
      "loss": 6.1279,
      "mean_token_accuracy": 0.1448235183954239,
      "num_tokens": 6539048.0,
      "step": 2985
    },
    {
      "entropy": 6.595535612106323,
      "epoch": 3.191137213027229,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044002154187797003,
      "loss": 6.1979,
      "mean_token_accuracy": 0.1410246193408966,
      "num_tokens": 6550152.0,
      "step": 2990
    },
    {
      "entropy": 6.677579689025879,
      "epoch": 3.196476241324079,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043973387140725,
      "loss": 6.1773,
      "mean_token_accuracy": 0.14386973679065704,
      "num_tokens": 6561189.0,
      "step": 2995
    },
    {
      "entropy": 6.722335386276245,
      "epoch": 3.201815269620929,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004394456193558428,
      "loss": 6.2154,
      "mean_token_accuracy": 0.13892951905727385,
      "num_tokens": 6572548.0,
      "step": 3000
    },
    {
      "epoch": 3.201815269620929,
      "eval_entropy": 6.4393722497023544,
      "eval_loss": 6.7406086921691895,
      "eval_mean_token_accuracy": 0.12297290774521889,
      "eval_num_tokens": 6572548.0,
      "eval_runtime": 1.8114,
      "eval_samples_per_second": 1700.306,
      "eval_steps_per_second": 212.538,
      "step": 3000
    },
    {
      "entropy": 6.626015567779541,
      "epoch": 3.207154297917779,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004391567867414007,
      "loss": 6.2295,
      "mean_token_accuracy": 0.13846756219863893,
      "num_tokens": 6584176.0,
      "step": 3005
    },
    {
      "entropy": 6.582659435272217,
      "epoch": 3.212493326214629,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043886737458362596,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15341638177633285,
      "num_tokens": 6595070.0,
      "step": 3010
    },
    {
      "entropy": 6.523849010467529,
      "epoch": 3.217832354511479,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004385773839042666,
      "loss": 6.1148,
      "mean_token_accuracy": 0.14375541731715202,
      "num_tokens": 6606556.0,
      "step": 3015
    },
    {
      "entropy": 6.617887306213379,
      "epoch": 3.223171382808329,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043828681572711316,
      "loss": 6.0635,
      "mean_token_accuracy": 0.14926337003707885,
      "num_tokens": 6616944.0,
      "step": 3020
    },
    {
      "entropy": 6.489473152160644,
      "epoch": 3.228510411105179,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043799567107799504,
      "loss": 6.1453,
      "mean_token_accuracy": 0.15048904716968536,
      "num_tokens": 6627541.0,
      "step": 3025
    },
    {
      "entropy": 6.605879449844361,
      "epoch": 3.2338494394020287,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00043770395098477666,
      "loss": 6.1314,
      "mean_token_accuracy": 0.14454084560275077,
      "num_tokens": 6639102.0,
      "step": 3030
    },
    {
      "entropy": 6.652470588684082,
      "epoch": 3.2391884676988787,
      "grad_norm": 1.0,
      "learning_rate": 0.0004374116564773543,
      "loss": 6.1631,
      "mean_token_accuracy": 0.14270546436309814,
      "num_tokens": 6650737.0,
      "step": 3035
    },
    {
      "entropy": 6.537238025665284,
      "epoch": 3.2445274959957286,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004371187885876519,
      "loss": 6.1842,
      "mean_token_accuracy": 0.138457190990448,
      "num_tokens": 6661812.0,
      "step": 3040
    },
    {
      "entropy": 6.6336798667907715,
      "epoch": 3.2498665242925786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043682534834961766,
      "loss": 6.2576,
      "mean_token_accuracy": 0.1425720900297165,
      "num_tokens": 6672122.0,
      "step": 3045
    },
    {
      "entropy": 6.620451641082764,
      "epoch": 3.2552055525894286,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043653133679922077,
      "loss": 6.2141,
      "mean_token_accuracy": 0.14090549126267432,
      "num_tokens": 6684219.0,
      "step": 3050
    },
    {
      "entropy": 6.561987209320068,
      "epoch": 3.2605445808862785,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043623675497444695,
      "loss": 6.0454,
      "mean_token_accuracy": 0.14704931527376175,
      "num_tokens": 6694290.0,
      "step": 3055
    },
    {
      "entropy": 6.5945192813873295,
      "epoch": 3.2658836091831285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043594160391529545,
      "loss": 6.157,
      "mean_token_accuracy": 0.14867981225252153,
      "num_tokens": 6704313.0,
      "step": 3060
    },
    {
      "entropy": 6.563622903823853,
      "epoch": 3.271222637479979,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043564588466377536,
      "loss": 6.1617,
      "mean_token_accuracy": 0.14300615936517716,
      "num_tokens": 6715424.0,
      "step": 3065
    },
    {
      "entropy": 6.504296970367432,
      "epoch": 3.2765616657768284,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004353495982639012,
      "loss": 6.0365,
      "mean_token_accuracy": 0.144931760430336,
      "num_tokens": 6727025.0,
      "step": 3070
    },
    {
      "entropy": 6.58922119140625,
      "epoch": 3.281900694073679,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043505274576169037,
      "loss": 6.1521,
      "mean_token_accuracy": 0.14766536206007003,
      "num_tokens": 6738344.0,
      "step": 3075
    },
    {
      "entropy": 6.5671083450317385,
      "epoch": 3.2872397223705287,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00043475532820515836,
      "loss": 6.1378,
      "mean_token_accuracy": 0.14321689307689667,
      "num_tokens": 6748812.0,
      "step": 3080
    },
    {
      "entropy": 6.579545545578003,
      "epoch": 3.2925787506673787,
      "grad_norm": 0.9375,
      "learning_rate": 0.00043445734664431575,
      "loss": 6.1458,
      "mean_token_accuracy": 0.14259575307369232,
      "num_tokens": 6759776.0,
      "step": 3085
    },
    {
      "entropy": 6.557349061965942,
      "epoch": 3.2979177789642287,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004341588021311644,
      "loss": 6.1432,
      "mean_token_accuracy": 0.14480729252099991,
      "num_tokens": 6770702.0,
      "step": 3090
    },
    {
      "entropy": 6.5284239768981935,
      "epoch": 3.3032568072610786,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00043385969571969353,
      "loss": 6.14,
      "mean_token_accuracy": 0.1491951197385788,
      "num_tokens": 6781599.0,
      "step": 3095
    },
    {
      "entropy": 6.5264880657196045,
      "epoch": 3.3085958355579286,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00043356002846587597,
      "loss": 6.1786,
      "mean_token_accuracy": 0.14892054349184036,
      "num_tokens": 6792385.0,
      "step": 3100
    },
    {
      "entropy": 6.645450162887573,
      "epoch": 3.3139348638547785,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004332598014276647,
      "loss": 6.1443,
      "mean_token_accuracy": 0.1407065249979496,
      "num_tokens": 6803805.0,
      "step": 3105
    },
    {
      "entropy": 6.503142929077148,
      "epoch": 3.3192738921516285,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00043295901566498907,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1454508811235428,
      "num_tokens": 6815460.0,
      "step": 3110
    },
    {
      "entropy": 6.593047142028809,
      "epoch": 3.3246129204484784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043265767223975076,
      "loss": 6.1151,
      "mean_token_accuracy": 0.1487767554819584,
      "num_tokens": 6825995.0,
      "step": 3115
    },
    {
      "entropy": 6.522263669967652,
      "epoch": 3.3299519487453284,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004323557722158204,
      "loss": 6.0856,
      "mean_token_accuracy": 0.14423727318644525,
      "num_tokens": 6836793.0,
      "step": 3120
    },
    {
      "entropy": 6.56120867729187,
      "epoch": 3.3352909770421784,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004320533166590335,
      "loss": 6.1374,
      "mean_token_accuracy": 0.14478903859853745,
      "num_tokens": 6849063.0,
      "step": 3125
    },
    {
      "entropy": 6.615983200073242,
      "epoch": 3.3406300053390283,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000431750306637187,
      "loss": 6.1393,
      "mean_token_accuracy": 0.13982861638069152,
      "num_tokens": 6859821.0,
      "step": 3130
    },
    {
      "entropy": 6.547687005996704,
      "epoch": 3.3459690336358783,
      "grad_norm": 0.953125,
      "learning_rate": 0.00043144674322003525,
      "loss": 6.1659,
      "mean_token_accuracy": 0.14122217744588852,
      "num_tokens": 6871026.0,
      "step": 3135
    },
    {
      "entropy": 6.6049974918365475,
      "epoch": 3.3513080619327282,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004311426274792863,
      "loss": 6.2092,
      "mean_token_accuracy": 0.14053697511553764,
      "num_tokens": 6883135.0,
      "step": 3140
    },
    {
      "entropy": 6.531812572479248,
      "epoch": 3.356647090229578,
      "grad_norm": 0.875,
      "learning_rate": 0.0004308379604885981,
      "loss": 6.1506,
      "mean_token_accuracy": 0.14196842312812805,
      "num_tokens": 6893943.0,
      "step": 3145
    },
    {
      "entropy": 6.480563163757324,
      "epoch": 3.361986118526428,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004305327433235749,
      "loss": 6.17,
      "mean_token_accuracy": 0.14052093848586084,
      "num_tokens": 6904172.0,
      "step": 3150
    },
    {
      "entropy": 6.660596752166748,
      "epoch": 3.367325146823278,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004302269770617632,
      "loss": 6.174,
      "mean_token_accuracy": 0.14555246010422707,
      "num_tokens": 6915440.0,
      "step": 3155
    },
    {
      "entropy": 6.630261611938477,
      "epoch": 3.372664175120128,
      "grad_norm": 0.984375,
      "learning_rate": 0.000429920662782648,
      "loss": 6.2339,
      "mean_token_accuracy": 0.1399409957230091,
      "num_tokens": 6927762.0,
      "step": 3160
    },
    {
      "entropy": 6.615788125991822,
      "epoch": 3.378003203416978,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042961380156764925,
      "loss": 6.1738,
      "mean_token_accuracy": 0.1380746841430664,
      "num_tokens": 6938139.0,
      "step": 3165
    },
    {
      "entropy": 6.513877248764038,
      "epoch": 3.383342231713828,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004293063945001175,
      "loss": 6.1746,
      "mean_token_accuracy": 0.14551160410046576,
      "num_tokens": 6949463.0,
      "step": 3170
    },
    {
      "entropy": 6.643848180770874,
      "epoch": 3.388681260010678,
      "grad_norm": 0.921875,
      "learning_rate": 0.00042899844266533054,
      "loss": 6.2151,
      "mean_token_accuracy": 0.14019910618662834,
      "num_tokens": 6960660.0,
      "step": 3175
    },
    {
      "entropy": 6.584480094909668,
      "epoch": 3.394020288307528,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004286899471504896,
      "loss": 6.1518,
      "mean_token_accuracy": 0.14778395593166352,
      "num_tokens": 6971283.0,
      "step": 3180
    },
    {
      "entropy": 6.641484785079956,
      "epoch": 3.399359316604378,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042838090904471495,
      "loss": 6.1865,
      "mean_token_accuracy": 0.14859624207019806,
      "num_tokens": 6981248.0,
      "step": 3185
    },
    {
      "entropy": 6.482734870910645,
      "epoch": 3.404698344901228,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004280713294390428,
      "loss": 6.153,
      "mean_token_accuracy": 0.1452343285083771,
      "num_tokens": 6991574.0,
      "step": 3190
    },
    {
      "entropy": 6.56207766532898,
      "epoch": 3.4100373731980778,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004277612094264209,
      "loss": 6.1664,
      "mean_token_accuracy": 0.1509738191962242,
      "num_tokens": 7002548.0,
      "step": 3195
    },
    {
      "entropy": 6.571753215789795,
      "epoch": 3.415376401494928,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004274505501017048,
      "loss": 6.1943,
      "mean_token_accuracy": 0.14039533361792564,
      "num_tokens": 7013815.0,
      "step": 3200
    },
    {
      "entropy": 6.619806480407715,
      "epoch": 3.4207154297917777,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004271393525616544,
      "loss": 6.1581,
      "mean_token_accuracy": 0.145014002174139,
      "num_tokens": 7024212.0,
      "step": 3205
    },
    {
      "entropy": 6.439695072174072,
      "epoch": 3.426054458088628,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004268276179049292,
      "loss": 6.1545,
      "mean_token_accuracy": 0.14754147827625275,
      "num_tokens": 7035167.0,
      "step": 3210
    },
    {
      "entropy": 6.619048595428467,
      "epoch": 3.431393486385478,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265153472320854,
      "loss": 6.1243,
      "mean_token_accuracy": 0.14939205199480057,
      "num_tokens": 7045036.0,
      "step": 3215
    },
    {
      "entropy": 6.576473379135132,
      "epoch": 3.436732514682328,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042620254164557137,
      "loss": 6.1894,
      "mean_token_accuracy": 0.14514195621013642,
      "num_tokens": 7054938.0,
      "step": 3220
    },
    {
      "entropy": 6.575670194625855,
      "epoch": 3.442071542979178,
      "grad_norm": 1.078125,
      "learning_rate": 0.000425889202249724,
      "loss": 6.1979,
      "mean_token_accuracy": 0.14288137555122377,
      "num_tokens": 7065645.0,
      "step": 3225
    },
    {
      "entropy": 6.573948764801026,
      "epoch": 3.447410571276028,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00042557533015076456,
      "loss": 6.1069,
      "mean_token_accuracy": 0.14459412023425103,
      "num_tokens": 7075909.0,
      "step": 3230
    },
    {
      "entropy": 6.556999588012696,
      "epoch": 3.452749599572878,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042526092645679535,
      "loss": 6.1364,
      "mean_token_accuracy": 0.15125184655189514,
      "num_tokens": 7085761.0,
      "step": 3235
    },
    {
      "entropy": 6.537608480453491,
      "epoch": 3.458088627869728,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042494599227779523,
      "loss": 6.2128,
      "mean_token_accuracy": 0.1416289672255516,
      "num_tokens": 7096357.0,
      "step": 3240
    },
    {
      "entropy": 6.5380041122436525,
      "epoch": 3.4634276561665778,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042463052872561587,
      "loss": 6.2078,
      "mean_token_accuracy": 0.14668214172124863,
      "num_tokens": 7107534.0,
      "step": 3245
    },
    {
      "entropy": 6.580185413360596,
      "epoch": 3.4687666844634277,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042431453691397787,
      "loss": 6.2045,
      "mean_token_accuracy": 0.13919102773070335,
      "num_tokens": 7119186.0,
      "step": 3250
    },
    {
      "entropy": 6.587459516525269,
      "epoch": 3.4741057127602777,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004239980179584669,
      "loss": 6.0809,
      "mean_token_accuracy": 0.1546573482453823,
      "num_tokens": 7130187.0,
      "step": 3255
    },
    {
      "entropy": 6.536509609222412,
      "epoch": 3.4794447410571276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004236809729765295,
      "loss": 6.1595,
      "mean_token_accuracy": 0.14584122598171234,
      "num_tokens": 7140734.0,
      "step": 3260
    },
    {
      "entropy": 6.479880380630493,
      "epoch": 3.4847837693539776,
      "grad_norm": 0.96875,
      "learning_rate": 0.00042336340308746944,
      "loss": 6.1511,
      "mean_token_accuracy": 0.1437065228819847,
      "num_tokens": 7151642.0,
      "step": 3265
    },
    {
      "entropy": 6.549475431442261,
      "epoch": 3.4901227976508276,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004230453094124436,
      "loss": 6.1861,
      "mean_token_accuracy": 0.13861185908317566,
      "num_tokens": 7162650.0,
      "step": 3270
    },
    {
      "entropy": 6.549925708770752,
      "epoch": 3.4954618259476775,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00042272669307445823,
      "loss": 6.161,
      "mean_token_accuracy": 0.14364441037178038,
      "num_tokens": 7174095.0,
      "step": 3275
    },
    {
      "entropy": 6.569450330734253,
      "epoch": 3.5008008542445275,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00042240755519836436,
      "loss": 6.1446,
      "mean_token_accuracy": 0.15000987201929092,
      "num_tokens": 7185923.0,
      "step": 3280
    },
    {
      "entropy": 6.502327013015747,
      "epoch": 3.5061398825413774,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00042208789691085464,
      "loss": 6.1405,
      "mean_token_accuracy": 0.14843530356884002,
      "num_tokens": 7197287.0,
      "step": 3285
    },
    {
      "entropy": 6.547327613830566,
      "epoch": 3.5114789108382274,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000421767719340459,
      "loss": 6.1417,
      "mean_token_accuracy": 0.14269062951207162,
      "num_tokens": 7208143.0,
      "step": 3290
    },
    {
      "entropy": 6.547702884674072,
      "epoch": 3.5168179391350773,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004214470236175404,
      "loss": 6.1023,
      "mean_token_accuracy": 0.15332724750041962,
      "num_tokens": 7218516.0,
      "step": 3295
    },
    {
      "entropy": 6.547658395767212,
      "epoch": 3.5221569674319273,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042112581087429133,
      "loss": 6.1804,
      "mean_token_accuracy": 0.14948860853910445,
      "num_tokens": 7229066.0,
      "step": 3300
    },
    {
      "entropy": 6.561883544921875,
      "epoch": 3.5274959957287773,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042080408224472946,
      "loss": 6.1691,
      "mean_token_accuracy": 0.14923621192574502,
      "num_tokens": 7239612.0,
      "step": 3305
    },
    {
      "entropy": 6.505513095855713,
      "epoch": 3.532835024025627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042048183886469386,
      "loss": 6.08,
      "mean_token_accuracy": 0.14673556685447692,
      "num_tokens": 7250266.0,
      "step": 3310
    },
    {
      "entropy": 6.513124561309814,
      "epoch": 3.538174052322477,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042015908187184077,
      "loss": 6.1603,
      "mean_token_accuracy": 0.14263225272297858,
      "num_tokens": 7261183.0,
      "step": 3315
    },
    {
      "entropy": 6.538923549652099,
      "epoch": 3.543513080619327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041983581240563985,
      "loss": 6.1348,
      "mean_token_accuracy": 0.1486171156167984,
      "num_tokens": 7271557.0,
      "step": 3320
    },
    {
      "entropy": 6.535774898529053,
      "epoch": 3.5488521089161775,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004195120316073698,
      "loss": 6.1998,
      "mean_token_accuracy": 0.1453086718916893,
      "num_tokens": 7282540.0,
      "step": 3325
    },
    {
      "entropy": 6.511842823028564,
      "epoch": 3.554191137213027,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041918774062011467,
      "loss": 6.1107,
      "mean_token_accuracy": 0.14593104943633078,
      "num_tokens": 7293568.0,
      "step": 3330
    },
    {
      "entropy": 6.523481845855713,
      "epoch": 3.5595301655098774,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041886294058875975,
      "loss": 6.2032,
      "mean_token_accuracy": 0.1386525124311447,
      "num_tokens": 7304745.0,
      "step": 3335
    },
    {
      "entropy": 6.553685283660888,
      "epoch": 3.564869193806727,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004185376326599874,
      "loss": 6.2036,
      "mean_token_accuracy": 0.14021066576242447,
      "num_tokens": 7316547.0,
      "step": 3340
    },
    {
      "entropy": 6.578783273696899,
      "epoch": 3.5702082221035774,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041821181798227305,
      "loss": 6.2193,
      "mean_token_accuracy": 0.14152044877409936,
      "num_tokens": 7328493.0,
      "step": 3345
    },
    {
      "entropy": 6.494209146499633,
      "epoch": 3.575547250400427,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004178854977058812,
      "loss": 6.1514,
      "mean_token_accuracy": 0.14675125181674958,
      "num_tokens": 7338831.0,
      "step": 3350
    },
    {
      "entropy": 6.58004937171936,
      "epoch": 3.5808862786972773,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004175586729828614,
      "loss": 6.2568,
      "mean_token_accuracy": 0.13772836774587632,
      "num_tokens": 7350884.0,
      "step": 3355
    },
    {
      "entropy": 6.56378321647644,
      "epoch": 3.5862253069941272,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004172313449670439,
      "loss": 6.1245,
      "mean_token_accuracy": 0.14097074419260025,
      "num_tokens": 7360674.0,
      "step": 3360
    },
    {
      "entropy": 6.479941940307617,
      "epoch": 3.591564335290977,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00041690351481403606,
      "loss": 6.1567,
      "mean_token_accuracy": 0.1383296102285385,
      "num_tokens": 7371965.0,
      "step": 3365
    },
    {
      "entropy": 6.642097520828247,
      "epoch": 3.596903363587827,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004165751836812178,
      "loss": 6.241,
      "mean_token_accuracy": 0.145929317176342,
      "num_tokens": 7383649.0,
      "step": 3370
    },
    {
      "entropy": 6.6187879085540775,
      "epoch": 3.602242391884677,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004162463527277378,
      "loss": 6.1849,
      "mean_token_accuracy": 0.1447222724556923,
      "num_tokens": 7393692.0,
      "step": 3375
    },
    {
      "entropy": 6.562582874298096,
      "epoch": 3.607581420181527,
      "grad_norm": 0.953125,
      "learning_rate": 0.00041591702311450924,
      "loss": 6.2116,
      "mean_token_accuracy": 0.13732682764530182,
      "num_tokens": 7404149.0,
      "step": 3380
    },
    {
      "entropy": 6.533840751647949,
      "epoch": 3.612920448478377,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004155871960042059,
      "loss": 6.1192,
      "mean_token_accuracy": 0.1485688403248787,
      "num_tokens": 7415011.0,
      "step": 3385
    },
    {
      "entropy": 6.53869137763977,
      "epoch": 3.618259476775227,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00041525687256125786,
      "loss": 6.1796,
      "mean_token_accuracy": 0.14087817072868347,
      "num_tokens": 7426297.0,
      "step": 3390
    },
    {
      "entropy": 6.520381307601928,
      "epoch": 3.623598505072077,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004149260539518475,
      "loss": 6.054,
      "mean_token_accuracy": 0.1445208042860031,
      "num_tokens": 7437718.0,
      "step": 3395
    },
    {
      "entropy": 6.488810968399048,
      "epoch": 3.628937533368927,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041459474134390527,
      "loss": 6.1478,
      "mean_token_accuracy": 0.1479748159646988,
      "num_tokens": 7448395.0,
      "step": 3400
    },
    {
      "entropy": 6.43403959274292,
      "epoch": 3.634276561665777,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00041426293590710587,
      "loss": 6.1326,
      "mean_token_accuracy": 0.15362447500228882,
      "num_tokens": 7458374.0,
      "step": 3405
    },
    {
      "entropy": 6.588842344284058,
      "epoch": 3.639615589962627,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00041393063881286346,
      "loss": 6.1325,
      "mean_token_accuracy": 0.15165682807564734,
      "num_tokens": 7469386.0,
      "step": 3410
    },
    {
      "entropy": 6.571090030670166,
      "epoch": 3.6449546182594768,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004135978512343284,
      "loss": 6.1911,
      "mean_token_accuracy": 0.1359236106276512,
      "num_tokens": 7479788.0,
      "step": 3415
    },
    {
      "entropy": 6.470835828781128,
      "epoch": 3.6502936465563267,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041326457434638246,
      "loss": 6.1273,
      "mean_token_accuracy": 0.14886193200945855,
      "num_tokens": 7490329.0,
      "step": 3420
    },
    {
      "entropy": 6.532852983474731,
      "epoch": 3.6556326748531767,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000412930809325635,
      "loss": 6.175,
      "mean_token_accuracy": 0.139870285987854,
      "num_tokens": 7502206.0,
      "step": 3425
    },
    {
      "entropy": 6.530126094818115,
      "epoch": 3.6609717031500266,
      "grad_norm": 0.921875,
      "learning_rate": 0.00041259655735041847,
      "loss": 6.1883,
      "mean_token_accuracy": 0.14210368692874908,
      "num_tokens": 7513704.0,
      "step": 3430
    },
    {
      "entropy": 6.4876954555511475,
      "epoch": 3.6663107314468766,
      "grad_norm": 0.96875,
      "learning_rate": 0.00041226181960078473,
      "loss": 6.0956,
      "mean_token_accuracy": 0.1456264317035675,
      "num_tokens": 7524320.0,
      "step": 3435
    },
    {
      "entropy": 6.542640733718872,
      "epoch": 3.6716497597437265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004119265972585005,
      "loss": 6.1752,
      "mean_token_accuracy": 0.14658585488796233,
      "num_tokens": 7533962.0,
      "step": 3440
    },
    {
      "entropy": 6.579928541183472,
      "epoch": 3.6769887880405765,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041159089150704325,
      "loss": 6.2128,
      "mean_token_accuracy": 0.14553995281457902,
      "num_tokens": 7545841.0,
      "step": 3445
    },
    {
      "entropy": 6.596597528457641,
      "epoch": 3.6823278163374265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041125470353159735,
      "loss": 6.1607,
      "mean_token_accuracy": 0.14802061766386032,
      "num_tokens": 7555533.0,
      "step": 3450
    },
    {
      "entropy": 6.503862905502319,
      "epoch": 3.6876668446342764,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00041091803451904927,
      "loss": 6.1554,
      "mean_token_accuracy": 0.1387634038925171,
      "num_tokens": 7566228.0,
      "step": 3455
    },
    {
      "entropy": 6.505353736877441,
      "epoch": 3.693005872931127,
      "grad_norm": 0.90625,
      "learning_rate": 0.00041058088565798403,
      "loss": 6.198,
      "mean_token_accuracy": 0.14261291325092315,
      "num_tokens": 7577694.0,
      "step": 3460
    },
    {
      "entropy": 6.62959246635437,
      "epoch": 3.6983449012279763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041024325813868065,
      "loss": 6.1753,
      "mean_token_accuracy": 0.14205058813095092,
      "num_tokens": 7588141.0,
      "step": 3465
    },
    {
      "entropy": 6.598307275772095,
      "epoch": 3.7036839295248267,
      "grad_norm": 0.96875,
      "learning_rate": 0.00040990515315310794,
      "loss": 6.1758,
      "mean_token_accuracy": 0.14348702132701874,
      "num_tokens": 7599831.0,
      "step": 3470
    },
    {
      "entropy": 6.537150287628174,
      "epoch": 3.7090229578216762,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040956657189492046,
      "loss": 6.0966,
      "mean_token_accuracy": 0.14398248717188836,
      "num_tokens": 7609567.0,
      "step": 3475
    },
    {
      "entropy": 6.556712341308594,
      "epoch": 3.7143619861185266,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004092275155594541,
      "loss": 6.2319,
      "mean_token_accuracy": 0.13728955760598183,
      "num_tokens": 7620722.0,
      "step": 3480
    },
    {
      "entropy": 6.567393589019775,
      "epoch": 3.719701014415376,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004088879853437221,
      "loss": 6.25,
      "mean_token_accuracy": 0.13819843381643296,
      "num_tokens": 7633036.0,
      "step": 3485
    },
    {
      "entropy": 6.479170513153076,
      "epoch": 3.7250400427122266,
      "grad_norm": 1.0,
      "learning_rate": 0.00040854798244641067,
      "loss": 6.1627,
      "mean_token_accuracy": 0.14585598707199096,
      "num_tokens": 7642789.0,
      "step": 3490
    },
    {
      "entropy": 6.506835269927978,
      "epoch": 3.7303790710090765,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004082075080678748,
      "loss": 6.1908,
      "mean_token_accuracy": 0.14425090700387955,
      "num_tokens": 7654234.0,
      "step": 3495
    },
    {
      "entropy": 6.539835405349732,
      "epoch": 3.7357180993059265,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040786656341013376,
      "loss": 6.1273,
      "mean_token_accuracy": 0.14244442880153657,
      "num_tokens": 7665994.0,
      "step": 3500
    },
    {
      "epoch": 3.7357180993059265,
      "eval_entropy": 6.402802858724223,
      "eval_loss": 6.652650833129883,
      "eval_mean_token_accuracy": 0.12644047731315936,
      "eval_num_tokens": 7665994.0,
      "eval_runtime": 1.8024,
      "eval_samples_per_second": 1708.804,
      "eval_steps_per_second": 213.6,
      "step": 3500
    },
    {
      "entropy": 6.511250829696655,
      "epoch": 3.7410571276027764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040752514967686747,
      "loss": 6.084,
      "mean_token_accuracy": 0.14790418446063996,
      "num_tokens": 7676813.0,
      "step": 3505
    },
    {
      "entropy": 6.438376808166504,
      "epoch": 3.7463961558996264,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004071832680734118,
      "loss": 6.1267,
      "mean_token_accuracy": 0.1470203071832657,
      "num_tokens": 7687775.0,
      "step": 3510
    },
    {
      "entropy": 6.558165264129639,
      "epoch": 3.7517351841964763,
      "grad_norm": 0.953125,
      "learning_rate": 0.00040684091980675426,
      "loss": 6.2229,
      "mean_token_accuracy": 0.1389820992946625,
      "num_tokens": 7698909.0,
      "step": 3515
    },
    {
      "entropy": 6.513540887832642,
      "epoch": 3.7570742124933263,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040649810608552994,
      "loss": 6.132,
      "mean_token_accuracy": 0.1510733187198639,
      "num_tokens": 7708783.0,
      "step": 3520
    },
    {
      "entropy": 6.544998741149902,
      "epoch": 3.7624132407901762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040615482812001726,
      "loss": 6.174,
      "mean_token_accuracy": 0.14355052411556243,
      "num_tokens": 7719010.0,
      "step": 3525
    },
    {
      "entropy": 6.5300548553466795,
      "epoch": 3.767752269087026,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004058110871221335,
      "loss": 6.2019,
      "mean_token_accuracy": 0.1421898864209652,
      "num_tokens": 7730231.0,
      "step": 3530
    },
    {
      "entropy": 6.484360885620117,
      "epoch": 3.773091297383876,
      "grad_norm": 0.953125,
      "learning_rate": 0.00040546688430543074,
      "loss": 6.1209,
      "mean_token_accuracy": 0.14670493602752685,
      "num_tokens": 7741656.0,
      "step": 3535
    },
    {
      "entropy": 6.542676115036011,
      "epoch": 3.778430325680726,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00040512222088509147,
      "loss": 6.1385,
      "mean_token_accuracy": 0.1478129133582115,
      "num_tokens": 7751956.0,
      "step": 3540
    },
    {
      "entropy": 6.508578777313232,
      "epoch": 3.783769353977576,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004047770980779242,
      "loss": 6.1029,
      "mean_token_accuracy": 0.1496267482638359,
      "num_tokens": 7761944.0,
      "step": 3545
    },
    {
      "entropy": 6.446523857116699,
      "epoch": 3.789108382274426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004044315171023593,
      "loss": 6.0753,
      "mean_token_accuracy": 0.14807819575071335,
      "num_tokens": 7771370.0,
      "step": 3550
    },
    {
      "entropy": 6.563878965377808,
      "epoch": 3.794447410571276,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004040854791784448,
      "loss": 6.2957,
      "mean_token_accuracy": 0.13254996612668038,
      "num_tokens": 7783615.0,
      "step": 3555
    },
    {
      "entropy": 6.600985717773438,
      "epoch": 3.799786438868126,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004037389855278419,
      "loss": 6.1559,
      "mean_token_accuracy": 0.1492430627346039,
      "num_tokens": 7794236.0,
      "step": 3560
    },
    {
      "entropy": 6.468111991882324,
      "epoch": 3.805125467164976,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040339203737382065,
      "loss": 6.166,
      "mean_token_accuracy": 0.14735100716352462,
      "num_tokens": 7806158.0,
      "step": 3565
    },
    {
      "entropy": 6.527578115463257,
      "epoch": 3.810464495461826,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004030446359412556,
      "loss": 6.12,
      "mean_token_accuracy": 0.14553360044956207,
      "num_tokens": 7818283.0,
      "step": 3570
    },
    {
      "entropy": 6.579574012756348,
      "epoch": 3.815803523758676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004026967824566218,
      "loss": 6.1631,
      "mean_token_accuracy": 0.1417527861893177,
      "num_tokens": 7828683.0,
      "step": 3575
    },
    {
      "entropy": 6.49411449432373,
      "epoch": 3.8211425520555258,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004023484781479902,
      "loss": 6.1374,
      "mean_token_accuracy": 0.14596590846776963,
      "num_tokens": 7839116.0,
      "step": 3580
    },
    {
      "entropy": 6.544578886032104,
      "epoch": 3.8264815803523757,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004019997242450231,
      "loss": 6.1942,
      "mean_token_accuracy": 0.1430832676589489,
      "num_tokens": 7850300.0,
      "step": 3585
    },
    {
      "entropy": 6.535199403762817,
      "epoch": 3.8318206086492257,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004016505219789703,
      "loss": 6.096,
      "mean_token_accuracy": 0.145581466704607,
      "num_tokens": 7861432.0,
      "step": 3590
    },
    {
      "entropy": 6.481894445419312,
      "epoch": 3.837159636946076,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004013008725826644,
      "loss": 6.19,
      "mean_token_accuracy": 0.145276015996933,
      "num_tokens": 7872472.0,
      "step": 3595
    },
    {
      "entropy": 6.576520872116089,
      "epoch": 3.8424986652429256,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004009507772905167,
      "loss": 6.1948,
      "mean_token_accuracy": 0.146095909178257,
      "num_tokens": 7883554.0,
      "step": 3600
    },
    {
      "entropy": 6.5780809879302975,
      "epoch": 3.847837693539776,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00040060023733851254,
      "loss": 6.1698,
      "mean_token_accuracy": 0.14025285094976425,
      "num_tokens": 7894853.0,
      "step": 3605
    },
    {
      "entropy": 6.464986371994018,
      "epoch": 3.8531767218366255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040024925396420707,
      "loss": 6.1337,
      "mean_token_accuracy": 0.1404825381934643,
      "num_tokens": 7905459.0,
      "step": 3610
    },
    {
      "entropy": 6.486498022079468,
      "epoch": 3.858515750133476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039989782840672123,
      "loss": 6.0861,
      "mean_token_accuracy": 0.14591265097260475,
      "num_tokens": 7916646.0,
      "step": 3615
    },
    {
      "entropy": 6.47977614402771,
      "epoch": 3.8638547784303254,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003995459619067366,
      "loss": 6.1397,
      "mean_token_accuracy": 0.14808139503002166,
      "num_tokens": 7928119.0,
      "step": 3620
    },
    {
      "entropy": 6.526146650314331,
      "epoch": 3.869193806727176,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039919365570649187,
      "loss": 6.2565,
      "mean_token_accuracy": 0.13531762585043908,
      "num_tokens": 7939295.0,
      "step": 3625
    },
    {
      "entropy": 6.55569486618042,
      "epoch": 3.874532835024026,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00039884091104977774,
      "loss": 6.0711,
      "mean_token_accuracy": 0.14545215293765068,
      "num_tokens": 7950094.0,
      "step": 3630
    },
    {
      "entropy": 6.515868520736694,
      "epoch": 3.8798718633208757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039848772918193315,
      "loss": 6.0973,
      "mean_token_accuracy": 0.1533176302909851,
      "num_tokens": 7960767.0,
      "step": 3635
    },
    {
      "entropy": 6.44457221031189,
      "epoch": 3.8852108916177257,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003981341113498404,
      "loss": 6.1581,
      "mean_token_accuracy": 0.14428275674581528,
      "num_tokens": 7971660.0,
      "step": 3640
    },
    {
      "entropy": 6.5656482696533205,
      "epoch": 3.8905499199145757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003977800588019209,
      "loss": 6.158,
      "mean_token_accuracy": 0.1490553878247738,
      "num_tokens": 7983337.0,
      "step": 3645
    },
    {
      "entropy": 6.527936792373657,
      "epoch": 3.8958889482114256,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0003974255727881307,
      "loss": 6.1845,
      "mean_token_accuracy": 0.14226604625582695,
      "num_tokens": 7994540.0,
      "step": 3650
    },
    {
      "entropy": 6.515545797348023,
      "epoch": 3.9012279765082756,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003970706545599566,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1402305468916893,
      "num_tokens": 8005396.0,
      "step": 3655
    },
    {
      "entropy": 6.4922425746917725,
      "epoch": 3.9065670048051255,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039671530537041094,
      "loss": 6.1029,
      "mean_token_accuracy": 0.15106893181800843,
      "num_tokens": 8015411.0,
      "step": 3660
    },
    {
      "entropy": 6.487470006942749,
      "epoch": 3.9119060331019755,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039635952647402756,
      "loss": 6.0656,
      "mean_token_accuracy": 0.15119688287377359,
      "num_tokens": 8025720.0,
      "step": 3665
    },
    {
      "entropy": 6.427022790908813,
      "epoch": 3.9172450613988254,
      "grad_norm": 1.0,
      "learning_rate": 0.0003960033191268574,
      "loss": 6.1439,
      "mean_token_accuracy": 0.1427379496395588,
      "num_tokens": 8036081.0,
      "step": 3670
    },
    {
      "entropy": 6.473222827911377,
      "epoch": 3.9225840896956754,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003956466845864641,
      "loss": 6.0318,
      "mean_token_accuracy": 0.14493608772754668,
      "num_tokens": 8046714.0,
      "step": 3675
    },
    {
      "entropy": 6.415006065368653,
      "epoch": 3.9279231179925254,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0003952896241119194,
      "loss": 6.0935,
      "mean_token_accuracy": 0.15388759225606918,
      "num_tokens": 8057174.0,
      "step": 3680
    },
    {
      "entropy": 6.449224328994751,
      "epoch": 3.9332621462893753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003949321389637986,
      "loss": 6.0499,
      "mean_token_accuracy": 0.15181737542152404,
      "num_tokens": 8067697.0,
      "step": 3685
    },
    {
      "entropy": 6.493015480041504,
      "epoch": 3.9386011745862253,
      "grad_norm": 1.0,
      "learning_rate": 0.0003945742304041766,
      "loss": 6.1459,
      "mean_token_accuracy": 0.14583780020475387,
      "num_tokens": 8077800.0,
      "step": 3690
    },
    {
      "entropy": 6.556552982330322,
      "epoch": 3.9439402028830752,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0003942158996966229,
      "loss": 6.2258,
      "mean_token_accuracy": 0.1387764908373356,
      "num_tokens": 8090319.0,
      "step": 3695
    },
    {
      "entropy": 6.55411057472229,
      "epoch": 3.949279231179925,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003938571481061975,
      "loss": 6.1572,
      "mean_token_accuracy": 0.14131653159856797,
      "num_tokens": 8100498.0,
      "step": 3700
    },
    {
      "entropy": 6.481984329223633,
      "epoch": 3.954618259476775,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039349797689944613,
      "loss": 6.0891,
      "mean_token_accuracy": 0.14142293035984038,
      "num_tokens": 8110506.0,
      "step": 3705
    },
    {
      "entropy": 6.5264497756958,
      "epoch": 3.959957287773625,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00039313838734439605,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14747968167066575,
      "num_tokens": 8121999.0,
      "step": 3710
    },
    {
      "entropy": 6.481518793106079,
      "epoch": 3.965296316070475,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039277838071055145,
      "loss": 6.1913,
      "mean_token_accuracy": 0.14093517512083054,
      "num_tokens": 8132884.0,
      "step": 3715
    },
    {
      "entropy": 6.576885175704956,
      "epoch": 3.970635344367325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000392417958268889,
      "loss": 6.1302,
      "mean_token_accuracy": 0.14702757447957993,
      "num_tokens": 8143344.0,
      "step": 3720
    },
    {
      "entropy": 6.490785789489746,
      "epoch": 3.975974372664175,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039205712129185337,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14834580570459366,
      "num_tokens": 8154658.0,
      "step": 3725
    },
    {
      "entropy": 6.4275837421417235,
      "epoch": 3.981313400961025,
      "grad_norm": 0.875,
      "learning_rate": 0.0003916958710533528,
      "loss": 6.082,
      "mean_token_accuracy": 0.14838915020227433,
      "num_tokens": 8166057.0,
      "step": 3730
    },
    {
      "entropy": 6.574824666976928,
      "epoch": 3.986652429257875,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003913342088287541,
      "loss": 6.106,
      "mean_token_accuracy": 0.14580099284648895,
      "num_tokens": 8176470.0,
      "step": 3735
    },
    {
      "entropy": 6.523640108108521,
      "epoch": 3.9919914575547253,
      "grad_norm": 0.984375,
      "learning_rate": 0.000390972135894879,
      "loss": 6.1629,
      "mean_token_accuracy": 0.13956414237618447,
      "num_tokens": 8186458.0,
      "step": 3740
    },
    {
      "entropy": 6.550470685958862,
      "epoch": 3.997330485851575,
      "grad_norm": 0.984375,
      "learning_rate": 0.00039060965352999926,
      "loss": 6.25,
      "mean_token_accuracy": 0.13434221893548964,
      "num_tokens": 8197826.0,
      "step": 3745
    },
    {
      "entropy": 6.523906389872233,
      "epoch": 4.00213561131874,
      "grad_norm": 0.890625,
      "learning_rate": 0.0003902467630138317,
      "loss": 6.0554,
      "mean_token_accuracy": 0.14978978451755312,
      "num_tokens": 8208809.0,
      "step": 3750
    },
    {
      "entropy": 6.455614471435547,
      "epoch": 4.00747463961559,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038988346562753435,
      "loss": 5.8155,
      "mean_token_accuracy": 0.16545109897851945,
      "num_tokens": 8218740.0,
      "step": 3755
    },
    {
      "entropy": 6.479466819763184,
      "epoch": 4.0128136679124395,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00038951976265370176,
      "loss": 5.8186,
      "mean_token_accuracy": 0.1650289312005043,
      "num_tokens": 8230360.0,
      "step": 3760
    },
    {
      "entropy": 6.4658434867858885,
      "epoch": 4.01815269620929,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00038915565537636003,
      "loss": 5.8426,
      "mean_token_accuracy": 0.16136547774076462,
      "num_tokens": 8240886.0,
      "step": 3765
    },
    {
      "entropy": 6.439322376251221,
      "epoch": 4.0234917245061395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000388791145080963,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16565286070108415,
      "num_tokens": 8252103.0,
      "step": 3770
    },
    {
      "entropy": 6.506541919708252,
      "epoch": 4.02883075280299,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003884262330543872,
      "loss": 5.9089,
      "mean_token_accuracy": 0.15189366191625595,
      "num_tokens": 8262890.0,
      "step": 3775
    },
    {
      "entropy": 6.481657361984253,
      "epoch": 4.03416978109984,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00038806092058492727,
      "loss": 5.8308,
      "mean_token_accuracy": 0.15886562168598176,
      "num_tokens": 8273562.0,
      "step": 3780
    },
    {
      "entropy": 6.5099996566772464,
      "epoch": 4.03950880939669,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003876952089622918,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15011855959892273,
      "num_tokens": 8285104.0,
      "step": 3785
    },
    {
      "entropy": 6.495954179763794,
      "epoch": 4.04484783769354,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003873290994775984,
      "loss": 5.8799,
      "mean_token_accuracy": 0.15607954263687135,
      "num_tokens": 8296760.0,
      "step": 3790
    },
    {
      "entropy": 6.490471601486206,
      "epoch": 4.05018686599039,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00038696259342336966,
      "loss": 5.8882,
      "mean_token_accuracy": 0.16234036087989806,
      "num_tokens": 8306929.0,
      "step": 3795
    },
    {
      "entropy": 6.442892169952392,
      "epoch": 4.05552589428724,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038659569209352766,
      "loss": 5.9147,
      "mean_token_accuracy": 0.15829430669546127,
      "num_tokens": 8316736.0,
      "step": 3800
    },
    {
      "entropy": 6.505135583877563,
      "epoch": 4.06086492258409,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038622839678339034,
      "loss": 5.8808,
      "mean_token_accuracy": 0.157439686357975,
      "num_tokens": 8327688.0,
      "step": 3805
    },
    {
      "entropy": 6.4566991329193115,
      "epoch": 4.06620395088094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038586070878966663,
      "loss": 5.8106,
      "mean_token_accuracy": 0.16600643396377562,
      "num_tokens": 8337986.0,
      "step": 3810
    },
    {
      "entropy": 6.422807598114014,
      "epoch": 4.0715429791777895,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003854926294104514,
      "loss": 5.8251,
      "mean_token_accuracy": 0.1629556953907013,
      "num_tokens": 8348473.0,
      "step": 3815
    },
    {
      "entropy": 6.492142152786255,
      "epoch": 4.07688200747464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003851241599452218,
      "loss": 5.8228,
      "mean_token_accuracy": 0.15974331200122832,
      "num_tokens": 8359563.0,
      "step": 3820
    },
    {
      "entropy": 6.476008796691895,
      "epoch": 4.082221035771489,
      "grad_norm": 1.0,
      "learning_rate": 0.00038475530169483185,
      "loss": 5.941,
      "mean_token_accuracy": 0.15269246846437454,
      "num_tokens": 8370965.0,
      "step": 3825
    },
    {
      "entropy": 6.416006135940552,
      "epoch": 4.08756006406834,
      "grad_norm": 0.90625,
      "learning_rate": 0.00038438605596150807,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15478795990347863,
      "num_tokens": 8382606.0,
      "step": 3830
    },
    {
      "entropy": 6.487174367904663,
      "epoch": 4.092899092365189,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038401642404884513,
      "loss": 5.8667,
      "mean_token_accuracy": 0.1620238795876503,
      "num_tokens": 8392618.0,
      "step": 3835
    },
    {
      "entropy": 6.451450061798096,
      "epoch": 4.09823812066204,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038364640726180096,
      "loss": 5.827,
      "mean_token_accuracy": 0.15819164365530014,
      "num_tokens": 8403210.0,
      "step": 3840
    },
    {
      "entropy": 6.457551288604736,
      "epoch": 4.103577148958889,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038327600690669244,
      "loss": 5.9271,
      "mean_token_accuracy": 0.15056874752044677,
      "num_tokens": 8413838.0,
      "step": 3845
    },
    {
      "entropy": 6.452711200714111,
      "epoch": 4.10891617725574,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00038290522429119034,
      "loss": 5.8383,
      "mean_token_accuracy": 0.1614674672484398,
      "num_tokens": 8425282.0,
      "step": 3850
    },
    {
      "entropy": 6.410413837432861,
      "epoch": 4.114255205552589,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038253406072431525,
      "loss": 5.8392,
      "mean_token_accuracy": 0.16616302728652954,
      "num_tokens": 8436762.0,
      "step": 3855
    },
    {
      "entropy": 6.500837230682373,
      "epoch": 4.11959423384944,
      "grad_norm": 1.0,
      "learning_rate": 0.00038216251751643245,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15862398594617844,
      "num_tokens": 8448636.0,
      "step": 3860
    },
    {
      "entropy": 6.445639896392822,
      "epoch": 4.124933262146289,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003817905959792477,
      "loss": 5.8186,
      "mean_token_accuracy": 0.16568127423524856,
      "num_tokens": 8459730.0,
      "step": 3865
    },
    {
      "entropy": 6.432496213912964,
      "epoch": 4.1302722904431395,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003814182974258023,
      "loss": 5.836,
      "mean_token_accuracy": 0.16377303451299668,
      "num_tokens": 8470628.0,
      "step": 3870
    },
    {
      "entropy": 6.410172033309936,
      "epoch": 4.135611318739989,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038104562317046873,
      "loss": 5.8404,
      "mean_token_accuracy": 0.1582864850759506,
      "num_tokens": 8480785.0,
      "step": 3875
    },
    {
      "entropy": 6.400460815429687,
      "epoch": 4.140950347036839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003806725745289455,
      "loss": 5.7772,
      "mean_token_accuracy": 0.16919722110033036,
      "num_tokens": 8491596.0,
      "step": 3880
    },
    {
      "entropy": 6.427718925476074,
      "epoch": 4.146289375333689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038029915281825334,
      "loss": 5.8855,
      "mean_token_accuracy": 0.1604493737220764,
      "num_tokens": 8502458.0,
      "step": 3885
    },
    {
      "entropy": 6.4377576351165775,
      "epoch": 4.151628403630539,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037992535935672974,
      "loss": 5.8356,
      "mean_token_accuracy": 0.1650594249367714,
      "num_tokens": 8513018.0,
      "step": 3890
    },
    {
      "entropy": 6.484112358093261,
      "epoch": 4.156967431927389,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00037955119546402474,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16013461947441102,
      "num_tokens": 8523757.0,
      "step": 3895
    },
    {
      "entropy": 6.320549201965332,
      "epoch": 4.162306460224239,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003791766624610961,
      "loss": 5.7758,
      "mean_token_accuracy": 0.16299208998680115,
      "num_tokens": 8535446.0,
      "step": 3900
    },
    {
      "entropy": 6.370971965789795,
      "epoch": 4.167645488521089,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003788017616702048,
      "loss": 5.7648,
      "mean_token_accuracy": 0.168772192299366,
      "num_tokens": 8545868.0,
      "step": 3905
    },
    {
      "entropy": 6.428561162948609,
      "epoch": 4.172984516817939,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003784264944149102,
      "loss": 5.8338,
      "mean_token_accuracy": 0.15697346329689027,
      "num_tokens": 8556465.0,
      "step": 3910
    },
    {
      "entropy": 6.4201459884643555,
      "epoch": 4.1783235451147895,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003780508620200655,
      "loss": 5.866,
      "mean_token_accuracy": 0.16035554856061934,
      "num_tokens": 8568227.0,
      "step": 3915
    },
    {
      "entropy": 6.408050775527954,
      "epoch": 4.183662573411639,
      "grad_norm": 1.0,
      "learning_rate": 0.0003776748658118128,
      "loss": 5.8827,
      "mean_token_accuracy": 0.16062171459198,
      "num_tokens": 8579309.0,
      "step": 3920
    },
    {
      "entropy": 6.407215690612793,
      "epoch": 4.1890016017084895,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00037729850711757903,
      "loss": 5.8097,
      "mean_token_accuracy": 0.16254528909921645,
      "num_tokens": 8590487.0,
      "step": 3925
    },
    {
      "entropy": 6.501875162124634,
      "epoch": 4.194340630005339,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003769217872660703,
      "loss": 5.9053,
      "mean_token_accuracy": 0.15888245403766632,
      "num_tokens": 8600940.0,
      "step": 3930
    },
    {
      "entropy": 6.375162887573242,
      "epoch": 4.199679658302189,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037654470758726816,
      "loss": 5.6835,
      "mean_token_accuracy": 0.1780616506934166,
      "num_tokens": 8611322.0,
      "step": 3935
    },
    {
      "entropy": 6.3994794368743895,
      "epoch": 4.205018686599039,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037616726941242445,
      "loss": 5.8733,
      "mean_token_accuracy": 0.15676409006118774,
      "num_tokens": 8622055.0,
      "step": 3940
    },
    {
      "entropy": 6.383853530883789,
      "epoch": 4.210357714895889,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003757894740740565,
      "loss": 5.9048,
      "mean_token_accuracy": 0.16065036207437516,
      "num_tokens": 8633515.0,
      "step": 3945
    },
    {
      "entropy": 6.382141780853272,
      "epoch": 4.215696743192739,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003754113229059426,
      "loss": 5.827,
      "mean_token_accuracy": 0.16290465593338013,
      "num_tokens": 8644199.0,
      "step": 3950
    },
    {
      "entropy": 6.386103391647339,
      "epoch": 4.221035771489589,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003750328172431176,
      "loss": 5.902,
      "mean_token_accuracy": 0.15742135792970657,
      "num_tokens": 8655025.0,
      "step": 3955
    },
    {
      "entropy": 6.384013700485229,
      "epoch": 4.226374799786439,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003746539584218673,
      "loss": 5.852,
      "mean_token_accuracy": 0.16765398383140565,
      "num_tokens": 8666251.0,
      "step": 3960
    },
    {
      "entropy": 6.393583822250366,
      "epoch": 4.231713828083289,
      "grad_norm": 0.984375,
      "learning_rate": 0.00037427474777972477,
      "loss": 5.8105,
      "mean_token_accuracy": 0.1669989585876465,
      "num_tokens": 8676727.0,
      "step": 3965
    },
    {
      "entropy": 6.466755676269531,
      "epoch": 4.237052856380139,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000373895186655465,
      "loss": 5.8869,
      "mean_token_accuracy": 0.16504256725311278,
      "num_tokens": 8687498.0,
      "step": 3970
    },
    {
      "entropy": 6.458343982696533,
      "epoch": 4.242391884676989,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037351527638910035,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16048015058040618,
      "num_tokens": 8698505.0,
      "step": 3975
    },
    {
      "entropy": 6.432484912872314,
      "epoch": 4.2477309129738385,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003731350183218755,
      "loss": 5.9565,
      "mean_token_accuracy": 0.1561967760324478,
      "num_tokens": 8709349.0,
      "step": 3980
    },
    {
      "entropy": 6.471627855300904,
      "epoch": 4.253069941270689,
      "grad_norm": 1.0,
      "learning_rate": 0.0003727544137962637,
      "loss": 5.7887,
      "mean_token_accuracy": 0.1591472879052162,
      "num_tokens": 8719515.0,
      "step": 3985
    },
    {
      "entropy": 6.417425298690796,
      "epoch": 4.2584089695675384,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003723734641559607,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15346258357167245,
      "num_tokens": 8731030.0,
      "step": 3990
    },
    {
      "entropy": 6.447788286209106,
      "epoch": 4.263747997864389,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003719921707458809,
      "loss": 5.8551,
      "mean_token_accuracy": 0.15848195552825928,
      "num_tokens": 8742776.0,
      "step": 3995
    },
    {
      "entropy": 6.39237151145935,
      "epoch": 4.269087026161238,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003716105349121523,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1576450362801552,
      "num_tokens": 8753994.0,
      "step": 4000
    },
    {
      "epoch": 4.269087026161238,
      "eval_entropy": 6.213875951395407,
      "eval_loss": 6.626263618469238,
      "eval_mean_token_accuracy": 0.12890922446529587,
      "eval_num_tokens": 8753994.0,
      "eval_runtime": 1.8138,
      "eval_samples_per_second": 1698.115,
      "eval_steps_per_second": 212.264,
      "step": 4000
    },
    {
      "entropy": 6.4048929691314695,
      "epoch": 4.274426054458089,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037122855800211194,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15603681802749633,
      "num_tokens": 8765166.0,
      "step": 4005
    },
    {
      "entropy": 6.335198831558228,
      "epoch": 4.279765082754938,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003708462413643009,
      "loss": 5.7886,
      "mean_token_accuracy": 0.16002152264118194,
      "num_tokens": 8777690.0,
      "step": 4010
    },
    {
      "entropy": 6.390238523483276,
      "epoch": 4.285104111051789,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003704635863484596,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1600464291870594,
      "num_tokens": 8789176.0,
      "step": 4015
    },
    {
      "entropy": 6.406296682357788,
      "epoch": 4.290443139348638,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037008059430552315,
      "loss": 5.9404,
      "mean_token_accuracy": 0.1529441237449646,
      "num_tokens": 8799810.0,
      "step": 4020
    },
    {
      "entropy": 6.389851474761963,
      "epoch": 4.295782167645489,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003696972665876165,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15673968195915222,
      "num_tokens": 8811226.0,
      "step": 4025
    },
    {
      "entropy": 6.507057332992554,
      "epoch": 4.301121195942338,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003693136045480496,
      "loss": 5.9014,
      "mean_token_accuracy": 0.15902647078037263,
      "num_tokens": 8821623.0,
      "step": 4030
    },
    {
      "entropy": 6.451570701599121,
      "epoch": 4.3064602242391885,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036892960954131287,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15746306031942367,
      "num_tokens": 8832146.0,
      "step": 4035
    },
    {
      "entropy": 6.423214149475098,
      "epoch": 4.311799252536039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003685452829230721,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16206436604261398,
      "num_tokens": 8842087.0,
      "step": 4040
    },
    {
      "entropy": 6.462438488006592,
      "epoch": 4.317138280832888,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003681606260501637,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15548368096351622,
      "num_tokens": 8853193.0,
      "step": 4045
    },
    {
      "entropy": 6.397952508926392,
      "epoch": 4.322477309129738,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036777564028059036,
      "loss": 5.7812,
      "mean_token_accuracy": 0.16826205998659133,
      "num_tokens": 8863751.0,
      "step": 4050
    },
    {
      "entropy": 6.370702838897705,
      "epoch": 4.327816337426588,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003673903269735156,
      "loss": 5.8334,
      "mean_token_accuracy": 0.15850836783647537,
      "num_tokens": 8875127.0,
      "step": 4055
    },
    {
      "entropy": 6.426618528366089,
      "epoch": 4.333155365723439,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003670046874892595,
      "loss": 5.8234,
      "mean_token_accuracy": 0.1664125144481659,
      "num_tokens": 8885851.0,
      "step": 4060
    },
    {
      "entropy": 6.402348804473877,
      "epoch": 4.338494394020288,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003666187231892936,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15492249578237532,
      "num_tokens": 8898001.0,
      "step": 4065
    },
    {
      "entropy": 6.467628765106201,
      "epoch": 4.343833422317139,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036623243543623603,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1625946819782257,
      "num_tokens": 8908852.0,
      "step": 4070
    },
    {
      "entropy": 6.443071126937866,
      "epoch": 4.349172450613988,
      "grad_norm": 1.0,
      "learning_rate": 0.0003658458255938471,
      "loss": 5.8299,
      "mean_token_accuracy": 0.16284757256507873,
      "num_tokens": 8919580.0,
      "step": 4075
    },
    {
      "entropy": 6.377685451507569,
      "epoch": 4.354511478910839,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036545889502702405,
      "loss": 5.9086,
      "mean_token_accuracy": 0.1580785721540451,
      "num_tokens": 8930218.0,
      "step": 4080
    },
    {
      "entropy": 6.4977436542510985,
      "epoch": 4.359850507207688,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036507164510179665,
      "loss": 5.9094,
      "mean_token_accuracy": 0.15574317276477814,
      "num_tokens": 8941960.0,
      "step": 4085
    },
    {
      "entropy": 6.398830318450928,
      "epoch": 4.3651895355045385,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00036468407718532177,
      "loss": 5.8806,
      "mean_token_accuracy": 0.15769995599985123,
      "num_tokens": 8953010.0,
      "step": 4090
    },
    {
      "entropy": 6.401963901519776,
      "epoch": 4.370528563801388,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003642961926458792,
      "loss": 5.9305,
      "mean_token_accuracy": 0.16030178517103194,
      "num_tokens": 8964262.0,
      "step": 4095
    },
    {
      "entropy": 6.4202672958374025,
      "epoch": 4.375867592098238,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00036390799285286645,
      "loss": 5.9664,
      "mean_token_accuracy": 0.15303387492895126,
      "num_tokens": 8976737.0,
      "step": 4100
    },
    {
      "entropy": 6.386706018447876,
      "epoch": 4.381206620395088,
      "grad_norm": 0.921875,
      "learning_rate": 0.00036351947917679414,
      "loss": 5.9042,
      "mean_token_accuracy": 0.15970954447984695,
      "num_tokens": 8988149.0,
      "step": 4105
    },
    {
      "entropy": 6.458559894561768,
      "epoch": 4.386545648691938,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003631306529892808,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15646170675754548,
      "num_tokens": 8998367.0,
      "step": 4110
    },
    {
      "entropy": 6.402514553070068,
      "epoch": 4.391884676988788,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036274151566304837,
      "loss": 5.9144,
      "mean_token_accuracy": 0.15345945209264755,
      "num_tokens": 9009249.0,
      "step": 4115
    },
    {
      "entropy": 6.383094215393067,
      "epoch": 4.397223705285638,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00036235206857191743,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16084175705909728,
      "num_tokens": 9020083.0,
      "step": 4120
    },
    {
      "entropy": 6.472617673873901,
      "epoch": 4.402562733582488,
      "grad_norm": 0.9375,
      "learning_rate": 0.0003619623130908017,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16436166763305665,
      "num_tokens": 9031477.0,
      "step": 4125
    },
    {
      "entropy": 6.413021230697632,
      "epoch": 4.407901761879338,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003615722505957043,
      "loss": 5.905,
      "mean_token_accuracy": 0.1590799331665039,
      "num_tokens": 9041429.0,
      "step": 4130
    },
    {
      "entropy": 6.363875246047973,
      "epoch": 4.413240790176188,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00036118188246371156,
      "loss": 5.822,
      "mean_token_accuracy": 0.16231572031974792,
      "num_tokens": 9051776.0,
      "step": 4135
    },
    {
      "entropy": 6.37334098815918,
      "epoch": 4.418579818473038,
      "grad_norm": 0.953125,
      "learning_rate": 0.00036079121007298934,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16210116893053056,
      "num_tokens": 9063077.0,
      "step": 4140
    },
    {
      "entropy": 6.474695777893066,
      "epoch": 4.423918846769888,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036040023480277724,
      "loss": 5.9025,
      "mean_token_accuracy": 0.158287712931633,
      "num_tokens": 9074130.0,
      "step": 4145
    },
    {
      "entropy": 6.411297225952149,
      "epoch": 4.429257875066738,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036000895803338465,
      "loss": 5.8647,
      "mean_token_accuracy": 0.1616749420762062,
      "num_tokens": 9083976.0,
      "step": 4150
    },
    {
      "entropy": 6.52132248878479,
      "epoch": 4.4345969033635875,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000359617381146185,
      "loss": 5.9975,
      "mean_token_accuracy": 0.14539264291524887,
      "num_tokens": 9095029.0,
      "step": 4155
    },
    {
      "entropy": 6.491507244110108,
      "epoch": 4.439935931660438,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003592255055236113,
      "loss": 5.9447,
      "mean_token_accuracy": 0.15878929644823075,
      "num_tokens": 9105382.0,
      "step": 4160
    },
    {
      "entropy": 6.381310367584229,
      "epoch": 4.445274959957287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003588333325491515,
      "loss": 5.8687,
      "mean_token_accuracy": 0.15649351477622986,
      "num_tokens": 9116120.0,
      "step": 4165
    },
    {
      "entropy": 6.3444465637207035,
      "epoch": 4.450613988254138,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035844086360734304,
      "loss": 5.9347,
      "mean_token_accuracy": 0.15064095035195352,
      "num_tokens": 9127492.0,
      "step": 4170
    },
    {
      "entropy": 6.419450426101685,
      "epoch": 4.455953016550987,
      "grad_norm": 0.890625,
      "learning_rate": 0.0003580481000837683,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16420729458332062,
      "num_tokens": 9139382.0,
      "step": 4175
    },
    {
      "entropy": 6.37151312828064,
      "epoch": 4.461292044847838,
      "grad_norm": 1.0,
      "learning_rate": 0.00035765504336504974,
      "loss": 5.9323,
      "mean_token_accuracy": 0.15446227490901948,
      "num_tokens": 9150484.0,
      "step": 4180
    },
    {
      "entropy": 6.447879409790039,
      "epoch": 4.466631073144688,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035726169483884493,
      "loss": 5.9109,
      "mean_token_accuracy": 0.15431981533765793,
      "num_tokens": 9161110.0,
      "step": 4185
    },
    {
      "entropy": 6.455252599716187,
      "epoch": 4.471970101441538,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003568680558938415,
      "loss": 5.8833,
      "mean_token_accuracy": 0.1610197678208351,
      "num_tokens": 9171786.0,
      "step": 4190
    },
    {
      "entropy": 6.417928457260132,
      "epoch": 4.477309129738388,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003564741279197526,
      "loss": 5.9639,
      "mean_token_accuracy": 0.15531326979398727,
      "num_tokens": 9183250.0,
      "step": 4195
    },
    {
      "entropy": 6.45450611114502,
      "epoch": 4.4826481580352375,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035607991230731154,
      "loss": 5.8953,
      "mean_token_accuracy": 0.15608533322811127,
      "num_tokens": 9194261.0,
      "step": 4200
    },
    {
      "entropy": 6.38051290512085,
      "epoch": 4.487987186332088,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003556854104482673,
      "loss": 5.891,
      "mean_token_accuracy": 0.1608704909682274,
      "num_tokens": 9205696.0,
      "step": 4205
    },
    {
      "entropy": 6.395957183837891,
      "epoch": 4.493326214628937,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003552906237353793,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16094670444726944,
      "num_tokens": 9217027.0,
      "step": 4210
    },
    {
      "entropy": 6.42204008102417,
      "epoch": 4.498665242925788,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003548955535624128,
      "loss": 5.9264,
      "mean_token_accuracy": 0.15839983522891998,
      "num_tokens": 9227916.0,
      "step": 4215
    },
    {
      "entropy": 6.433207321166992,
      "epoch": 4.504004271222637,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003545002013241337,
      "loss": 5.8787,
      "mean_token_accuracy": 0.1681976228952408,
      "num_tokens": 9239401.0,
      "step": 4220
    },
    {
      "entropy": 6.345034646987915,
      "epoch": 4.509343299519488,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003541045684163035,
      "loss": 5.8769,
      "mean_token_accuracy": 0.15489550828933715,
      "num_tokens": 9249258.0,
      "step": 4225
    },
    {
      "entropy": 6.437056159973144,
      "epoch": 4.514682327816337,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035370865623567495,
      "loss": 5.894,
      "mean_token_accuracy": 0.15679936856031418,
      "num_tokens": 9259958.0,
      "step": 4230
    },
    {
      "entropy": 6.40920934677124,
      "epoch": 4.520021356113188,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035331246617998654,
      "loss": 5.8184,
      "mean_token_accuracy": 0.15945262610912322,
      "num_tokens": 9270380.0,
      "step": 4235
    },
    {
      "entropy": 6.376594591140747,
      "epoch": 4.525360384410037,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003529159996479577,
      "loss": 5.882,
      "mean_token_accuracy": 0.16308974474668503,
      "num_tokens": 9281178.0,
      "step": 4240
    },
    {
      "entropy": 6.373546457290649,
      "epoch": 4.530699412706888,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035251925803928414,
      "loss": 5.7705,
      "mean_token_accuracy": 0.1718821719288826,
      "num_tokens": 9290987.0,
      "step": 4245
    },
    {
      "entropy": 6.434765577316284,
      "epoch": 4.536038441003737,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003521222427546326,
      "loss": 5.9256,
      "mean_token_accuracy": 0.15994654297828675,
      "num_tokens": 9302195.0,
      "step": 4250
    },
    {
      "entropy": 6.361167621612549,
      "epoch": 4.5413774693005875,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000351724955195636,
      "loss": 5.8783,
      "mean_token_accuracy": 0.1654079496860504,
      "num_tokens": 9312851.0,
      "step": 4255
    },
    {
      "entropy": 6.421001005172729,
      "epoch": 4.546716497597437,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00035132739676488854,
      "loss": 5.9486,
      "mean_token_accuracy": 0.15135279595851897,
      "num_tokens": 9324001.0,
      "step": 4260
    },
    {
      "entropy": 6.422706508636475,
      "epoch": 4.552055525894287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003509295688659409,
      "loss": 5.868,
      "mean_token_accuracy": 0.16096135824918748,
      "num_tokens": 9334092.0,
      "step": 4265
    },
    {
      "entropy": 6.419775152206421,
      "epoch": 4.557394554191137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035053147290329455,
      "loss": 5.916,
      "mean_token_accuracy": 0.15535657107830048,
      "num_tokens": 9345096.0,
      "step": 4270
    },
    {
      "entropy": 6.274848413467407,
      "epoch": 4.562733582487987,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000350133110282398,
      "loss": 5.9262,
      "mean_token_accuracy": 0.15504762530326843,
      "num_tokens": 9357184.0,
      "step": 4275
    },
    {
      "entropy": 6.463329076766968,
      "epoch": 4.568072610784837,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003497344824096407,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15729070752859114,
      "num_tokens": 9367392.0,
      "step": 4280
    },
    {
      "entropy": 6.472297477722168,
      "epoch": 4.573411639081687,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034933559069234876,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16340985745191575,
      "num_tokens": 9377138.0,
      "step": 4285
    },
    {
      "entropy": 6.411571645736695,
      "epoch": 4.578750667378537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00034893643653877973,
      "loss": 5.9533,
      "mean_token_accuracy": 0.152741901576519,
      "num_tokens": 9387777.0,
      "step": 4290
    },
    {
      "entropy": 6.442379903793335,
      "epoch": 4.584089695675387,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003485370213581177,
      "loss": 5.919,
      "mean_token_accuracy": 0.16460502445697783,
      "num_tokens": 9397880.0,
      "step": 4295
    },
    {
      "entropy": 6.380690717697144,
      "epoch": 4.589428723972237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003481373465604683,
      "loss": 5.825,
      "mean_token_accuracy": 0.15915247350931166,
      "num_tokens": 9408899.0,
      "step": 4300
    },
    {
      "entropy": 6.376305961608887,
      "epoch": 4.594767752269087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003477374135568536,
      "loss": 5.874,
      "mean_token_accuracy": 0.16272423788905144,
      "num_tokens": 9419599.0,
      "step": 4305
    },
    {
      "entropy": 6.412158107757568,
      "epoch": 4.6001067805659375,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003473372237592075,
      "loss": 5.9253,
      "mean_token_accuracy": 0.15421468615531922,
      "num_tokens": 9431055.0,
      "step": 4310
    },
    {
      "entropy": 6.416840648651123,
      "epoch": 4.605445808862787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034693677858037015,
      "loss": 5.86,
      "mean_token_accuracy": 0.16356259137392043,
      "num_tokens": 9442193.0,
      "step": 4315
    },
    {
      "entropy": 6.4224800109863285,
      "epoch": 4.6107848371596365,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003465360794340836,
      "loss": 5.914,
      "mean_token_accuracy": 0.15818674862384796,
      "num_tokens": 9453340.0,
      "step": 4320
    },
    {
      "entropy": 6.298796987533569,
      "epoch": 4.616123865456487,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034613512773498634,
      "loss": 5.8637,
      "mean_token_accuracy": 0.15737705677747726,
      "num_tokens": 9464163.0,
      "step": 4325
    },
    {
      "entropy": 6.346707963943482,
      "epoch": 4.621462893753337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003457339248986086,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16036440283060074,
      "num_tokens": 9473376.0,
      "step": 4330
    },
    {
      "entropy": 6.358528995513916,
      "epoch": 4.626801922050187,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003453324723413671,
      "loss": 5.819,
      "mean_token_accuracy": 0.17424627095460893,
      "num_tokens": 9484571.0,
      "step": 4335
    },
    {
      "entropy": 6.4063362121582035,
      "epoch": 4.632140950347037,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003449307714805603,
      "loss": 5.8956,
      "mean_token_accuracy": 0.16029897183179856,
      "num_tokens": 9495702.0,
      "step": 4340
    },
    {
      "entropy": 6.424567985534668,
      "epoch": 4.637479978643887,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003445288237343632,
      "loss": 5.9546,
      "mean_token_accuracy": 0.15446095466613768,
      "num_tokens": 9507296.0,
      "step": 4345
    },
    {
      "entropy": 6.3962554931640625,
      "epoch": 4.642819006940737,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003441266305218224,
      "loss": 5.9453,
      "mean_token_accuracy": 0.15589361935853957,
      "num_tokens": 9518198.0,
      "step": 4350
    },
    {
      "entropy": 6.459865522384644,
      "epoch": 4.648158035237587,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00034372419326285126,
      "loss": 5.9588,
      "mean_token_accuracy": 0.1505233347415924,
      "num_tokens": 9529711.0,
      "step": 4355
    },
    {
      "entropy": 6.455620670318604,
      "epoch": 4.653497063534437,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034332151337822454,
      "loss": 5.8618,
      "mean_token_accuracy": 0.1631107211112976,
      "num_tokens": 9540543.0,
      "step": 4360
    },
    {
      "entropy": 6.328192329406738,
      "epoch": 4.6588360918312866,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003429185922895736,
      "loss": 5.8707,
      "mean_token_accuracy": 0.1605295106768608,
      "num_tokens": 9550502.0,
      "step": 4365
    },
    {
      "entropy": 6.371403598785401,
      "epoch": 4.664175120128137,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034251543141938144,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15743122845888138,
      "num_tokens": 9561909.0,
      "step": 4370
    },
    {
      "entropy": 6.36637396812439,
      "epoch": 4.6695141484249865,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003421120321909776,
      "loss": 5.8222,
      "mean_token_accuracy": 0.1606006696820259,
      "num_tokens": 9572380.0,
      "step": 4375
    },
    {
      "entropy": 6.3657632827758786,
      "epoch": 4.674853176721837,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003417083960285329,
      "loss": 5.82,
      "mean_token_accuracy": 0.1607291281223297,
      "num_tokens": 9584188.0,
      "step": 4380
    },
    {
      "entropy": 6.361291742324829,
      "epoch": 4.680192205018686,
      "grad_norm": 1.0625,
      "learning_rate": 0.000341304524357055,
      "loss": 5.8803,
      "mean_token_accuracy": 0.1607501268386841,
      "num_tokens": 9594449.0,
      "step": 4385
    },
    {
      "entropy": 6.455878353118896,
      "epoch": 4.685531233315537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003409004186023826,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16451315730810165,
      "num_tokens": 9604655.0,
      "step": 4390
    },
    {
      "entropy": 6.440843200683593,
      "epoch": 4.690870261612386,
      "grad_norm": 1.0,
      "learning_rate": 0.0003404960801911812,
      "loss": 5.9775,
      "mean_token_accuracy": 0.15684316754341127,
      "num_tokens": 9615715.0,
      "step": 4395
    },
    {
      "entropy": 6.388810873031616,
      "epoch": 4.696209289909237,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003400915105509375,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15807077884674073,
      "num_tokens": 9626143.0,
      "step": 4400
    },
    {
      "entropy": 6.403064346313476,
      "epoch": 4.701548318206086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003396867111099544,
      "loss": 5.9027,
      "mean_token_accuracy": 0.15824245810508727,
      "num_tokens": 9638196.0,
      "step": 4405
    },
    {
      "entropy": 6.420948076248169,
      "epoch": 4.706887346502937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003392816832973464,
      "loss": 5.8988,
      "mean_token_accuracy": 0.15165719091892244,
      "num_tokens": 9649444.0,
      "step": 4410
    },
    {
      "entropy": 6.393687057495117,
      "epoch": 4.712226374799786,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000338876428543034,
      "loss": 5.9455,
      "mean_token_accuracy": 0.15852398574352264,
      "num_tokens": 9662529.0,
      "step": 4415
    },
    {
      "entropy": 6.482411241531372,
      "epoch": 4.7175654030966365,
      "grad_norm": 1.078125,
      "learning_rate": 0.000338470948277739,
      "loss": 5.9431,
      "mean_token_accuracy": 0.1514291748404503,
      "num_tokens": 9673335.0,
      "step": 4420
    },
    {
      "entropy": 6.447452783584595,
      "epoch": 4.722904431393486,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0003380652439329793,
      "loss": 5.968,
      "mean_token_accuracy": 0.15461814552545547,
      "num_tokens": 9685427.0,
      "step": 4425
    },
    {
      "entropy": 6.402762079238892,
      "epoch": 4.728243459690336,
      "grad_norm": 1.0,
      "learning_rate": 0.0003376593169410639,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16269552260637282,
      "num_tokens": 9696759.0,
      "step": 4430
    },
    {
      "entropy": 6.435091543197632,
      "epoch": 4.733582487987186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033725316873508805,
      "loss": 5.9588,
      "mean_token_accuracy": 0.15220657736063004,
      "num_tokens": 9707548.0,
      "step": 4435
    },
    {
      "entropy": 6.347317123413086,
      "epoch": 4.738921516284036,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003368468007489276,
      "loss": 5.9024,
      "mean_token_accuracy": 0.15936826914548874,
      "num_tokens": 9718663.0,
      "step": 4440
    },
    {
      "entropy": 6.454331493377685,
      "epoch": 4.744260544580886,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003364402144172346,
      "loss": 5.938,
      "mean_token_accuracy": 0.15357836484909057,
      "num_tokens": 9729742.0,
      "step": 4445
    },
    {
      "entropy": 6.449036645889282,
      "epoch": 4.749599572877736,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003360334111754318,
      "loss": 5.8918,
      "mean_token_accuracy": 0.16091630458831788,
      "num_tokens": 9739907.0,
      "step": 4450
    },
    {
      "entropy": 6.392539215087891,
      "epoch": 4.754938601174587,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033562639245970807,
      "loss": 5.8765,
      "mean_token_accuracy": 0.1536475419998169,
      "num_tokens": 9750813.0,
      "step": 4455
    },
    {
      "entropy": 6.446838760375977,
      "epoch": 4.760277629471436,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003352191597070123,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15809690654277803,
      "num_tokens": 9761200.0,
      "step": 4460
    },
    {
      "entropy": 6.389144659042358,
      "epoch": 4.765616657768286,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033481171435504985,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15566133260726928,
      "num_tokens": 9772075.0,
      "step": 4465
    },
    {
      "entropy": 6.391507959365844,
      "epoch": 4.770955686065136,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00033440405784227606,
      "loss": 5.9611,
      "mean_token_accuracy": 0.1576541468501091,
      "num_tokens": 9784391.0,
      "step": 4470
    },
    {
      "entropy": 6.4490166187286375,
      "epoch": 4.7762947143619865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033399619160789186,
      "loss": 5.8904,
      "mean_token_accuracy": 0.15303320437669754,
      "num_tokens": 9795781.0,
      "step": 4475
    },
    {
      "entropy": 6.378712892532349,
      "epoch": 4.781633742658836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00033358811709183883,
      "loss": 5.8793,
      "mean_token_accuracy": 0.15918490439653396,
      "num_tokens": 9807114.0,
      "step": 4480
    },
    {
      "entropy": 6.36537618637085,
      "epoch": 4.786972770955686,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003331798357347937,
      "loss": 5.8999,
      "mean_token_accuracy": 0.1550626814365387,
      "num_tokens": 9817936.0,
      "step": 4485
    },
    {
      "entropy": 6.399729681015015,
      "epoch": 4.792311799252536,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033277134897816336,
      "loss": 5.8617,
      "mean_token_accuracy": 0.15773180574178697,
      "num_tokens": 9828264.0,
      "step": 4490
    },
    {
      "entropy": 6.409576654434204,
      "epoch": 4.797650827549386,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033236265826407996,
      "loss": 5.8669,
      "mean_token_accuracy": 0.1598842889070511,
      "num_tokens": 9840543.0,
      "step": 4495
    },
    {
      "entropy": 6.422646856307983,
      "epoch": 4.802989855846236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003319537650353958,
      "loss": 5.9755,
      "mean_token_accuracy": 0.15414716601371764,
      "num_tokens": 9851995.0,
      "step": 4500
    },
    {
      "epoch": 4.802989855846236,
      "eval_entropy": 6.197350484674627,
      "eval_loss": 6.585744857788086,
      "eval_mean_token_accuracy": 0.13170837133929328,
      "eval_num_tokens": 9851995.0,
      "eval_runtime": 2.0221,
      "eval_samples_per_second": 1523.147,
      "eval_steps_per_second": 190.393,
      "step": 4500
    },
    {
      "entropy": 6.369800138473511,
      "epoch": 4.808328884143086,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003315446707356778,
      "loss": 5.9459,
      "mean_token_accuracy": 0.1581083983182907,
      "num_tokens": 9863523.0,
      "step": 4505
    },
    {
      "entropy": 6.387823820114136,
      "epoch": 4.813667912439936,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0003311353768092032,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15540545284748078,
      "num_tokens": 9875104.0,
      "step": 4510
    },
    {
      "entropy": 6.379478693008423,
      "epoch": 4.819006940736786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033072588470095364,
      "loss": 5.9428,
      "mean_token_accuracy": 0.15790754109621047,
      "num_tokens": 9886163.0,
      "step": 4515
    },
    {
      "entropy": 6.373296785354614,
      "epoch": 4.824345969033636,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003303161958566107,
      "loss": 5.8736,
      "mean_token_accuracy": 0.15861729979515077,
      "num_tokens": 9897548.0,
      "step": 4520
    },
    {
      "entropy": 6.405317878723144,
      "epoch": 4.829684997330486,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00032990631172255026,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1649320423603058,
      "num_tokens": 9908047.0,
      "step": 4525
    },
    {
      "entropy": 6.356839275360107,
      "epoch": 4.835024025627336,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003294962337458379,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16006208807229996,
      "num_tokens": 9919652.0,
      "step": 4530
    },
    {
      "entropy": 6.406173419952393,
      "epoch": 4.840363053924186,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003290859633742233,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16061726957559586,
      "num_tokens": 9930755.0,
      "step": 4535
    },
    {
      "entropy": 6.35872449874878,
      "epoch": 4.8457020822210355,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003286755020561356,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16003422439098358,
      "num_tokens": 9941525.0,
      "step": 4540
    },
    {
      "entropy": 6.37477560043335,
      "epoch": 4.851041110517886,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032826485124067803,
      "loss": 5.8577,
      "mean_token_accuracy": 0.16541478484869004,
      "num_tokens": 9953137.0,
      "step": 4545
    },
    {
      "entropy": 6.448293924331665,
      "epoch": 4.856380138814735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032785401237762256,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15062763392925263,
      "num_tokens": 9963089.0,
      "step": 4550
    },
    {
      "entropy": 6.412472105026245,
      "epoch": 4.861719167111586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032744298691740544,
      "loss": 5.949,
      "mean_token_accuracy": 0.15738110840320588,
      "num_tokens": 9974158.0,
      "step": 4555
    },
    {
      "entropy": 6.415805149078369,
      "epoch": 4.867058195408435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032703177631112154,
      "loss": 5.8822,
      "mean_token_accuracy": 0.16327403783798217,
      "num_tokens": 9985379.0,
      "step": 4560
    },
    {
      "entropy": 6.372489023208618,
      "epoch": 4.872397223705286,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00032662038201051915,
      "loss": 5.887,
      "mean_token_accuracy": 0.15769575014710427,
      "num_tokens": 9996683.0,
      "step": 4565
    },
    {
      "entropy": 6.359514284133911,
      "epoch": 4.877736252002135,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032620880546799534,
      "loss": 5.9138,
      "mean_token_accuracy": 0.16005149185657502,
      "num_tokens": 10006515.0,
      "step": 4570
    },
    {
      "entropy": 6.41174430847168,
      "epoch": 4.883075280298986,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00032579704813659054,
      "loss": 5.846,
      "mean_token_accuracy": 0.16435935646295546,
      "num_tokens": 10017154.0,
      "step": 4575
    },
    {
      "entropy": 6.370229864120484,
      "epoch": 4.888414308595836,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00032538511146998345,
      "loss": 5.8942,
      "mean_token_accuracy": 0.1569717898964882,
      "num_tokens": 10027962.0,
      "step": 4580
    },
    {
      "entropy": 6.360958433151245,
      "epoch": 4.8937533368926855,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032497299692248573,
      "loss": 5.8677,
      "mean_token_accuracy": 0.15199009478092193,
      "num_tokens": 10039636.0,
      "step": 4585
    },
    {
      "entropy": 6.382066631317139,
      "epoch": 4.899092365189535,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003245607059490372,
      "loss": 5.8499,
      "mean_token_accuracy": 0.15818813741207122,
      "num_tokens": 10049824.0,
      "step": 4590
    },
    {
      "entropy": 6.351303911209106,
      "epoch": 4.9044313934863855,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003241482400052005,
      "loss": 5.9098,
      "mean_token_accuracy": 0.15453121364116668,
      "num_tokens": 10060276.0,
      "step": 4595
    },
    {
      "entropy": 6.409463262557983,
      "epoch": 4.909770421783236,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032373560054715597,
      "loss": 5.9491,
      "mean_token_accuracy": 0.16075047105550766,
      "num_tokens": 10070724.0,
      "step": 4600
    },
    {
      "entropy": 6.450956201553344,
      "epoch": 4.915109450080085,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003233227890316965,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15876972898840905,
      "num_tokens": 10081989.0,
      "step": 4605
    },
    {
      "entropy": 6.3411389827728275,
      "epoch": 4.920448478376936,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003229098069162226,
      "loss": 5.8758,
      "mean_token_accuracy": 0.16423702090978623,
      "num_tokens": 10093105.0,
      "step": 4610
    },
    {
      "entropy": 6.377899312973023,
      "epoch": 4.925787506673785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003224966556587368,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1600843369960785,
      "num_tokens": 10103669.0,
      "step": 4615
    },
    {
      "entropy": 6.4453178405761715,
      "epoch": 4.931126534970636,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032208333671783897,
      "loss": 5.8947,
      "mean_token_accuracy": 0.16705036610364915,
      "num_tokens": 10113333.0,
      "step": 4620
    },
    {
      "entropy": 6.363779735565186,
      "epoch": 4.936465563267485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003216698515527209,
      "loss": 5.8659,
      "mean_token_accuracy": 0.15492594242095947,
      "num_tokens": 10125182.0,
      "step": 4625
    },
    {
      "entropy": 6.456665849685669,
      "epoch": 4.941804591564336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032125620162316116,
      "loss": 5.9578,
      "mean_token_accuracy": 0.16319140791893005,
      "num_tokens": 10136634.0,
      "step": 4630
    },
    {
      "entropy": 6.4088178157806395,
      "epoch": 4.947143619861185,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032084238838952015,
      "loss": 5.9326,
      "mean_token_accuracy": 0.15632062256336213,
      "num_tokens": 10147010.0,
      "step": 4635
    },
    {
      "entropy": 6.409919023513794,
      "epoch": 4.9524826481580355,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032042841331273467,
      "loss": 5.9426,
      "mean_token_accuracy": 0.1570739448070526,
      "num_tokens": 10158106.0,
      "step": 4640
    },
    {
      "entropy": 6.353962516784668,
      "epoch": 4.957821676454885,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032001427785431305,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16374620348215102,
      "num_tokens": 10168259.0,
      "step": 4645
    },
    {
      "entropy": 6.42985429763794,
      "epoch": 4.963160704751735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031959998347632956,
      "loss": 6.006,
      "mean_token_accuracy": 0.1434712141752243,
      "num_tokens": 10180508.0,
      "step": 4650
    },
    {
      "entropy": 6.492809057235718,
      "epoch": 4.968499733048585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003191855316414199,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15816870629787444,
      "num_tokens": 10190262.0,
      "step": 4655
    },
    {
      "entropy": 6.418329763412475,
      "epoch": 4.973838761345435,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031877092381277537,
      "loss": 5.9772,
      "mean_token_accuracy": 0.15402303636074066,
      "num_tokens": 10201523.0,
      "step": 4660
    },
    {
      "entropy": 6.317938280105591,
      "epoch": 4.979177789642285,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031835616145413814,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15760272294282912,
      "num_tokens": 10212831.0,
      "step": 4665
    },
    {
      "entropy": 6.404195642471313,
      "epoch": 4.984516817939135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003179412460297959,
      "loss": 5.9397,
      "mean_token_accuracy": 0.15713729113340377,
      "num_tokens": 10224321.0,
      "step": 4670
    },
    {
      "entropy": 6.39815320968628,
      "epoch": 4.989855846235985,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031752617900457656,
      "loss": 5.8457,
      "mean_token_accuracy": 0.16098662912845613,
      "num_tokens": 10234991.0,
      "step": 4675
    },
    {
      "entropy": 6.392170476913452,
      "epoch": 4.995194874532835,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031711096184384356,
      "loss": 5.913,
      "mean_token_accuracy": 0.1588407799601555,
      "num_tokens": 10245792.0,
      "step": 4680
    },
    {
      "entropy": 6.382403903537327,
      "epoch": 5.0,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003166955960134901,
      "loss": 5.929,
      "mean_token_accuracy": 0.15361620485782623,
      "num_tokens": 10255235.0,
      "step": 4685
    },
    {
      "entropy": 6.418357086181641,
      "epoch": 5.00533902829685,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003162800829799346,
      "loss": 5.7082,
      "mean_token_accuracy": 0.1671856552362442,
      "num_tokens": 10266176.0,
      "step": 4690
    },
    {
      "entropy": 6.424757814407348,
      "epoch": 5.0106780565937,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00031586442421011464,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17020085901021959,
      "num_tokens": 10276925.0,
      "step": 4695
    },
    {
      "entropy": 6.3981077671051025,
      "epoch": 5.01601708489055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031544862117148265,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16531572341918946,
      "num_tokens": 10288951.0,
      "step": 4700
    },
    {
      "entropy": 6.414116668701172,
      "epoch": 5.0213561131874,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003150326753320005,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1673135429620743,
      "num_tokens": 10299416.0,
      "step": 4705
    },
    {
      "entropy": 6.406787347793579,
      "epoch": 5.02669514148425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003146165881601341,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17430951297283173,
      "num_tokens": 10310065.0,
      "step": 4710
    },
    {
      "entropy": 6.389986848831176,
      "epoch": 5.0320341697811,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000314200361124848,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17484350502490997,
      "num_tokens": 10320859.0,
      "step": 4715
    },
    {
      "entropy": 6.363846254348755,
      "epoch": 5.03737319807795,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031378399569560104,
      "loss": 5.7124,
      "mean_token_accuracy": 0.17121145725250245,
      "num_tokens": 10332785.0,
      "step": 4720
    },
    {
      "entropy": 6.4665141105651855,
      "epoch": 5.0427122263748,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003133674933423404,
      "loss": 5.6947,
      "mean_token_accuracy": 0.17320917546749115,
      "num_tokens": 10343245.0,
      "step": 4725
    },
    {
      "entropy": 6.372145271301269,
      "epoch": 5.04805125467165,
      "grad_norm": 1.0,
      "learning_rate": 0.00031295085553549655,
      "loss": 5.6017,
      "mean_token_accuracy": 0.17550942450761794,
      "num_tokens": 10354819.0,
      "step": 4730
    },
    {
      "entropy": 6.339980840682983,
      "epoch": 5.0533902829685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003125340837459784,
      "loss": 5.675,
      "mean_token_accuracy": 0.17248251140117646,
      "num_tokens": 10364212.0,
      "step": 4735
    },
    {
      "entropy": 6.357005596160889,
      "epoch": 5.05872931126535,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003121171794451678,
      "loss": 5.6891,
      "mean_token_accuracy": 0.17237879633903502,
      "num_tokens": 10375749.0,
      "step": 4740
    },
    {
      "entropy": 6.390599584579467,
      "epoch": 5.0640683395621995,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003117001441049143,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17662997543811798,
      "num_tokens": 10386706.0,
      "step": 4745
    },
    {
      "entropy": 6.407583522796631,
      "epoch": 5.06940736785905,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031128297919753015,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1671500787138939,
      "num_tokens": 10397937.0,
      "step": 4750
    },
    {
      "entropy": 6.340428447723388,
      "epoch": 5.074746396155899,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003108656861957852,
      "loss": 5.6541,
      "mean_token_accuracy": 0.17233022451400756,
      "num_tokens": 10408472.0,
      "step": 4755
    },
    {
      "entropy": 6.367531251907349,
      "epoch": 5.08008542445275,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003104482665729012,
      "loss": 5.7561,
      "mean_token_accuracy": 0.16906343400478363,
      "num_tokens": 10418982.0,
      "step": 4760
    },
    {
      "entropy": 6.448597955703735,
      "epoch": 5.085424452749599,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003100307218025472,
      "loss": 5.7205,
      "mean_token_accuracy": 0.1782952591776848,
      "num_tokens": 10428889.0,
      "step": 4765
    },
    {
      "entropy": 6.3812178611755375,
      "epoch": 5.09076348104645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030961305335883393,
      "loss": 5.6373,
      "mean_token_accuracy": 0.17270799428224565,
      "num_tokens": 10439205.0,
      "step": 4770
    },
    {
      "entropy": 6.347117567062378,
      "epoch": 5.096102509343299,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003091952627163089,
      "loss": 5.7298,
      "mean_token_accuracy": 0.17093968838453294,
      "num_tokens": 10450923.0,
      "step": 4775
    },
    {
      "entropy": 6.4076886653900145,
      "epoch": 5.10144153764015,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003087773513499507,
      "loss": 5.6534,
      "mean_token_accuracy": 0.16741634607315065,
      "num_tokens": 10462110.0,
      "step": 4780
    },
    {
      "entropy": 6.417945766448975,
      "epoch": 5.106780565936999,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003083593207351644,
      "loss": 5.7396,
      "mean_token_accuracy": 0.16935442984104157,
      "num_tokens": 10473453.0,
      "step": 4785
    },
    {
      "entropy": 6.288882541656494,
      "epoch": 5.1121195942338495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030794117234777624,
      "loss": 5.5515,
      "mean_token_accuracy": 0.18606837838888168,
      "num_tokens": 10483228.0,
      "step": 4790
    },
    {
      "entropy": 6.3034350872039795,
      "epoch": 5.117458622530699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003075229076640278,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17581222504377364,
      "num_tokens": 10493767.0,
      "step": 4795
    },
    {
      "entropy": 6.403956604003906,
      "epoch": 5.1227976508275495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030710452816057137,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1753307983279228,
      "num_tokens": 10504618.0,
      "step": 4800
    },
    {
      "entropy": 6.377118015289307,
      "epoch": 5.128136679124399,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003066860353144649,
      "loss": 5.7459,
      "mean_token_accuracy": 0.16523779481649398,
      "num_tokens": 10516342.0,
      "step": 4805
    },
    {
      "entropy": 6.4060032844543455,
      "epoch": 5.133475707421249,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003062674306031662,
      "loss": 5.6746,
      "mean_token_accuracy": 0.17781568616628646,
      "num_tokens": 10526243.0,
      "step": 4810
    },
    {
      "entropy": 6.341823101043701,
      "epoch": 5.138814735718099,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000305848715504528,
      "loss": 5.6337,
      "mean_token_accuracy": 0.17808617651462555,
      "num_tokens": 10538557.0,
      "step": 4815
    },
    {
      "entropy": 6.391098070144653,
      "epoch": 5.144153764014949,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000305429891496793,
      "loss": 5.6862,
      "mean_token_accuracy": 0.1737893134355545,
      "num_tokens": 10549248.0,
      "step": 4820
    },
    {
      "entropy": 6.3253425598144535,
      "epoch": 5.1494927923118,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003050109600585881,
      "loss": 5.6163,
      "mean_token_accuracy": 0.18013107180595397,
      "num_tokens": 10560697.0,
      "step": 4825
    },
    {
      "entropy": 6.323402166366577,
      "epoch": 5.154831820608649,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003045919226689196,
      "loss": 5.6614,
      "mean_token_accuracy": 0.1754286155104637,
      "num_tokens": 10571880.0,
      "step": 4830
    },
    {
      "entropy": 6.394438791275024,
      "epoch": 5.1601708489055,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003041727808071679,
      "loss": 5.7492,
      "mean_token_accuracy": 0.1688705712556839,
      "num_tokens": 10582478.0,
      "step": 4835
    },
    {
      "entropy": 6.376315784454346,
      "epoch": 5.165509877202349,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0003037535359530823,
      "loss": 5.7517,
      "mean_token_accuracy": 0.16800296902656556,
      "num_tokens": 10594017.0,
      "step": 4840
    },
    {
      "entropy": 6.383328914642334,
      "epoch": 5.1708489054991995,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030333418958677525,
      "loss": 5.6554,
      "mean_token_accuracy": 0.17329463213682175,
      "num_tokens": 10604590.0,
      "step": 4845
    },
    {
      "entropy": 6.339723014831543,
      "epoch": 5.176187933796049,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0003029147431887183,
      "loss": 5.724,
      "mean_token_accuracy": 0.16704942733049394,
      "num_tokens": 10616702.0,
      "step": 4850
    },
    {
      "entropy": 6.435148477554321,
      "epoch": 5.181526962092899,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030249519823973546,
      "loss": 5.6996,
      "mean_token_accuracy": 0.16877569556236266,
      "num_tokens": 10627266.0,
      "step": 4855
    },
    {
      "entropy": 6.274662113189697,
      "epoch": 5.186865990389749,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030207555622099923,
      "loss": 5.6042,
      "mean_token_accuracy": 0.17321005761623381,
      "num_tokens": 10638052.0,
      "step": 4860
    },
    {
      "entropy": 6.367066812515259,
      "epoch": 5.192205018686599,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00030165581861402445,
      "loss": 5.689,
      "mean_token_accuracy": 0.1742100700736046,
      "num_tokens": 10649341.0,
      "step": 4865
    },
    {
      "entropy": 6.341131687164307,
      "epoch": 5.197544046983449,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00030123598690066363,
      "loss": 5.6132,
      "mean_token_accuracy": 0.17496493011713027,
      "num_tokens": 10660109.0,
      "step": 4870
    },
    {
      "entropy": 6.34923529624939,
      "epoch": 5.202883075280299,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00030081606256310127,
      "loss": 5.6668,
      "mean_token_accuracy": 0.1677444115281105,
      "num_tokens": 10670744.0,
      "step": 4875
    },
    {
      "entropy": 6.309982538223267,
      "epoch": 5.208222103577149,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030039604708384916,
      "loss": 5.6442,
      "mean_token_accuracy": 0.17178915590047836,
      "num_tokens": 10682191.0,
      "step": 4880
    },
    {
      "entropy": 6.340310096740723,
      "epoch": 5.213561131873999,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002999759419457406,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17591886818408967,
      "num_tokens": 10693776.0,
      "step": 4885
    },
    {
      "entropy": 6.352523374557495,
      "epoch": 5.218900160170849,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002995557486319257,
      "loss": 5.7681,
      "mean_token_accuracy": 0.16551783084869384,
      "num_tokens": 10706133.0,
      "step": 4890
    },
    {
      "entropy": 6.333683156967163,
      "epoch": 5.224239188467699,
      "grad_norm": 1.046875,
      "learning_rate": 0.00029913546862586567,
      "loss": 5.6941,
      "mean_token_accuracy": 0.16658874601125717,
      "num_tokens": 10717101.0,
      "step": 4895
    },
    {
      "entropy": 6.272641658782959,
      "epoch": 5.229578216764549,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029871510341132775,
      "loss": 5.6556,
      "mean_token_accuracy": 0.16933980137109755,
      "num_tokens": 10729084.0,
      "step": 4900
    },
    {
      "entropy": 6.375098752975464,
      "epoch": 5.234917245061399,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029829465447238015,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1725642204284668,
      "num_tokens": 10738222.0,
      "step": 4905
    },
    {
      "entropy": 6.346765041351318,
      "epoch": 5.2402562733582485,
      "grad_norm": 1.015625,
      "learning_rate": 0.00029787412329338657,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17934679388999938,
      "num_tokens": 10749660.0,
      "step": 4910
    },
    {
      "entropy": 6.3313685894012455,
      "epoch": 5.245595301655099,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029745351135900114,
      "loss": 5.6755,
      "mean_token_accuracy": 0.1738262429833412,
      "num_tokens": 10759674.0,
      "step": 4915
    },
    {
      "entropy": 6.31716775894165,
      "epoch": 5.250934329951948,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002970328201541629,
      "loss": 5.7311,
      "mean_token_accuracy": 0.16669611185789107,
      "num_tokens": 10771128.0,
      "step": 4920
    },
    {
      "entropy": 6.362551927566528,
      "epoch": 5.256273358248799,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029661205116409106,
      "loss": 5.6929,
      "mean_token_accuracy": 0.1685841992497444,
      "num_tokens": 10781892.0,
      "step": 4925
    },
    {
      "entropy": 6.381035041809082,
      "epoch": 5.261612386545648,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029619120587427906,
      "loss": 5.7166,
      "mean_token_accuracy": 0.16831838488578796,
      "num_tokens": 10793582.0,
      "step": 4930
    },
    {
      "entropy": 6.2977722644805905,
      "epoch": 5.266951414842499,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002957702857704901,
      "loss": 5.5889,
      "mean_token_accuracy": 0.18098255395889282,
      "num_tokens": 10803789.0,
      "step": 4935
    },
    {
      "entropy": 6.347255229949951,
      "epoch": 5.272290443139348,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029534929233875117,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1723804607987404,
      "num_tokens": 10815087.0,
      "step": 4940
    },
    {
      "entropy": 6.349794673919678,
      "epoch": 5.277629471436199,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002949282270653483,
      "loss": 5.6737,
      "mean_token_accuracy": 0.17483874708414077,
      "num_tokens": 10825885.0,
      "step": 4945
    },
    {
      "entropy": 6.404815149307251,
      "epoch": 5.282968499733048,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002945070914368212,
      "loss": 5.7014,
      "mean_token_accuracy": 0.16592416912317276,
      "num_tokens": 10835956.0,
      "step": 4950
    },
    {
      "entropy": 6.379074764251709,
      "epoch": 5.288307528029899,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002940858869399579,
      "loss": 5.6806,
      "mean_token_accuracy": 0.1729910969734192,
      "num_tokens": 10846460.0,
      "step": 4955
    },
    {
      "entropy": 6.343271350860595,
      "epoch": 5.293646556326749,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029366461506178954,
      "loss": 5.7067,
      "mean_token_accuracy": 0.16899541318416594,
      "num_tokens": 10857021.0,
      "step": 4960
    },
    {
      "entropy": 6.343428754806519,
      "epoch": 5.2989855846235985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002932432772895851,
      "loss": 5.7336,
      "mean_token_accuracy": 0.17285263240337373,
      "num_tokens": 10866283.0,
      "step": 4965
    },
    {
      "entropy": 6.278393840789795,
      "epoch": 5.304324612920449,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029282187511084644,
      "loss": 5.5869,
      "mean_token_accuracy": 0.18448489904403687,
      "num_tokens": 10876993.0,
      "step": 4970
    },
    {
      "entropy": 6.378061532974243,
      "epoch": 5.309663641217298,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00029240041001330256,
      "loss": 5.6984,
      "mean_token_accuracy": 0.1693331405520439,
      "num_tokens": 10887981.0,
      "step": 4975
    },
    {
      "entropy": 6.280534744262695,
      "epoch": 5.315002669514149,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029197888348490446,
      "loss": 5.6122,
      "mean_token_accuracy": 0.1767708420753479,
      "num_tokens": 10898889.0,
      "step": 4980
    },
    {
      "entropy": 6.317689323425293,
      "epoch": 5.320341697810998,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029155729701382045,
      "loss": 5.719,
      "mean_token_accuracy": 0.1729544296860695,
      "num_tokens": 10909982.0,
      "step": 4985
    },
    {
      "entropy": 6.370178127288819,
      "epoch": 5.325680726107849,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00029113565208843017,
      "loss": 5.7206,
      "mean_token_accuracy": 0.1736098438501358,
      "num_tokens": 10921245.0,
      "step": 4990
    },
    {
      "entropy": 6.353579378128051,
      "epoch": 5.331019754404698,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002907139501973196,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17264861315488816,
      "num_tokens": 10931660.0,
      "step": 4995
    },
    {
      "entropy": 6.374445533752441,
      "epoch": 5.336358782701549,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029029219282927603,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17012456506490709,
      "num_tokens": 10943534.0,
      "step": 5000
    },
    {
      "epoch": 5.336358782701549,
      "eval_entropy": 6.1904587758051886,
      "eval_loss": 6.577048301696777,
      "eval_mean_token_accuracy": 0.13215796733637908,
      "eval_num_tokens": 10943534.0,
      "eval_runtime": 1.8138,
      "eval_samples_per_second": 1698.066,
      "eval_steps_per_second": 212.258,
      "step": 5000
    },
    {
      "entropy": 6.345319509506226,
      "epoch": 5.341697810998398,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002898703814732824,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1756830707192421,
      "num_tokens": 10954519.0,
      "step": 5005
    },
    {
      "entropy": 6.229328632354736,
      "epoch": 5.3470368392952485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002894485176185124,
      "loss": 5.6316,
      "mean_token_accuracy": 0.1782810002565384,
      "num_tokens": 10966236.0,
      "step": 5010
    },
    {
      "entropy": 6.332392406463623,
      "epoch": 5.352375867592098,
      "grad_norm": 0.9375,
      "learning_rate": 0.00028902660275432503,
      "loss": 5.6885,
      "mean_token_accuracy": 0.16952528804540634,
      "num_tokens": 10978070.0,
      "step": 5015
    },
    {
      "entropy": 6.3167695045471195,
      "epoch": 5.3577148958889484,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028860463837025935,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1742299184203148,
      "num_tokens": 10988651.0,
      "step": 5020
    },
    {
      "entropy": 6.326428937911987,
      "epoch": 5.363053924185798,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002881826259560292,
      "loss": 5.7326,
      "mean_token_accuracy": 0.16955900341272354,
      "num_tokens": 11001226.0,
      "step": 5025
    },
    {
      "entropy": 6.35105037689209,
      "epoch": 5.368392952482648,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000287760567001518,
      "loss": 5.616,
      "mean_token_accuracy": 0.17740404307842256,
      "num_tokens": 11012025.0,
      "step": 5030
    },
    {
      "entropy": 6.346265268325806,
      "epoch": 5.373731980779498,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002873384629967738,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1710437223315239,
      "num_tokens": 11023202.0,
      "step": 5035
    },
    {
      "entropy": 6.345341157913208,
      "epoch": 5.379071009076348,
      "grad_norm": 1.0625,
      "learning_rate": 0.000286916315432003,
      "loss": 5.6687,
      "mean_token_accuracy": 0.172770456969738,
      "num_tokens": 11034003.0,
      "step": 5040
    },
    {
      "entropy": 6.375285863876343,
      "epoch": 5.384410037373198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028649412579756647,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1692378044128418,
      "num_tokens": 11046083.0,
      "step": 5045
    },
    {
      "entropy": 6.409449243545533,
      "epoch": 5.389749065670048,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028607189558397313,
      "loss": 5.7742,
      "mean_token_accuracy": 0.1709982544183731,
      "num_tokens": 11057378.0,
      "step": 5050
    },
    {
      "entropy": 6.381783294677734,
      "epoch": 5.395088093966898,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002856496262818755,
      "loss": 5.7099,
      "mean_token_accuracy": 0.17474423944950104,
      "num_tokens": 11068389.0,
      "step": 5055
    },
    {
      "entropy": 6.341726016998291,
      "epoch": 5.400427122263748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002852273193820641,
      "loss": 5.7448,
      "mean_token_accuracy": 0.1749459683895111,
      "num_tokens": 11079699.0,
      "step": 5060
    },
    {
      "entropy": 6.379783010482788,
      "epoch": 5.405766150560598,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002848049763754617,
      "loss": 5.7553,
      "mean_token_accuracy": 0.16467145085334778,
      "num_tokens": 11090245.0,
      "step": 5065
    },
    {
      "entropy": 6.295734596252442,
      "epoch": 5.411105178857448,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002843825987531192,
      "loss": 5.724,
      "mean_token_accuracy": 0.1723369762301445,
      "num_tokens": 11101810.0,
      "step": 5070
    },
    {
      "entropy": 6.336012506484986,
      "epoch": 5.4164442071542975,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028396018800620927,
      "loss": 5.6506,
      "mean_token_accuracy": 0.18089286237955093,
      "num_tokens": 11111849.0,
      "step": 5075
    },
    {
      "entropy": 6.281029510498047,
      "epoch": 5.421783235451148,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028353774562602164,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1751598000526428,
      "num_tokens": 11122920.0,
      "step": 5080
    },
    {
      "entropy": 6.314212656021118,
      "epoch": 5.427122263747998,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028311527310395776,
      "loss": 5.6814,
      "mean_token_accuracy": 0.17217377573251724,
      "num_tokens": 11133227.0,
      "step": 5085
    },
    {
      "entropy": 6.2608273983001705,
      "epoch": 5.432461292044848,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002826927719315254,
      "loss": 5.657,
      "mean_token_accuracy": 0.16982914507389069,
      "num_tokens": 11145206.0,
      "step": 5090
    },
    {
      "entropy": 6.364982986450196,
      "epoch": 5.437800320341697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028227024360033353,
      "loss": 5.6904,
      "mean_token_accuracy": 0.16514935940504075,
      "num_tokens": 11155739.0,
      "step": 5095
    },
    {
      "entropy": 6.329564189910888,
      "epoch": 5.443139348638548,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0002818476896020872,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17437167465686798,
      "num_tokens": 11166769.0,
      "step": 5100
    },
    {
      "entropy": 6.363117694854736,
      "epoch": 5.448478376935398,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002814251114285816,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1687398821115494,
      "num_tokens": 11177398.0,
      "step": 5105
    },
    {
      "entropy": 6.315731954574585,
      "epoch": 5.453817405232248,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002810025105716979,
      "loss": 5.6592,
      "mean_token_accuracy": 0.1719701200723648,
      "num_tokens": 11188576.0,
      "step": 5110
    },
    {
      "entropy": 6.334354066848755,
      "epoch": 5.459156433529098,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028057988852339677,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17892009466886521,
      "num_tokens": 11198761.0,
      "step": 5115
    },
    {
      "entropy": 6.304895687103271,
      "epoch": 5.464495461825948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000280157246775714,
      "loss": 5.6424,
      "mean_token_accuracy": 0.17465301007032394,
      "num_tokens": 11208584.0,
      "step": 5120
    },
    {
      "entropy": 6.3745725631713865,
      "epoch": 5.469834490122798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002797345868207549,
      "loss": 5.7293,
      "mean_token_accuracy": 0.1720472276210785,
      "num_tokens": 11219112.0,
      "step": 5125
    },
    {
      "entropy": 6.357462358474732,
      "epoch": 5.4751735184196475,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002793119101506889,
      "loss": 5.6811,
      "mean_token_accuracy": 0.1729859247803688,
      "num_tokens": 11229803.0,
      "step": 5130
    },
    {
      "entropy": 6.339340782165527,
      "epoch": 5.480512546716498,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002788892182577447,
      "loss": 5.6658,
      "mean_token_accuracy": 0.17343320995569228,
      "num_tokens": 11240834.0,
      "step": 5135
    },
    {
      "entropy": 6.286582279205322,
      "epoch": 5.485851575013347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00027846651263420456,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1690528392791748,
      "num_tokens": 11251357.0,
      "step": 5140
    },
    {
      "entropy": 6.346102952957153,
      "epoch": 5.491190603310198,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027804379477239927,
      "loss": 5.709,
      "mean_token_accuracy": 0.17001670449972153,
      "num_tokens": 11262187.0,
      "step": 5145
    },
    {
      "entropy": 6.346243953704834,
      "epoch": 5.496529631607047,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027762106616470276,
      "loss": 5.7154,
      "mean_token_accuracy": 0.16876792460680007,
      "num_tokens": 11272711.0,
      "step": 5150
    },
    {
      "entropy": 6.369475269317627,
      "epoch": 5.501868659903898,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000277198328303527,
      "loss": 5.6705,
      "mean_token_accuracy": 0.17209891378879547,
      "num_tokens": 11283022.0,
      "step": 5155
    },
    {
      "entropy": 6.379570722579956,
      "epoch": 5.507207688200747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027677558268131655,
      "loss": 5.7516,
      "mean_token_accuracy": 0.16407181173563004,
      "num_tokens": 11293469.0,
      "step": 5160
    },
    {
      "entropy": 6.325233125686646,
      "epoch": 5.512546716497598,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002763528307905435,
      "loss": 5.6836,
      "mean_token_accuracy": 0.1714015170931816,
      "num_tokens": 11304555.0,
      "step": 5165
    },
    {
      "entropy": 6.3725221157073975,
      "epoch": 5.517885744794447,
      "grad_norm": 1.03125,
      "learning_rate": 0.00027593007412370194,
      "loss": 5.76,
      "mean_token_accuracy": 0.1653032273054123,
      "num_tokens": 11316089.0,
      "step": 5170
    },
    {
      "entropy": 6.396354389190674,
      "epoch": 5.523224773091298,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002755073141733027,
      "loss": 5.7527,
      "mean_token_accuracy": 0.1662668615579605,
      "num_tokens": 11327636.0,
      "step": 5175
    },
    {
      "entropy": 6.385932922363281,
      "epoch": 5.528563801388147,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002750845524318686,
      "loss": 5.7326,
      "mean_token_accuracy": 0.171891513466835,
      "num_tokens": 11338178.0,
      "step": 5180
    },
    {
      "entropy": 6.30861268043518,
      "epoch": 5.5339028296849975,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027466179039192833,
      "loss": 5.6418,
      "mean_token_accuracy": 0.17382311820983887,
      "num_tokens": 11349138.0,
      "step": 5185
    },
    {
      "entropy": 6.272126007080078,
      "epoch": 5.539241857981847,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000274239029546012,
      "loss": 5.6875,
      "mean_token_accuracy": 0.17363980263471604,
      "num_tokens": 11359926.0,
      "step": 5190
    },
    {
      "entropy": 6.331406784057617,
      "epoch": 5.544580886278697,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002738162713866453,
      "loss": 5.6715,
      "mean_token_accuracy": 0.1712009072303772,
      "num_tokens": 11371235.0,
      "step": 5195
    },
    {
      "entropy": 6.3823121070861815,
      "epoch": 5.549919914575547,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027339351740634436,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17318181097507476,
      "num_tokens": 11381545.0,
      "step": 5200
    },
    {
      "entropy": 6.321549987792968,
      "epoch": 5.555258942872397,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000272970769097611,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1728147089481354,
      "num_tokens": 11392785.0,
      "step": 5205
    },
    {
      "entropy": 6.35057806968689,
      "epoch": 5.560597971169247,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027254802795292633,
      "loss": 5.7731,
      "mean_token_accuracy": 0.16422995030879975,
      "num_tokens": 11403263.0,
      "step": 5210
    },
    {
      "entropy": 6.347456312179565,
      "epoch": 5.565936999466097,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00027212529546474676,
      "loss": 5.7354,
      "mean_token_accuracy": 0.16245418637990952,
      "num_tokens": 11414543.0,
      "step": 5215
    },
    {
      "entropy": 6.324865484237671,
      "epoch": 5.571276027762947,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002717025731254978,
      "loss": 5.6225,
      "mean_token_accuracy": 0.18328625112771987,
      "num_tokens": 11424942.0,
      "step": 5220
    },
    {
      "entropy": 6.362385845184326,
      "epoch": 5.576615056059797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027127986242756933,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17098136097192765,
      "num_tokens": 11435537.0,
      "step": 5225
    },
    {
      "entropy": 6.340441131591797,
      "epoch": 5.5819540843566475,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027085716486331007,
      "loss": 5.6594,
      "mean_token_accuracy": 0.17434974610805512,
      "num_tokens": 11446465.0,
      "step": 5230
    },
    {
      "entropy": 6.375800085067749,
      "epoch": 5.587293112653497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027043448192502225,
      "loss": 5.728,
      "mean_token_accuracy": 0.17118730247020722,
      "num_tokens": 11457618.0,
      "step": 5235
    },
    {
      "entropy": 6.3312904834747314,
      "epoch": 5.5926321409503466,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027001181510495664,
      "loss": 5.6882,
      "mean_token_accuracy": 0.17196964472532272,
      "num_tokens": 11468586.0,
      "step": 5240
    },
    {
      "entropy": 6.303575944900513,
      "epoch": 5.597971169247197,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026958916589530697,
      "loss": 5.6982,
      "mean_token_accuracy": 0.16687122881412506,
      "num_tokens": 11481160.0,
      "step": 5245
    },
    {
      "entropy": 6.36999282836914,
      "epoch": 5.603310197544047,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026916653578820486,
      "loss": 5.7427,
      "mean_token_accuracy": 0.17406813949346542,
      "num_tokens": 11493490.0,
      "step": 5250
    },
    {
      "entropy": 6.353348350524902,
      "epoch": 5.608649225840897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026874392627571455,
      "loss": 5.6679,
      "mean_token_accuracy": 0.1726611539721489,
      "num_tokens": 11505194.0,
      "step": 5255
    },
    {
      "entropy": 6.292462539672852,
      "epoch": 5.613988254137747,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002683213388498274,
      "loss": 5.6491,
      "mean_token_accuracy": 0.18231137245893478,
      "num_tokens": 11515710.0,
      "step": 5260
    },
    {
      "entropy": 6.312755346298218,
      "epoch": 5.619327282434597,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026789877500245704,
      "loss": 5.6932,
      "mean_token_accuracy": 0.17089592665433884,
      "num_tokens": 11525599.0,
      "step": 5265
    },
    {
      "entropy": 6.346545267105102,
      "epoch": 5.624666310731447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002674762362254337,
      "loss": 5.6728,
      "mean_token_accuracy": 0.16902205348014832,
      "num_tokens": 11536308.0,
      "step": 5270
    },
    {
      "entropy": 6.343692779541016,
      "epoch": 5.630005339028297,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026705372401049897,
      "loss": 5.6705,
      "mean_token_accuracy": 0.17369551062583924,
      "num_tokens": 11546530.0,
      "step": 5275
    },
    {
      "entropy": 6.331171655654908,
      "epoch": 5.635344367325147,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000266631239849301,
      "loss": 5.7249,
      "mean_token_accuracy": 0.17059761583805083,
      "num_tokens": 11558540.0,
      "step": 5280
    },
    {
      "entropy": 6.311120510101318,
      "epoch": 5.640683395621997,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002662087852333886,
      "loss": 5.7066,
      "mean_token_accuracy": 0.1674735203385353,
      "num_tokens": 11569750.0,
      "step": 5285
    },
    {
      "entropy": 6.379989147186279,
      "epoch": 5.646022423918847,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00026578636165420637,
      "loss": 5.7141,
      "mean_token_accuracy": 0.16615591198205948,
      "num_tokens": 11580773.0,
      "step": 5290
    },
    {
      "entropy": 6.374867630004883,
      "epoch": 5.6513614522156965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026536397060308945,
      "loss": 5.7459,
      "mean_token_accuracy": 0.1622788906097412,
      "num_tokens": 11592087.0,
      "step": 5295
    },
    {
      "entropy": 6.348657894134521,
      "epoch": 5.656700480512547,
      "grad_norm": 1.109375,
      "learning_rate": 0.000264941613571258,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1680184170603752,
      "num_tokens": 11603234.0,
      "step": 5300
    },
    {
      "entropy": 6.328213453292847,
      "epoch": 5.662039508809396,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002645192920498122,
      "loss": 5.6764,
      "mean_token_accuracy": 0.17510985732078552,
      "num_tokens": 11613849.0,
      "step": 5305
    },
    {
      "entropy": 6.363373613357544,
      "epoch": 5.667378537106247,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026409700752972665,
      "loss": 5.6916,
      "mean_token_accuracy": 0.17176212668418883,
      "num_tokens": 11623972.0,
      "step": 5310
    },
    {
      "entropy": 6.272594881057739,
      "epoch": 5.672717565403096,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026367476150184556,
      "loss": 5.6319,
      "mean_token_accuracy": 0.1784251555800438,
      "num_tokens": 11634701.0,
      "step": 5315
    },
    {
      "entropy": 6.238541460037231,
      "epoch": 5.678056593699947,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026325255545687706,
      "loss": 5.6208,
      "mean_token_accuracy": 0.17861277014017105,
      "num_tokens": 11644138.0,
      "step": 5320
    },
    {
      "entropy": 6.30879054069519,
      "epoch": 5.683395621996796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002628303908853882,
      "loss": 5.6758,
      "mean_token_accuracy": 0.1733354464173317,
      "num_tokens": 11655807.0,
      "step": 5325
    },
    {
      "entropy": 6.308476877212525,
      "epoch": 5.688734650293647,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026240826927779976,
      "loss": 5.6959,
      "mean_token_accuracy": 0.16817405223846435,
      "num_tokens": 11667111.0,
      "step": 5330
    },
    {
      "entropy": 6.402845287322998,
      "epoch": 5.694073678590496,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0002619861921243806,
      "loss": 5.7185,
      "mean_token_accuracy": 0.17394309788942336,
      "num_tokens": 11678205.0,
      "step": 5335
    },
    {
      "entropy": 6.3389698505401615,
      "epoch": 5.699412706887347,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026156416091524276,
      "loss": 5.673,
      "mean_token_accuracy": 0.17586482614278792,
      "num_tokens": 11688334.0,
      "step": 5340
    },
    {
      "entropy": 6.3289553165435795,
      "epoch": 5.704751735184196,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000261142177140336,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17235073149204255,
      "num_tokens": 11699395.0,
      "step": 5345
    },
    {
      "entropy": 6.266600704193115,
      "epoch": 5.7100907634810465,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026072024228944267,
      "loss": 5.6061,
      "mean_token_accuracy": 0.18524847328662872,
      "num_tokens": 11710454.0,
      "step": 5350
    },
    {
      "entropy": 6.344654512405396,
      "epoch": 5.715429791777897,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002602983578521725,
      "loss": 5.6318,
      "mean_token_accuracy": 0.17262564450502396,
      "num_tokens": 11721980.0,
      "step": 5355
    },
    {
      "entropy": 6.333229446411133,
      "epoch": 5.720768820074746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025987652531795697,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1802364245057106,
      "num_tokens": 11733826.0,
      "step": 5360
    },
    {
      "entropy": 6.3234271049499515,
      "epoch": 5.726107848371596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025945474617604456,
      "loss": 5.7411,
      "mean_token_accuracy": 0.17659412920475007,
      "num_tokens": 11744729.0,
      "step": 5365
    },
    {
      "entropy": 6.2692749977111815,
      "epoch": 5.731446876668446,
      "grad_norm": 1.125,
      "learning_rate": 0.0002590330219154954,
      "loss": 5.6853,
      "mean_token_accuracy": 0.17409733533859253,
      "num_tokens": 11755934.0,
      "step": 5370
    },
    {
      "entropy": 6.382311153411865,
      "epoch": 5.736785904965297,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002586113540251753,
      "loss": 5.7085,
      "mean_token_accuracy": 0.178350730240345,
      "num_tokens": 11766501.0,
      "step": 5375
    },
    {
      "entropy": 6.392430973052979,
      "epoch": 5.742124933262146,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002581897439937515,
      "loss": 5.7553,
      "mean_token_accuracy": 0.16536930203437805,
      "num_tokens": 11777731.0,
      "step": 5380
    },
    {
      "entropy": 6.312212419509888,
      "epoch": 5.747463961558997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002577681933096869,
      "loss": 5.7048,
      "mean_token_accuracy": 0.17177869379520416,
      "num_tokens": 11788061.0,
      "step": 5385
    },
    {
      "entropy": 6.266093730926514,
      "epoch": 5.752802989855846,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025734670346123494,
      "loss": 5.5841,
      "mean_token_accuracy": 0.177488249540329,
      "num_tokens": 11797783.0,
      "step": 5390
    },
    {
      "entropy": 6.325652122497559,
      "epoch": 5.7581420181526966,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00025692527593643405,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16489865630865097,
      "num_tokens": 11809112.0,
      "step": 5395
    },
    {
      "entropy": 6.369477891921997,
      "epoch": 5.763481046449546,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025650391222310293,
      "loss": 5.6616,
      "mean_token_accuracy": 0.16763915121555328,
      "num_tokens": 11819804.0,
      "step": 5400
    },
    {
      "entropy": 6.344639253616333,
      "epoch": 5.7688200747463965,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002560826138088347,
      "loss": 5.7006,
      "mean_token_accuracy": 0.17343638390302657,
      "num_tokens": 11831397.0,
      "step": 5405
    },
    {
      "entropy": 6.286445760726929,
      "epoch": 5.774159103043246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025566138218099217,
      "loss": 5.6198,
      "mean_token_accuracy": 0.17345215380191803,
      "num_tokens": 11842395.0,
      "step": 5410
    },
    {
      "entropy": 6.297845935821533,
      "epoch": 5.779498131340096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002552402188267023,
      "loss": 5.7067,
      "mean_token_accuracy": 0.17351611852645873,
      "num_tokens": 11853314.0,
      "step": 5415
    },
    {
      "entropy": 6.262566709518433,
      "epoch": 5.784837159636946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025481912523285093,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17077697217464446,
      "num_tokens": 11863836.0,
      "step": 5420
    },
    {
      "entropy": 6.265786457061767,
      "epoch": 5.790176187933796,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025439810288607783,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1804139643907547,
      "num_tokens": 11875198.0,
      "step": 5425
    },
    {
      "entropy": 6.317636013031006,
      "epoch": 5.795515216230646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025397715327277095,
      "loss": 5.7225,
      "mean_token_accuracy": 0.1702480509877205,
      "num_tokens": 11885653.0,
      "step": 5430
    },
    {
      "entropy": 6.372118854522705,
      "epoch": 5.800854244527496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025355627787906177,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17125837802886962,
      "num_tokens": 11896393.0,
      "step": 5435
    },
    {
      "entropy": 6.249027299880981,
      "epoch": 5.806193272824346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002531354781908194,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17184996008872985,
      "num_tokens": 11907935.0,
      "step": 5440
    },
    {
      "entropy": 6.373139715194702,
      "epoch": 5.811532301121196,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00025271475569364586,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16964334696531297,
      "num_tokens": 11918388.0,
      "step": 5445
    },
    {
      "entropy": 6.3742577075958256,
      "epoch": 5.816871329418046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002522941118728709,
      "loss": 5.6929,
      "mean_token_accuracy": 0.1735077306628227,
      "num_tokens": 11928438.0,
      "step": 5450
    },
    {
      "entropy": 6.368045902252197,
      "epoch": 5.822210357714896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025187354821354596,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16163920015096664,
      "num_tokens": 11939344.0,
      "step": 5455
    },
    {
      "entropy": 6.306874752044678,
      "epoch": 5.8275493860117455,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025145306620043994,
      "loss": 5.6972,
      "mean_token_accuracy": 0.16952828019857408,
      "num_tokens": 11950968.0,
      "step": 5460
    },
    {
      "entropy": 6.356746006011963,
      "epoch": 5.832888414308596,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002510326673180333,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17609124928712844,
      "num_tokens": 11962832.0,
      "step": 5465
    },
    {
      "entropy": 6.325391578674316,
      "epoch": 5.8382274426054455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002506123530505132,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17825517505407334,
      "num_tokens": 11972528.0,
      "step": 5470
    },
    {
      "entropy": 6.346975374221802,
      "epoch": 5.843566470902296,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00025019212488176757,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1736564815044403,
      "num_tokens": 11983786.0,
      "step": 5475
    },
    {
      "entropy": 6.294334602355957,
      "epoch": 5.848905499199145,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024977198429538096,
      "loss": 5.6388,
      "mean_token_accuracy": 0.16715028882026672,
      "num_tokens": 11995299.0,
      "step": 5480
    },
    {
      "entropy": 6.325953817367553,
      "epoch": 5.854244527495996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024935193277462835,
      "loss": 5.6638,
      "mean_token_accuracy": 0.17183291912078857,
      "num_tokens": 12006846.0,
      "step": 5485
    },
    {
      "entropy": 6.31584792137146,
      "epoch": 5.859583555792845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024893197180247056,
      "loss": 5.5542,
      "mean_token_accuracy": 0.1834078386425972,
      "num_tokens": 12016937.0,
      "step": 5490
    },
    {
      "entropy": 6.333651113510132,
      "epoch": 5.864922584089696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024851210286154854,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17022206485271454,
      "num_tokens": 12026860.0,
      "step": 5495
    },
    {
      "entropy": 6.316498041152954,
      "epoch": 5.870261612386546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002480923274341783,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1723729819059372,
      "num_tokens": 12036169.0,
      "step": 5500
    },
    {
      "epoch": 5.870261612386546,
      "eval_entropy": 6.132839763938606,
      "eval_loss": 6.555144786834717,
      "eval_mean_token_accuracy": 0.13326003254427538,
      "eval_num_tokens": 12036169.0,
      "eval_runtime": 1.8244,
      "eval_samples_per_second": 1688.262,
      "eval_steps_per_second": 211.033,
      "step": 5500
    },
    {
      "entropy": 6.314624404907226,
      "epoch": 5.875600640683396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002476726470023459,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17477745413780213,
      "num_tokens": 12047086.0,
      "step": 5505
    },
    {
      "entropy": 6.289700126647949,
      "epoch": 5.880939668980245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002472530630477017,
      "loss": 5.6497,
      "mean_token_accuracy": 0.17475370615720748,
      "num_tokens": 12057407.0,
      "step": 5510
    },
    {
      "entropy": 6.396420860290528,
      "epoch": 5.8862786972770955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002468335770515559,
      "loss": 5.7059,
      "mean_token_accuracy": 0.16928656101226808,
      "num_tokens": 12068028.0,
      "step": 5515
    },
    {
      "entropy": 6.312612390518188,
      "epoch": 5.891617725573946,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024641419049487246,
      "loss": 5.6024,
      "mean_token_accuracy": 0.17720211148262024,
      "num_tokens": 12078362.0,
      "step": 5520
    },
    {
      "entropy": 6.308467864990234,
      "epoch": 5.896956753870795,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00024599490485826446,
      "loss": 5.6803,
      "mean_token_accuracy": 0.17163989543914795,
      "num_tokens": 12089854.0,
      "step": 5525
    },
    {
      "entropy": 6.3653336524963375,
      "epoch": 5.902295782167646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024557572162198883,
      "loss": 5.7196,
      "mean_token_accuracy": 0.1672852575778961,
      "num_tokens": 12100886.0,
      "step": 5530
    },
    {
      "entropy": 6.348902845382691,
      "epoch": 5.907634810464495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002451566422659407,
      "loss": 5.7,
      "mean_token_accuracy": 0.17318319976329805,
      "num_tokens": 12112747.0,
      "step": 5535
    },
    {
      "entropy": 6.347790002822876,
      "epoch": 5.912973838761346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002447376682696485,
      "loss": 5.7365,
      "mean_token_accuracy": 0.16582491248846054,
      "num_tokens": 12123730.0,
      "step": 5540
    },
    {
      "entropy": 6.374050569534302,
      "epoch": 5.918312867058195,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002443188011122689,
      "loss": 5.8037,
      "mean_token_accuracy": 0.16277724206447602,
      "num_tokens": 12135870.0,
      "step": 5545
    },
    {
      "entropy": 6.372762632369995,
      "epoch": 5.923651895355046,
      "grad_norm": 1.0625,
      "learning_rate": 0.00024390004227258123,
      "loss": 5.6862,
      "mean_token_accuracy": 0.18058872371912002,
      "num_tokens": 12146826.0,
      "step": 5550
    },
    {
      "entropy": 6.291637754440307,
      "epoch": 5.928990923651895,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002434813932289825,
      "loss": 5.7145,
      "mean_token_accuracy": 0.16549645662307738,
      "num_tokens": 12159586.0,
      "step": 5555
    },
    {
      "entropy": 6.307714939117432,
      "epoch": 5.934329951948746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024306285545948192,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1726786732673645,
      "num_tokens": 12170162.0,
      "step": 5560
    },
    {
      "entropy": 6.381466817855835,
      "epoch": 5.939668980245595,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00024264443044169615,
      "loss": 5.738,
      "mean_token_accuracy": 0.1712544173002243,
      "num_tokens": 12181208.0,
      "step": 5565
    },
    {
      "entropy": 6.326228046417237,
      "epoch": 5.9450080085424455,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024222611965284341,
      "loss": 5.7158,
      "mean_token_accuracy": 0.17157405018806457,
      "num_tokens": 12192438.0,
      "step": 5570
    },
    {
      "entropy": 6.284603929519653,
      "epoch": 5.950347036839295,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024180792456973894,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17318402975797653,
      "num_tokens": 12203863.0,
      "step": 5575
    },
    {
      "entropy": 6.314238405227661,
      "epoch": 5.955686065136145,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002413898466687894,
      "loss": 5.6864,
      "mean_token_accuracy": 0.1683388128876686,
      "num_tokens": 12213561.0,
      "step": 5580
    },
    {
      "entropy": 6.3632911205291744,
      "epoch": 5.961025093432995,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024097188742598768,
      "loss": 5.7269,
      "mean_token_accuracy": 0.16977357864379883,
      "num_tokens": 12224140.0,
      "step": 5585
    },
    {
      "entropy": 6.374677467346191,
      "epoch": 5.966364121729845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002405540483169079,
      "loss": 5.6601,
      "mean_token_accuracy": 0.17816300988197326,
      "num_tokens": 12235654.0,
      "step": 5590
    },
    {
      "entropy": 6.33419132232666,
      "epoch": 5.971703150026695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024013633081669994,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1661536157131195,
      "num_tokens": 12247255.0,
      "step": 5595
    },
    {
      "entropy": 6.31337947845459,
      "epoch": 5.977042178323545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002397187364000844,
      "loss": 5.6823,
      "mean_token_accuracy": 0.17041108906269073,
      "num_tokens": 12258651.0,
      "step": 5600
    },
    {
      "entropy": 6.351841115951538,
      "epoch": 5.982381206620395,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023930126654134732,
      "loss": 5.6295,
      "mean_token_accuracy": 0.1734935835003853,
      "num_tokens": 12269005.0,
      "step": 5605
    },
    {
      "entropy": 6.288322877883911,
      "epoch": 5.987720234917245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002388839227143349,
      "loss": 5.7215,
      "mean_token_accuracy": 0.16935435086488723,
      "num_tokens": 12279727.0,
      "step": 5610
    },
    {
      "entropy": 6.2979058742523195,
      "epoch": 5.993059263214095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00023846670639244866,
      "loss": 5.6719,
      "mean_token_accuracy": 0.17470595985651016,
      "num_tokens": 12290765.0,
      "step": 5615
    },
    {
      "entropy": 6.295067024230957,
      "epoch": 5.998398291510945,
      "grad_norm": 1.125,
      "learning_rate": 0.00023804961904863969,
      "loss": 5.7007,
      "mean_token_accuracy": 0.17165803164243698,
      "num_tokens": 12303014.0,
      "step": 5620
    },
    {
      "entropy": 6.364009963141547,
      "epoch": 6.00320341697811,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002376326621554039,
      "loss": 5.6442,
      "mean_token_accuracy": 0.17100713153680167,
      "num_tokens": 12312009.0,
      "step": 5625
    },
    {
      "entropy": 6.327949523925781,
      "epoch": 6.00854244527496,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00023721583718477673,
      "loss": 5.52,
      "mean_token_accuracy": 0.18202630132436753,
      "num_tokens": 12323417.0,
      "step": 5630
    },
    {
      "entropy": 6.3348822593688965,
      "epoch": 6.01388147357181,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002367991456083275,
      "loss": 5.5289,
      "mean_token_accuracy": 0.19252616167068481,
      "num_tokens": 12333611.0,
      "step": 5635
    },
    {
      "entropy": 6.343252229690552,
      "epoch": 6.01922050186866,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000236382588897155,
      "loss": 5.5416,
      "mean_token_accuracy": 0.17564734518527986,
      "num_tokens": 12343790.0,
      "step": 5640
    },
    {
      "entropy": 6.236243009567261,
      "epoch": 6.02455953016551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023596616852188164,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1880975544452667,
      "num_tokens": 12353667.0,
      "step": 5645
    },
    {
      "entropy": 6.301973009109497,
      "epoch": 6.02989855846236,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002355498859526487,
      "loss": 5.4312,
      "mean_token_accuracy": 0.19881273210048675,
      "num_tokens": 12363582.0,
      "step": 5650
    },
    {
      "entropy": 6.354191589355469,
      "epoch": 6.03523758675921,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023513374265911086,
      "loss": 5.5048,
      "mean_token_accuracy": 0.18463512510061264,
      "num_tokens": 12373821.0,
      "step": 5655
    },
    {
      "entropy": 6.3050848007202145,
      "epoch": 6.04057661505606,
      "grad_norm": 1.125,
      "learning_rate": 0.000234717740110431,
      "loss": 5.5051,
      "mean_token_accuracy": 0.18619876503944396,
      "num_tokens": 12385056.0,
      "step": 5660
    },
    {
      "entropy": 6.331363725662231,
      "epoch": 6.0459156433529095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023430187977527533,
      "loss": 5.5146,
      "mean_token_accuracy": 0.18862221390008926,
      "num_tokens": 12396694.0,
      "step": 5665
    },
    {
      "entropy": 6.216295480728149,
      "epoch": 6.05125467164976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023388616312180766,
      "loss": 5.4623,
      "mean_token_accuracy": 0.18716894537210466,
      "num_tokens": 12408688.0,
      "step": 5670
    },
    {
      "entropy": 6.360234975814819,
      "epoch": 6.0565936999466095,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002334705916176848,
      "loss": 5.5172,
      "mean_token_accuracy": 0.18779598027467728,
      "num_tokens": 12419789.0,
      "step": 5675
    },
    {
      "entropy": 6.36201434135437,
      "epoch": 6.06193272824346,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023305516673005112,
      "loss": 5.5486,
      "mean_token_accuracy": 0.18714220821857452,
      "num_tokens": 12431380.0,
      "step": 5680
    },
    {
      "entropy": 6.357383584976196,
      "epoch": 6.067271756540309,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002326398899255332,
      "loss": 5.4941,
      "mean_token_accuracy": 0.18233438730239868,
      "num_tokens": 12442314.0,
      "step": 5685
    },
    {
      "entropy": 6.314443349838257,
      "epoch": 6.07261078483716,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023222476267023508,
      "loss": 5.6087,
      "mean_token_accuracy": 0.17443327456712723,
      "num_tokens": 12454371.0,
      "step": 5690
    },
    {
      "entropy": 6.342440605163574,
      "epoch": 6.077949813134009,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023180978642973248,
      "loss": 5.525,
      "mean_token_accuracy": 0.18542657196521758,
      "num_tokens": 12465923.0,
      "step": 5695
    },
    {
      "entropy": 6.264341878890991,
      "epoch": 6.08328884143086,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023139496266906835,
      "loss": 5.4366,
      "mean_token_accuracy": 0.192969910800457,
      "num_tokens": 12476152.0,
      "step": 5700
    },
    {
      "entropy": 6.230353879928589,
      "epoch": 6.088627869727709,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023098029285274703,
      "loss": 5.4205,
      "mean_token_accuracy": 0.19613498598337173,
      "num_tokens": 12487546.0,
      "step": 5705
    },
    {
      "entropy": 6.316088533401489,
      "epoch": 6.09396689802456,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023056577844472948,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18258665949106218,
      "num_tokens": 12498851.0,
      "step": 5710
    },
    {
      "entropy": 6.330278158187866,
      "epoch": 6.099305926321409,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023015142090842807,
      "loss": 5.4701,
      "mean_token_accuracy": 0.18922051340341567,
      "num_tokens": 12510005.0,
      "step": 5715
    },
    {
      "entropy": 6.313366460800171,
      "epoch": 6.1046449546182595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002297372217067012,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1863625854253769,
      "num_tokens": 12521058.0,
      "step": 5720
    },
    {
      "entropy": 6.237168836593628,
      "epoch": 6.109983982915109,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022932318230184845,
      "loss": 5.4051,
      "mean_token_accuracy": 0.18609307259321212,
      "num_tokens": 12533207.0,
      "step": 5725
    },
    {
      "entropy": 6.276350736618042,
      "epoch": 6.115323011211959,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00022890930415560523,
      "loss": 5.5515,
      "mean_token_accuracy": 0.177360463142395,
      "num_tokens": 12544689.0,
      "step": 5730
    },
    {
      "entropy": 6.287495279312134,
      "epoch": 6.12066203950881,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022849558872913733,
      "loss": 5.512,
      "mean_token_accuracy": 0.18004681915044785,
      "num_tokens": 12555502.0,
      "step": 5735
    },
    {
      "entropy": 6.306123733520508,
      "epoch": 6.126001067805659,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002280820374830364,
      "loss": 5.5281,
      "mean_token_accuracy": 0.17885361164808272,
      "num_tokens": 12567012.0,
      "step": 5740
    },
    {
      "entropy": 6.303791046142578,
      "epoch": 6.13134009610251,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022766865187731438,
      "loss": 5.5278,
      "mean_token_accuracy": 0.18501299619674683,
      "num_tokens": 12577851.0,
      "step": 5745
    },
    {
      "entropy": 6.28567943572998,
      "epoch": 6.136679124399359,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002272554333713986,
      "loss": 5.5241,
      "mean_token_accuracy": 0.18573147505521775,
      "num_tokens": 12588382.0,
      "step": 5750
    },
    {
      "entropy": 6.271179485321045,
      "epoch": 6.14201815269621,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000226842383424126,
      "loss": 5.5142,
      "mean_token_accuracy": 0.17466562241315842,
      "num_tokens": 12600374.0,
      "step": 5755
    },
    {
      "entropy": 6.318966388702393,
      "epoch": 6.147357180993059,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022642950349373897,
      "loss": 5.4968,
      "mean_token_accuracy": 0.18892965614795684,
      "num_tokens": 12611228.0,
      "step": 5760
    },
    {
      "entropy": 6.280329751968384,
      "epoch": 6.15269620928991,
      "grad_norm": 1.0,
      "learning_rate": 0.00022601679503787925,
      "loss": 5.5105,
      "mean_token_accuracy": 0.18153341859579086,
      "num_tokens": 12623096.0,
      "step": 5765
    },
    {
      "entropy": 6.284728384017944,
      "epoch": 6.158035237586759,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022560425951358333,
      "loss": 5.5769,
      "mean_token_accuracy": 0.17436710447072984,
      "num_tokens": 12633864.0,
      "step": 5770
    },
    {
      "entropy": 6.330235004425049,
      "epoch": 6.1633742658836095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022519189837727727,
      "loss": 5.5593,
      "mean_token_accuracy": 0.18184018284082412,
      "num_tokens": 12644832.0,
      "step": 5775
    },
    {
      "entropy": 6.215789794921875,
      "epoch": 6.168713294180459,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022477971308477147,
      "loss": 5.4848,
      "mean_token_accuracy": 0.18760353922843934,
      "num_tokens": 12655353.0,
      "step": 5780
    },
    {
      "entropy": 6.2849150657653805,
      "epoch": 6.174052322477309,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022436770509125538,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18099713921546937,
      "num_tokens": 12666754.0,
      "step": 5785
    },
    {
      "entropy": 6.3148720264434814,
      "epoch": 6.179391350774159,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022395587585129264,
      "loss": 5.6045,
      "mean_token_accuracy": 0.17165170907974242,
      "num_tokens": 12677550.0,
      "step": 5790
    },
    {
      "entropy": 6.315364980697632,
      "epoch": 6.184730379071009,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022354422681881576,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19243646562099456,
      "num_tokens": 12688414.0,
      "step": 5795
    },
    {
      "entropy": 6.331755542755127,
      "epoch": 6.190069407367859,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002231327594471211,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18594738394021987,
      "num_tokens": 12699998.0,
      "step": 5800
    },
    {
      "entropy": 6.262127447128296,
      "epoch": 6.195408435664709,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022272147518886343,
      "loss": 5.4805,
      "mean_token_accuracy": 0.186960868537426,
      "num_tokens": 12711043.0,
      "step": 5805
    },
    {
      "entropy": 6.298642778396607,
      "epoch": 6.200747463961559,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002223103754960515,
      "loss": 5.5483,
      "mean_token_accuracy": 0.1816814735531807,
      "num_tokens": 12721425.0,
      "step": 5810
    },
    {
      "entropy": 6.304767847061157,
      "epoch": 6.206086492258409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022189946182004207,
      "loss": 5.5381,
      "mean_token_accuracy": 0.17976996153593064,
      "num_tokens": 12731908.0,
      "step": 5815
    },
    {
      "entropy": 6.287419080734253,
      "epoch": 6.211425520555259,
      "grad_norm": 1.015625,
      "learning_rate": 0.00022148873561153538,
      "loss": 5.511,
      "mean_token_accuracy": 0.1849981963634491,
      "num_tokens": 12742783.0,
      "step": 5820
    },
    {
      "entropy": 6.2587018489837645,
      "epoch": 6.216764548852109,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002210781983205698,
      "loss": 5.4744,
      "mean_token_accuracy": 0.18162271231412888,
      "num_tokens": 12753599.0,
      "step": 5825
    },
    {
      "entropy": 6.363664674758911,
      "epoch": 6.222103577148959,
      "grad_norm": 1.25,
      "learning_rate": 0.00022066785139651662,
      "loss": 5.6172,
      "mean_token_accuracy": 0.17767306864261628,
      "num_tokens": 12765310.0,
      "step": 5830
    },
    {
      "entropy": 6.358264780044555,
      "epoch": 6.227442605445809,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022025769628807523,
      "loss": 5.5677,
      "mean_token_accuracy": 0.18100647628307343,
      "num_tokens": 12776254.0,
      "step": 5835
    },
    {
      "entropy": 6.380987882614136,
      "epoch": 6.2327816337426585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00021984773444326772,
      "loss": 5.6027,
      "mean_token_accuracy": 0.173652546107769,
      "num_tokens": 12787963.0,
      "step": 5840
    },
    {
      "entropy": 6.306836652755737,
      "epoch": 6.238120662039509,
      "grad_norm": 1.25,
      "learning_rate": 0.0002194379673094339,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18403929471969604,
      "num_tokens": 12798733.0,
      "step": 5845
    },
    {
      "entropy": 6.297771739959717,
      "epoch": 6.243459690336358,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021902839633322618,
      "loss": 5.547,
      "mean_token_accuracy": 0.18292764127254485,
      "num_tokens": 12809485.0,
      "step": 5850
    },
    {
      "entropy": 6.2989256381988525,
      "epoch": 6.248798718633209,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021861902296060448,
      "loss": 5.5211,
      "mean_token_accuracy": 0.18578432649374008,
      "num_tokens": 12820079.0,
      "step": 5855
    },
    {
      "entropy": 6.329894781112671,
      "epoch": 6.254137746930059,
      "grad_norm": 1.21875,
      "learning_rate": 0.00021820984863683113,
      "loss": 5.5886,
      "mean_token_accuracy": 0.1779401034116745,
      "num_tokens": 12830495.0,
      "step": 5860
    },
    {
      "entropy": 6.249867868423462,
      "epoch": 6.259476775226909,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021780087480646548,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1794280156493187,
      "num_tokens": 12841362.0,
      "step": 5865
    },
    {
      "entropy": 6.23549256324768,
      "epoch": 6.264815803523758,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021739210291335937,
      "loss": 5.3842,
      "mean_token_accuracy": 0.19200924336910247,
      "num_tokens": 12852227.0,
      "step": 5870
    },
    {
      "entropy": 6.34181489944458,
      "epoch": 6.270154831820609,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002169835344006515,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18425769805908204,
      "num_tokens": 12862853.0,
      "step": 5875
    },
    {
      "entropy": 6.316803693771362,
      "epoch": 6.275493860117459,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00021657517071076266,
      "loss": 5.5947,
      "mean_token_accuracy": 0.1759370505809784,
      "num_tokens": 12874092.0,
      "step": 5880
    },
    {
      "entropy": 6.28931770324707,
      "epoch": 6.2808328884143085,
      "grad_norm": 0.96875,
      "learning_rate": 0.00021616701328539057,
      "loss": 5.5461,
      "mean_token_accuracy": 0.18204266130924224,
      "num_tokens": 12885265.0,
      "step": 5885
    },
    {
      "entropy": 6.337766838073731,
      "epoch": 6.286171916711159,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021575906356550463,
      "loss": 5.5164,
      "mean_token_accuracy": 0.18724005669355392,
      "num_tokens": 12896029.0,
      "step": 5890
    },
    {
      "entropy": 6.351672267913818,
      "epoch": 6.2915109450080084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00021535132299134097,
      "loss": 5.538,
      "mean_token_accuracy": 0.1854406401515007,
      "num_tokens": 12906920.0,
      "step": 5895
    },
    {
      "entropy": 6.313889074325561,
      "epoch": 6.296849973304859,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002149437930023974,
      "loss": 5.5461,
      "mean_token_accuracy": 0.1822608396410942,
      "num_tokens": 12918053.0,
      "step": 5900
    },
    {
      "entropy": 6.323455667495727,
      "epoch": 6.302189001601708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021453647503742824,
      "loss": 5.5518,
      "mean_token_accuracy": 0.1901797831058502,
      "num_tokens": 12929231.0,
      "step": 5905
    },
    {
      "entropy": 6.28606595993042,
      "epoch": 6.307528029898559,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002141293705344393,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1860789716243744,
      "num_tokens": 12939765.0,
      "step": 5910
    },
    {
      "entropy": 6.356584215164185,
      "epoch": 6.312867058195408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021372248093068274,
      "loss": 5.5381,
      "mean_token_accuracy": 0.1795216053724289,
      "num_tokens": 12949620.0,
      "step": 5915
    },
    {
      "entropy": 6.24603009223938,
      "epoch": 6.318206086492259,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021331580766265212,
      "loss": 5.5549,
      "mean_token_accuracy": 0.18385580331087112,
      "num_tokens": 12960464.0,
      "step": 5920
    },
    {
      "entropy": 6.282250261306762,
      "epoch": 6.323545114789108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021290935216607726,
      "loss": 5.5007,
      "mean_token_accuracy": 0.18764480501413344,
      "num_tokens": 12971156.0,
      "step": 5925
    },
    {
      "entropy": 6.335037517547607,
      "epoch": 6.328884143085959,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021250311587591887,
      "loss": 5.5386,
      "mean_token_accuracy": 0.17930494695901872,
      "num_tokens": 12981583.0,
      "step": 5930
    },
    {
      "entropy": 6.31908802986145,
      "epoch": 6.334223171382808,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021209710022636403,
      "loss": 5.59,
      "mean_token_accuracy": 0.18331025689840316,
      "num_tokens": 12991937.0,
      "step": 5935
    },
    {
      "entropy": 6.311692571640014,
      "epoch": 6.3395621996796585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002116913066508208,
      "loss": 5.5469,
      "mean_token_accuracy": 0.18102213740348816,
      "num_tokens": 13002570.0,
      "step": 5940
    },
    {
      "entropy": 6.344061946868896,
      "epoch": 6.344901227976508,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021128573658191335,
      "loss": 5.5049,
      "mean_token_accuracy": 0.1811664655804634,
      "num_tokens": 13014096.0,
      "step": 5945
    },
    {
      "entropy": 6.368574094772339,
      "epoch": 6.350240256273358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021088039145147653,
      "loss": 5.5944,
      "mean_token_accuracy": 0.18091420978307723,
      "num_tokens": 13023944.0,
      "step": 5950
    },
    {
      "entropy": 6.333920907974243,
      "epoch": 6.355579284570208,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021047527269055138,
      "loss": 5.6037,
      "mean_token_accuracy": 0.18233564198017121,
      "num_tokens": 13034943.0,
      "step": 5955
    },
    {
      "entropy": 6.265704250335693,
      "epoch": 6.360918312867058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021007038172937936,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18883284628391267,
      "num_tokens": 13046361.0,
      "step": 5960
    },
    {
      "entropy": 6.278402328491211,
      "epoch": 6.366257341163908,
      "grad_norm": 1.125,
      "learning_rate": 0.00020966571999739798,
      "loss": 5.5597,
      "mean_token_accuracy": 0.17997330725193023,
      "num_tokens": 13056795.0,
      "step": 5965
    },
    {
      "entropy": 6.26501317024231,
      "epoch": 6.371596369460758,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020926128892323538,
      "loss": 5.4512,
      "mean_token_accuracy": 0.18857761174440385,
      "num_tokens": 13066849.0,
      "step": 5970
    },
    {
      "entropy": 6.4005210399627686,
      "epoch": 6.376935397757608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020885708993470547,
      "loss": 5.6603,
      "mean_token_accuracy": 0.17295733839273453,
      "num_tokens": 13079071.0,
      "step": 5975
    },
    {
      "entropy": 6.280942773818969,
      "epoch": 6.382274426054458,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002084531244588027,
      "loss": 5.5652,
      "mean_token_accuracy": 0.17425670325756074,
      "num_tokens": 13089781.0,
      "step": 5980
    },
    {
      "entropy": 6.390124797821045,
      "epoch": 6.387613454351308,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00020804939392169713,
      "loss": 5.6063,
      "mean_token_accuracy": 0.177083620429039,
      "num_tokens": 13100814.0,
      "step": 5985
    },
    {
      "entropy": 6.322206830978393,
      "epoch": 6.392952482648158,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002076458997487295,
      "loss": 5.5611,
      "mean_token_accuracy": 0.18351718485355378,
      "num_tokens": 13112712.0,
      "step": 5990
    },
    {
      "entropy": 6.322111177444458,
      "epoch": 6.398291510945008,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020724264336440584,
      "loss": 5.4872,
      "mean_token_accuracy": 0.18441599905490874,
      "num_tokens": 13123560.0,
      "step": 5995
    },
    {
      "entropy": 6.331759405136109,
      "epoch": 6.403630539241858,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020683962619239294,
      "loss": 5.5804,
      "mean_token_accuracy": 0.18275608122348785,
      "num_tokens": 13135879.0,
      "step": 6000
    },
    {
      "epoch": 6.403630539241858,
      "eval_entropy": 6.163486111628544,
      "eval_loss": 6.554488182067871,
      "eval_mean_token_accuracy": 0.13392661240967837,
      "eval_num_tokens": 13135879.0,
      "eval_runtime": 1.819,
      "eval_samples_per_second": 1693.217,
      "eval_steps_per_second": 211.652,
      "step": 6000
    },
    {
      "entropy": 6.352371311187744,
      "epoch": 6.408969567538708,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020643684965551297,
      "loss": 5.5199,
      "mean_token_accuracy": 0.18580380380153655,
      "num_tokens": 13146585.0,
      "step": 6005
    },
    {
      "entropy": 6.315131568908692,
      "epoch": 6.414308595835558,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002060343151757385,
      "loss": 5.48,
      "mean_token_accuracy": 0.18755016028881072,
      "num_tokens": 13156715.0,
      "step": 6010
    },
    {
      "entropy": 6.337861347198486,
      "epoch": 6.419647624132408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00020563202417418763,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17808083295822144,
      "num_tokens": 13168191.0,
      "step": 6015
    },
    {
      "entropy": 6.262475395202637,
      "epoch": 6.424986652429258,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00020522997807111888,
      "loss": 5.4674,
      "mean_token_accuracy": 0.19102598279714583,
      "num_tokens": 13179056.0,
      "step": 6020
    },
    {
      "entropy": 6.319896697998047,
      "epoch": 6.430325680726108,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002048281782859261,
      "loss": 5.562,
      "mean_token_accuracy": 0.17918537855148314,
      "num_tokens": 13190090.0,
      "step": 6025
    },
    {
      "entropy": 6.333342885971069,
      "epoch": 6.435664709022958,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020442662623713348,
      "loss": 5.5598,
      "mean_token_accuracy": 0.17563522905111312,
      "num_tokens": 13199842.0,
      "step": 6030
    },
    {
      "entropy": 6.263208198547363,
      "epoch": 6.441003737319808,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020402532334239072,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1844705805182457,
      "num_tokens": 13210655.0,
      "step": 6035
    },
    {
      "entropy": 6.25391845703125,
      "epoch": 6.446342765616658,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020362427101846785,
      "loss": 5.5483,
      "mean_token_accuracy": 0.18805767744779586,
      "num_tokens": 13221458.0,
      "step": 6040
    },
    {
      "entropy": 6.374371194839478,
      "epoch": 6.451681793913508,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020322347068125037,
      "loss": 5.6416,
      "mean_token_accuracy": 0.17785127460956573,
      "num_tokens": 13231912.0,
      "step": 6045
    },
    {
      "entropy": 6.308995628356934,
      "epoch": 6.457020822210358,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020282292374573398,
      "loss": 5.5069,
      "mean_token_accuracy": 0.18875450044870376,
      "num_tokens": 13243383.0,
      "step": 6050
    },
    {
      "entropy": 6.293632078170776,
      "epoch": 6.462359850507208,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020242263162601991,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1814887836575508,
      "num_tokens": 13254747.0,
      "step": 6055
    },
    {
      "entropy": 6.290720367431641,
      "epoch": 6.4676988788040575,
      "grad_norm": 1.125,
      "learning_rate": 0.00020202259573530974,
      "loss": 5.5654,
      "mean_token_accuracy": 0.17937231659889222,
      "num_tokens": 13265842.0,
      "step": 6060
    },
    {
      "entropy": 6.236067438125611,
      "epoch": 6.473037907100908,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020162281748590039,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1790890023112297,
      "num_tokens": 13276760.0,
      "step": 6065
    },
    {
      "entropy": 6.308027267456055,
      "epoch": 6.478376935397757,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002012232982891793,
      "loss": 5.5553,
      "mean_token_accuracy": 0.18342843651771545,
      "num_tokens": 13287687.0,
      "step": 6070
    },
    {
      "entropy": 6.282391691207886,
      "epoch": 6.483715963694608,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020082403955561928,
      "loss": 5.5932,
      "mean_token_accuracy": 0.17987498492002488,
      "num_tokens": 13298409.0,
      "step": 6075
    },
    {
      "entropy": 6.3039881706237795,
      "epoch": 6.489054991991457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020042504269477359,
      "loss": 5.5223,
      "mean_token_accuracy": 0.18251995444297792,
      "num_tokens": 13309178.0,
      "step": 6080
    },
    {
      "entropy": 6.298363304138183,
      "epoch": 6.494394020288308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000200026309115271,
      "loss": 5.4803,
      "mean_token_accuracy": 0.18861500024795533,
      "num_tokens": 13319381.0,
      "step": 6085
    },
    {
      "entropy": 6.300890016555786,
      "epoch": 6.499733048585157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019962784022481083,
      "loss": 5.5704,
      "mean_token_accuracy": 0.18661998957395554,
      "num_tokens": 13330409.0,
      "step": 6090
    },
    {
      "entropy": 6.348041963577271,
      "epoch": 6.505072076882008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019922963743015782,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17736010551452636,
      "num_tokens": 13341988.0,
      "step": 6095
    },
    {
      "entropy": 6.337808609008789,
      "epoch": 6.510411105178857,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001988317021371373,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1819508343935013,
      "num_tokens": 13353222.0,
      "step": 6100
    },
    {
      "entropy": 6.2926851272583,
      "epoch": 6.5157501334757075,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019843403575063028,
      "loss": 5.539,
      "mean_token_accuracy": 0.1807304948568344,
      "num_tokens": 13363740.0,
      "step": 6105
    },
    {
      "entropy": 6.307255268096924,
      "epoch": 6.521089161772557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019803663967456843,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18084392547607422,
      "num_tokens": 13373972.0,
      "step": 6110
    },
    {
      "entropy": 6.311766338348389,
      "epoch": 6.526428190069407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019763951531192897,
      "loss": 5.5672,
      "mean_token_accuracy": 0.18313288539648057,
      "num_tokens": 13385031.0,
      "step": 6115
    },
    {
      "entropy": 6.350743532180786,
      "epoch": 6.531767218366257,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019724266406473013,
      "loss": 5.5698,
      "mean_token_accuracy": 0.18248473405838012,
      "num_tokens": 13396260.0,
      "step": 6120
    },
    {
      "entropy": 6.3166967868804935,
      "epoch": 6.537106246663107,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00019684608733402543,
      "loss": 5.5822,
      "mean_token_accuracy": 0.1851217567920685,
      "num_tokens": 13408106.0,
      "step": 6125
    },
    {
      "entropy": 6.3371728420257565,
      "epoch": 6.542445274959958,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019644978651989965,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18292033523321152,
      "num_tokens": 13419173.0,
      "step": 6130
    },
    {
      "entropy": 6.345329236984253,
      "epoch": 6.547784303256807,
      "grad_norm": 1.125,
      "learning_rate": 0.00019605376302146344,
      "loss": 5.6047,
      "mean_token_accuracy": 0.17939743846654893,
      "num_tokens": 13430421.0,
      "step": 6135
    },
    {
      "entropy": 6.275396823883057,
      "epoch": 6.553123331553657,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001956580182368483,
      "loss": 5.4882,
      "mean_token_accuracy": 0.18517961949110032,
      "num_tokens": 13441046.0,
      "step": 6140
    },
    {
      "entropy": 6.296225357055664,
      "epoch": 6.558462359850507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001952625535632017,
      "loss": 5.5271,
      "mean_token_accuracy": 0.18224766999483108,
      "num_tokens": 13452660.0,
      "step": 6145
    },
    {
      "entropy": 6.246510744094849,
      "epoch": 6.563801388147358,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001948673703966824,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19840070307254792,
      "num_tokens": 13464215.0,
      "step": 6150
    },
    {
      "entropy": 6.317129278182984,
      "epoch": 6.569140416444207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001944724701324551,
      "loss": 5.5587,
      "mean_token_accuracy": 0.18184617757797242,
      "num_tokens": 13475665.0,
      "step": 6155
    },
    {
      "entropy": 6.30324501991272,
      "epoch": 6.5744794447410575,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001940778541646858,
      "loss": 5.5446,
      "mean_token_accuracy": 0.17954391986131668,
      "num_tokens": 13487411.0,
      "step": 6160
    },
    {
      "entropy": 6.300690221786499,
      "epoch": 6.579818473037907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019368352388653693,
      "loss": 5.5232,
      "mean_token_accuracy": 0.18714903742074968,
      "num_tokens": 13498292.0,
      "step": 6165
    },
    {
      "entropy": 6.340859985351562,
      "epoch": 6.585157501334757,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00019328948069016218,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1790682151913643,
      "num_tokens": 13510046.0,
      "step": 6170
    },
    {
      "entropy": 6.349089908599853,
      "epoch": 6.590496529631607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019289572596670169,
      "loss": 5.6411,
      "mean_token_accuracy": 0.17593853771686555,
      "num_tokens": 13521003.0,
      "step": 6175
    },
    {
      "entropy": 6.262182283401489,
      "epoch": 6.595835557928457,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019250226110627733,
      "loss": 5.4865,
      "mean_token_accuracy": 0.19835155308246613,
      "num_tokens": 13531443.0,
      "step": 6180
    },
    {
      "entropy": 6.257317733764649,
      "epoch": 6.601174586225307,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001921090874979876,
      "loss": 5.5508,
      "mean_token_accuracy": 0.17542205154895782,
      "num_tokens": 13541783.0,
      "step": 6185
    },
    {
      "entropy": 6.2820525646209715,
      "epoch": 6.606513614522157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019171620652990256,
      "loss": 5.5006,
      "mean_token_accuracy": 0.17975751608610152,
      "num_tokens": 13552002.0,
      "step": 6190
    },
    {
      "entropy": 6.316552591323853,
      "epoch": 6.611852642819007,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001913236195890593,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1775813117623329,
      "num_tokens": 13563151.0,
      "step": 6195
    },
    {
      "entropy": 6.219785404205322,
      "epoch": 6.617191671115857,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001909313280614569,
      "loss": 5.4929,
      "mean_token_accuracy": 0.18668177276849746,
      "num_tokens": 13573542.0,
      "step": 6200
    },
    {
      "entropy": 6.309219741821289,
      "epoch": 6.622530699412707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019053933333205137,
      "loss": 5.5385,
      "mean_token_accuracy": 0.18093562871217728,
      "num_tokens": 13584023.0,
      "step": 6205
    },
    {
      "entropy": 6.316692113876343,
      "epoch": 6.627869727709557,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019014763678475095,
      "loss": 5.5075,
      "mean_token_accuracy": 0.18570243269205094,
      "num_tokens": 13594781.0,
      "step": 6210
    },
    {
      "entropy": 6.279546594619751,
      "epoch": 6.633208756006407,
      "grad_norm": 1.046875,
      "learning_rate": 0.00018975623980241124,
      "loss": 5.5343,
      "mean_token_accuracy": 0.1863781675696373,
      "num_tokens": 13606328.0,
      "step": 6215
    },
    {
      "entropy": 6.327500057220459,
      "epoch": 6.638547784303257,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001893651437668303,
      "loss": 5.51,
      "mean_token_accuracy": 0.18514935821294784,
      "num_tokens": 13617209.0,
      "step": 6220
    },
    {
      "entropy": 6.309375858306884,
      "epoch": 6.6438868126001065,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00018897435005874331,
      "loss": 5.5394,
      "mean_token_accuracy": 0.1805016040802002,
      "num_tokens": 13629314.0,
      "step": 6225
    },
    {
      "entropy": 6.312755393981933,
      "epoch": 6.649225840896957,
      "grad_norm": 1.203125,
      "learning_rate": 0.00018858386005781864,
      "loss": 5.5908,
      "mean_token_accuracy": 0.17755433320999145,
      "num_tokens": 13640552.0,
      "step": 6230
    },
    {
      "entropy": 6.303439903259277,
      "epoch": 6.654564869193806,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018819367514265206,
      "loss": 5.5633,
      "mean_token_accuracy": 0.18204109072685243,
      "num_tokens": 13650273.0,
      "step": 6235
    },
    {
      "entropy": 6.30832314491272,
      "epoch": 6.659903897490657,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018780379669076257,
      "loss": 5.5438,
      "mean_token_accuracy": 0.1898371934890747,
      "num_tokens": 13660251.0,
      "step": 6240
    },
    {
      "entropy": 6.269473314285278,
      "epoch": 6.665242925787506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00018741422607858693,
      "loss": 5.4726,
      "mean_token_accuracy": 0.19135648012161255,
      "num_tokens": 13671225.0,
      "step": 6245
    },
    {
      "entropy": 6.254298162460327,
      "epoch": 6.670581954084357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00018702496468147534,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18298991322517394,
      "num_tokens": 13682236.0,
      "step": 6250
    },
    {
      "entropy": 6.279804039001465,
      "epoch": 6.675920982381206,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018663601387368612,
      "loss": 5.5568,
      "mean_token_accuracy": 0.18257627189159392,
      "num_tokens": 13693510.0,
      "step": 6255
    },
    {
      "entropy": 6.36996750831604,
      "epoch": 6.681260010678057,
      "grad_norm": 1.03125,
      "learning_rate": 0.00018624737502838124,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1762879952788353,
      "num_tokens": 13706057.0,
      "step": 6260
    },
    {
      "entropy": 6.348255586624146,
      "epoch": 6.686599038974906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00018585904951762123,
      "loss": 5.5999,
      "mean_token_accuracy": 0.1818082019686699,
      "num_tokens": 13716674.0,
      "step": 6265
    },
    {
      "entropy": 6.308555364608765,
      "epoch": 6.6919380672717566,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00018547103871236048,
      "loss": 5.4897,
      "mean_token_accuracy": 0.18605282455682753,
      "num_tokens": 13727745.0,
      "step": 6270
    },
    {
      "entropy": 6.2941718101501465,
      "epoch": 6.697277095568607,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018508334398244225,
      "loss": 5.6056,
      "mean_token_accuracy": 0.1769578352570534,
      "num_tokens": 13738970.0,
      "step": 6275
    },
    {
      "entropy": 6.330140542984009,
      "epoch": 6.7026161238654565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00018469596669659394,
      "loss": 5.6379,
      "mean_token_accuracy": 0.17538753598928453,
      "num_tokens": 13750281.0,
      "step": 6280
    },
    {
      "entropy": 6.391390037536621,
      "epoch": 6.707955152162306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018430890822242242,
      "loss": 5.5846,
      "mean_token_accuracy": 0.18001397252082824,
      "num_tokens": 13760757.0,
      "step": 6285
    },
    {
      "entropy": 6.290466976165772,
      "epoch": 6.713294180459156,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018392216992640866,
      "loss": 5.5536,
      "mean_token_accuracy": 0.18084823042154313,
      "num_tokens": 13771639.0,
      "step": 6290
    },
    {
      "entropy": 6.289993000030518,
      "epoch": 6.718633208756007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018353575317390353,
      "loss": 5.5329,
      "mean_token_accuracy": 0.18451715558767318,
      "num_tokens": 13782387.0,
      "step": 6295
    },
    {
      "entropy": 6.327233743667603,
      "epoch": 6.723972237052856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00018314965932912265,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1848110541701317,
      "num_tokens": 13792693.0,
      "step": 6300
    },
    {
      "entropy": 6.3418426513671875,
      "epoch": 6.729311265349707,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001827638897551417,
      "loss": 5.6132,
      "mean_token_accuracy": 0.17475601583719252,
      "num_tokens": 13804553.0,
      "step": 6305
    },
    {
      "entropy": 6.335278129577636,
      "epoch": 6.734650293646556,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018237844581389147,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18327444195747375,
      "num_tokens": 13814339.0,
      "step": 6310
    },
    {
      "entropy": 6.298328924179077,
      "epoch": 6.739989321943407,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018199332886615312,
      "loss": 5.6229,
      "mean_token_accuracy": 0.17313117086887359,
      "num_tokens": 13824960.0,
      "step": 6315
    },
    {
      "entropy": 6.320883655548096,
      "epoch": 6.745328350240256,
      "grad_norm": 1.125,
      "learning_rate": 0.0001816085402715533,
      "loss": 5.6079,
      "mean_token_accuracy": 0.1743606612086296,
      "num_tokens": 13835296.0,
      "step": 6320
    },
    {
      "entropy": 6.229419183731079,
      "epoch": 6.7506673785371065,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018122408138855972,
      "loss": 5.4237,
      "mean_token_accuracy": 0.19275894314050673,
      "num_tokens": 13845932.0,
      "step": 6325
    },
    {
      "entropy": 6.319077062606811,
      "epoch": 6.756006406833956,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001808399535744757,
      "loss": 5.5593,
      "mean_token_accuracy": 0.18214440643787383,
      "num_tokens": 13857095.0,
      "step": 6330
    },
    {
      "entropy": 6.334871053695679,
      "epoch": 6.761345435130806,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018045615818543614,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1771411955356598,
      "num_tokens": 13867882.0,
      "step": 6335
    },
    {
      "entropy": 6.273644542694091,
      "epoch": 6.766684463427656,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001800726965764019,
      "loss": 5.4851,
      "mean_token_accuracy": 0.1845867559313774,
      "num_tokens": 13878208.0,
      "step": 6340
    },
    {
      "entropy": 6.261086702346802,
      "epoch": 6.772023491724506,
      "grad_norm": 1.109375,
      "learning_rate": 0.000179689570101156,
      "loss": 5.4102,
      "mean_token_accuracy": 0.19955071955919265,
      "num_tokens": 13889075.0,
      "step": 6345
    },
    {
      "entropy": 6.311852312088012,
      "epoch": 6.777362520021356,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00017930678011229774,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17336114048957824,
      "num_tokens": 13900305.0,
      "step": 6350
    },
    {
      "entropy": 6.275598430633545,
      "epoch": 6.782701548318206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017892432796123874,
      "loss": 5.5312,
      "mean_token_accuracy": 0.17904499024152756,
      "num_tokens": 13912407.0,
      "step": 6355
    },
    {
      "entropy": 6.317451429367066,
      "epoch": 6.788040576615056,
      "grad_norm": 1.1875,
      "learning_rate": 0.00017854221499819783,
      "loss": 5.5542,
      "mean_token_accuracy": 0.1790584310889244,
      "num_tokens": 13923777.0,
      "step": 6360
    },
    {
      "entropy": 6.29730372428894,
      "epoch": 6.793379604911906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001781604425721966,
      "loss": 5.5268,
      "mean_token_accuracy": 0.18457200676202773,
      "num_tokens": 13934634.0,
      "step": 6365
    },
    {
      "entropy": 6.306339073181152,
      "epoch": 6.798718633208756,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001777790120310541,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1794625535607338,
      "num_tokens": 13945092.0,
      "step": 6370
    },
    {
      "entropy": 6.3330113887786865,
      "epoch": 6.804057661505606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017739792472138248,
      "loss": 5.6036,
      "mean_token_accuracy": 0.1796196311712265,
      "num_tokens": 13955834.0,
      "step": 6375
    },
    {
      "entropy": 6.360069131851196,
      "epoch": 6.809396689802456,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017701718198858248,
      "loss": 5.6001,
      "mean_token_accuracy": 0.1716108053922653,
      "num_tokens": 13966749.0,
      "step": 6380
    },
    {
      "entropy": 6.284036350250244,
      "epoch": 6.814735718099306,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001766367851768375,
      "loss": 5.5121,
      "mean_token_accuracy": 0.18103928864002228,
      "num_tokens": 13977387.0,
      "step": 6385
    },
    {
      "entropy": 6.332955169677734,
      "epoch": 6.8200747463961555,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017625673562911054,
      "loss": 5.6054,
      "mean_token_accuracy": 0.180105060338974,
      "num_tokens": 13987949.0,
      "step": 6390
    },
    {
      "entropy": 6.319362831115723,
      "epoch": 6.825413774693006,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017587703468713807,
      "loss": 5.5248,
      "mean_token_accuracy": 0.1792665407061577,
      "num_tokens": 13998723.0,
      "step": 6395
    },
    {
      "entropy": 6.3338072299957275,
      "epoch": 6.830752802989856,
      "grad_norm": 1.125,
      "learning_rate": 0.00017549768369142633,
      "loss": 5.5418,
      "mean_token_accuracy": 0.18090006113052368,
      "num_tokens": 14009760.0,
      "step": 6400
    },
    {
      "entropy": 6.336702346801758,
      "epoch": 6.836091831286706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00017511868398124546,
      "loss": 5.5971,
      "mean_token_accuracy": 0.18475197553634642,
      "num_tokens": 14020122.0,
      "step": 6405
    },
    {
      "entropy": 6.296493196487427,
      "epoch": 6.841430859583555,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001747400368946261,
      "loss": 5.6277,
      "mean_token_accuracy": 0.1744499459862709,
      "num_tokens": 14031708.0,
      "step": 6410
    },
    {
      "entropy": 6.37297477722168,
      "epoch": 6.846769887880406,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017436174376835349,
      "loss": 5.6148,
      "mean_token_accuracy": 0.17940505146980285,
      "num_tokens": 14042468.0,
      "step": 6415
    },
    {
      "entropy": 6.285836553573608,
      "epoch": 6.852108916177256,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017398380593796326,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18315577358007432,
      "num_tokens": 14053254.0,
      "step": 6420
    },
    {
      "entropy": 6.311503219604492,
      "epoch": 6.857447944474106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017360622473773705,
      "loss": 5.5002,
      "mean_token_accuracy": 0.18174608796834946,
      "num_tokens": 14064732.0,
      "step": 6425
    },
    {
      "entropy": 6.2944371700286865,
      "epoch": 6.862786972770956,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00017322900150069703,
      "loss": 5.5576,
      "mean_token_accuracy": 0.1816406726837158,
      "num_tokens": 14075269.0,
      "step": 6430
    },
    {
      "entropy": 6.295346355438232,
      "epoch": 6.868126001067806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000172852137558602,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18191902190446854,
      "num_tokens": 14085402.0,
      "step": 6435
    },
    {
      "entropy": 6.378720998764038,
      "epoch": 6.873465029364656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00017247563424194188,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1773359015583992,
      "num_tokens": 14097183.0,
      "step": 6440
    },
    {
      "entropy": 6.349096584320068,
      "epoch": 6.8788040576615055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017209949287993394,
      "loss": 5.536,
      "mean_token_accuracy": 0.18651263564825057,
      "num_tokens": 14107870.0,
      "step": 6445
    },
    {
      "entropy": 6.345979166030884,
      "epoch": 6.884143085958356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017172371480051688,
      "loss": 5.5955,
      "mean_token_accuracy": 0.18064535558223724,
      "num_tokens": 14118859.0,
      "step": 6450
    },
    {
      "entropy": 6.2495989322662355,
      "epoch": 6.889482114255205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001713483013303476,
      "loss": 5.5555,
      "mean_token_accuracy": 0.18182558417320252,
      "num_tokens": 14130006.0,
      "step": 6455
    },
    {
      "entropy": 6.318955755233764,
      "epoch": 6.894821142552056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017097325379479528,
      "loss": 5.6223,
      "mean_token_accuracy": 0.1748135954141617,
      "num_tokens": 14141376.0,
      "step": 6460
    },
    {
      "entropy": 6.259933233261108,
      "epoch": 6.900160170848905,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001705985735179375,
      "loss": 5.4424,
      "mean_token_accuracy": 0.1841606691479683,
      "num_tokens": 14151825.0,
      "step": 6465
    },
    {
      "entropy": 6.3122239112854,
      "epoch": 6.905499199145756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017022426182255497,
      "loss": 5.6051,
      "mean_token_accuracy": 0.18254586160182953,
      "num_tokens": 14162509.0,
      "step": 6470
    },
    {
      "entropy": 6.370364761352539,
      "epoch": 6.910838227442605,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016985032003012738,
      "loss": 5.6154,
      "mean_token_accuracy": 0.17966496497392653,
      "num_tokens": 14173591.0,
      "step": 6475
    },
    {
      "entropy": 6.283137416839599,
      "epoch": 6.916177255739456,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016947674946082863,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18040162324905396,
      "num_tokens": 14185562.0,
      "step": 6480
    },
    {
      "entropy": 6.273719644546508,
      "epoch": 6.921516284036305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016910355143352147,
      "loss": 5.4709,
      "mean_token_accuracy": 0.18578019887208938,
      "num_tokens": 14196231.0,
      "step": 6485
    },
    {
      "entropy": 6.2712445735931395,
      "epoch": 6.9268553123331555,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016873072726575407,
      "loss": 5.4758,
      "mean_token_accuracy": 0.1880102276802063,
      "num_tokens": 14209085.0,
      "step": 6490
    },
    {
      "entropy": 6.3388467788696286,
      "epoch": 6.932194340630005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016835827827375426,
      "loss": 5.5858,
      "mean_token_accuracy": 0.17689670920372008,
      "num_tokens": 14219805.0,
      "step": 6495
    },
    {
      "entropy": 6.313488912582398,
      "epoch": 6.9375333689268555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016798620577242567,
      "loss": 5.476,
      "mean_token_accuracy": 0.19206169694662095,
      "num_tokens": 14230505.0,
      "step": 6500
    },
    {
      "epoch": 6.9375333689268555,
      "eval_entropy": 6.111162968425008,
      "eval_loss": 6.5421881675720215,
      "eval_mean_token_accuracy": 0.13504700539754583,
      "eval_num_tokens": 14230505.0,
      "eval_runtime": 1.8124,
      "eval_samples_per_second": 1699.416,
      "eval_steps_per_second": 212.427,
      "step": 6500
    },
    {
      "entropy": 6.271717882156372,
      "epoch": 6.942872397223705,
      "grad_norm": 1.125,
      "learning_rate": 0.00016761451107534249,
      "loss": 5.5262,
      "mean_token_accuracy": 0.19075504094362258,
      "num_tokens": 14241584.0,
      "step": 6505
    },
    {
      "entropy": 6.278272104263306,
      "epoch": 6.948211425520555,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016724319549474534,
      "loss": 5.5543,
      "mean_token_accuracy": 0.19189462959766387,
      "num_tokens": 14252571.0,
      "step": 6510
    },
    {
      "entropy": 6.322755002975464,
      "epoch": 6.953550453817405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016687226034153624,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17759112566709517,
      "num_tokens": 14263470.0,
      "step": 6515
    },
    {
      "entropy": 6.299075269699097,
      "epoch": 6.958889482114255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016650170692527407,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1799885153770447,
      "num_tokens": 14274241.0,
      "step": 6520
    },
    {
      "entropy": 6.338451814651489,
      "epoch": 6.964228510411105,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001661315365541702,
      "loss": 5.5802,
      "mean_token_accuracy": 0.1735849931836128,
      "num_tokens": 14284542.0,
      "step": 6525
    },
    {
      "entropy": 6.280116224288941,
      "epoch": 6.969567538707955,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016576175053508357,
      "loss": 5.5888,
      "mean_token_accuracy": 0.1822688788175583,
      "num_tokens": 14295504.0,
      "step": 6530
    },
    {
      "entropy": 6.25789065361023,
      "epoch": 6.974906567004805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016539235017351634,
      "loss": 5.571,
      "mean_token_accuracy": 0.18007161021232604,
      "num_tokens": 14306575.0,
      "step": 6535
    },
    {
      "entropy": 6.347195148468018,
      "epoch": 6.980245595301655,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016502333677360887,
      "loss": 5.5981,
      "mean_token_accuracy": 0.17393214553594588,
      "num_tokens": 14316677.0,
      "step": 6540
    },
    {
      "entropy": 6.3871876239776615,
      "epoch": 6.9855846235985055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016465471163813576,
      "loss": 5.5593,
      "mean_token_accuracy": 0.18010709881782533,
      "num_tokens": 14327027.0,
      "step": 6545
    },
    {
      "entropy": 6.316414451599121,
      "epoch": 6.990923651895355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016428647606850033,
      "loss": 5.5865,
      "mean_token_accuracy": 0.18866724222898484,
      "num_tokens": 14338771.0,
      "step": 6550
    },
    {
      "entropy": 6.340197610855102,
      "epoch": 6.9962626801922045,
      "grad_norm": 1.25,
      "learning_rate": 0.00016391863136473118,
      "loss": 5.5756,
      "mean_token_accuracy": 0.17773306369781494,
      "num_tokens": 14350028.0,
      "step": 6555
    },
    {
      "entropy": 6.26469612121582,
      "epoch": 7.00106780565937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001635511788254765,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18587267564402687,
      "num_tokens": 14359688.0,
      "step": 6560
    },
    {
      "entropy": 6.269435834884644,
      "epoch": 7.00640683395622,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016318411974800025,
      "loss": 5.4091,
      "mean_token_accuracy": 0.18694305717945098,
      "num_tokens": 14369892.0,
      "step": 6565
    },
    {
      "entropy": 6.2558362007141115,
      "epoch": 7.01174586225307,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016281745542817716,
      "loss": 5.3584,
      "mean_token_accuracy": 0.192064106464386,
      "num_tokens": 14381383.0,
      "step": 6570
    },
    {
      "entropy": 6.297444868087768,
      "epoch": 7.01708489054992,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001624511871604884,
      "loss": 5.4562,
      "mean_token_accuracy": 0.192567777633667,
      "num_tokens": 14391590.0,
      "step": 6575
    },
    {
      "entropy": 6.312979078292846,
      "epoch": 7.02242391884677,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016208531623801702,
      "loss": 5.3939,
      "mean_token_accuracy": 0.1895030602812767,
      "num_tokens": 14402345.0,
      "step": 6580
    },
    {
      "entropy": 6.3703536033630375,
      "epoch": 7.02776294714362,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016171984395244276,
      "loss": 5.4881,
      "mean_token_accuracy": 0.18503127098083497,
      "num_tokens": 14412769.0,
      "step": 6585
    },
    {
      "entropy": 6.305860710144043,
      "epoch": 7.03310197544047,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016135477159403866,
      "loss": 5.4148,
      "mean_token_accuracy": 0.2001012980937958,
      "num_tokens": 14424509.0,
      "step": 6590
    },
    {
      "entropy": 6.392051410675049,
      "epoch": 7.0384410037373195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016099010045166537,
      "loss": 5.5376,
      "mean_token_accuracy": 0.18047535121440889,
      "num_tokens": 14435189.0,
      "step": 6595
    },
    {
      "entropy": 6.353335762023926,
      "epoch": 7.04378003203417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001606258318127674,
      "loss": 5.5095,
      "mean_token_accuracy": 0.1861751824617386,
      "num_tokens": 14445676.0,
      "step": 6600
    },
    {
      "entropy": 6.348418712615967,
      "epoch": 7.049119060331019,
      "grad_norm": 1.25,
      "learning_rate": 0.00016026196696336786,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1838784009218216,
      "num_tokens": 14456848.0,
      "step": 6605
    },
    {
      "entropy": 6.287606906890869,
      "epoch": 7.05445808862787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015989850718806483,
      "loss": 5.4332,
      "mean_token_accuracy": 0.18585668951272966,
      "num_tokens": 14468939.0,
      "step": 6610
    },
    {
      "entropy": 6.322085952758789,
      "epoch": 7.059797116924719,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015953545377002593,
      "loss": 5.5085,
      "mean_token_accuracy": 0.18388541787862778,
      "num_tokens": 14480341.0,
      "step": 6615
    },
    {
      "entropy": 6.3109293460845945,
      "epoch": 7.06513614522157,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015917280799098414,
      "loss": 5.4129,
      "mean_token_accuracy": 0.19869672060012816,
      "num_tokens": 14491517.0,
      "step": 6620
    },
    {
      "entropy": 6.279620122909546,
      "epoch": 7.070475173518419,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001588105711312336,
      "loss": 5.4366,
      "mean_token_accuracy": 0.18694413006305693,
      "num_tokens": 14504351.0,
      "step": 6625
    },
    {
      "entropy": 6.318189525604248,
      "epoch": 7.07581420181527,
      "grad_norm": 1.0625,
      "learning_rate": 0.00015844874446962448,
      "loss": 5.5338,
      "mean_token_accuracy": 0.18216532617807388,
      "num_tokens": 14516431.0,
      "step": 6630
    },
    {
      "entropy": 6.295264530181885,
      "epoch": 7.081153230112119,
      "grad_norm": 1.09375,
      "learning_rate": 0.000158087329283559,
      "loss": 5.3816,
      "mean_token_accuracy": 0.19511567950248718,
      "num_tokens": 14527248.0,
      "step": 6635
    },
    {
      "entropy": 6.3042511463165285,
      "epoch": 7.08649225840897,
      "grad_norm": 1.25,
      "learning_rate": 0.00015772632684898653,
      "loss": 5.4665,
      "mean_token_accuracy": 0.18800214529037476,
      "num_tokens": 14538380.0,
      "step": 6640
    },
    {
      "entropy": 6.314582014083863,
      "epoch": 7.09183128670582,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015736573844039953,
      "loss": 5.4482,
      "mean_token_accuracy": 0.18293119072914124,
      "num_tokens": 14549710.0,
      "step": 6645
    },
    {
      "entropy": 6.282787704467774,
      "epoch": 7.0971703150026695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015700556533082826,
      "loss": 5.4284,
      "mean_token_accuracy": 0.1919356808066368,
      "num_tokens": 14560623.0,
      "step": 6650
    },
    {
      "entropy": 6.293274068832398,
      "epoch": 7.10250934329952,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015664580879183737,
      "loss": 5.5054,
      "mean_token_accuracy": 0.1837783768773079,
      "num_tokens": 14570816.0,
      "step": 6655
    },
    {
      "entropy": 6.316432905197144,
      "epoch": 7.107848371596369,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015628647009352037,
      "loss": 5.4329,
      "mean_token_accuracy": 0.18765368163585663,
      "num_tokens": 14582014.0,
      "step": 6660
    },
    {
      "entropy": 6.275792598724365,
      "epoch": 7.11318739989322,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015592755050449603,
      "loss": 5.3662,
      "mean_token_accuracy": 0.19992139488458632,
      "num_tokens": 14592100.0,
      "step": 6665
    },
    {
      "entropy": 6.173001956939697,
      "epoch": 7.118526428190069,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015556905129190309,
      "loss": 5.3337,
      "mean_token_accuracy": 0.20151432305574418,
      "num_tokens": 14603894.0,
      "step": 6670
    },
    {
      "entropy": 6.229399871826172,
      "epoch": 7.12386545648692,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00015521097372139654,
      "loss": 5.3648,
      "mean_token_accuracy": 0.19738561660051346,
      "num_tokens": 14614697.0,
      "step": 6675
    },
    {
      "entropy": 6.316095399856567,
      "epoch": 7.129204484783769,
      "grad_norm": 1.25,
      "learning_rate": 0.00015485331905714257,
      "loss": 5.3925,
      "mean_token_accuracy": 0.19631152153015136,
      "num_tokens": 14624302.0,
      "step": 6680
    },
    {
      "entropy": 6.291509103775025,
      "epoch": 7.13454351308062,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015449608856181422,
      "loss": 5.4087,
      "mean_token_accuracy": 0.19371372908353807,
      "num_tokens": 14635081.0,
      "step": 6685
    },
    {
      "entropy": 6.264547395706177,
      "epoch": 7.139882541377469,
      "grad_norm": 1.125,
      "learning_rate": 0.00015413928349658745,
      "loss": 5.3872,
      "mean_token_accuracy": 0.2043850302696228,
      "num_tokens": 14645817.0,
      "step": 6690
    },
    {
      "entropy": 6.287539434432984,
      "epoch": 7.1452215696743195,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001537829051211358,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1920837566256523,
      "num_tokens": 14656442.0,
      "step": 6695
    },
    {
      "entropy": 6.278778505325318,
      "epoch": 7.150560597971169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015342695469362682,
      "loss": 5.3925,
      "mean_token_accuracy": 0.19858739972114564,
      "num_tokens": 14666974.0,
      "step": 6700
    },
    {
      "entropy": 6.348932266235352,
      "epoch": 7.1558996262680195,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015307143347071688,
      "loss": 5.5826,
      "mean_token_accuracy": 0.18329962342977524,
      "num_tokens": 14677952.0,
      "step": 6705
    },
    {
      "entropy": 6.270998954772949,
      "epoch": 7.161238654564869,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001527163427075474,
      "loss": 5.4297,
      "mean_token_accuracy": 0.19348912090063095,
      "num_tokens": 14688858.0,
      "step": 6710
    },
    {
      "entropy": 6.274411296844482,
      "epoch": 7.166577682861719,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015236168365773985,
      "loss": 5.4505,
      "mean_token_accuracy": 0.18570196628570557,
      "num_tokens": 14699992.0,
      "step": 6715
    },
    {
      "entropy": 6.332328271865845,
      "epoch": 7.171916711158569,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015200745757339149,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19031400680541993,
      "num_tokens": 14710720.0,
      "step": 6720
    },
    {
      "entropy": 6.32621431350708,
      "epoch": 7.177255739455419,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00015165366570507138,
      "loss": 5.4728,
      "mean_token_accuracy": 0.18544712662696838,
      "num_tokens": 14721101.0,
      "step": 6725
    },
    {
      "entropy": 6.3247661113739015,
      "epoch": 7.182594767752269,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015130030930181527,
      "loss": 5.4501,
      "mean_token_accuracy": 0.19216947257518768,
      "num_tokens": 14731428.0,
      "step": 6730
    },
    {
      "entropy": 6.322046327590942,
      "epoch": 7.187933796049119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001509473896111218,
      "loss": 5.4599,
      "mean_token_accuracy": 0.18756082206964492,
      "num_tokens": 14743419.0,
      "step": 6735
    },
    {
      "entropy": 6.281675624847412,
      "epoch": 7.193272824345969,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015059490787894768,
      "loss": 5.4847,
      "mean_token_accuracy": 0.18561287969350815,
      "num_tokens": 14756162.0,
      "step": 6740
    },
    {
      "entropy": 6.386045646667481,
      "epoch": 7.198611852642819,
      "grad_norm": 1.25,
      "learning_rate": 0.0001502428653497034,
      "loss": 5.5361,
      "mean_token_accuracy": 0.18351231813430785,
      "num_tokens": 14767264.0,
      "step": 6745
    },
    {
      "entropy": 6.404717302322387,
      "epoch": 7.203950880939669,
      "grad_norm": 1.375,
      "learning_rate": 0.0001498912632662489,
      "loss": 5.4631,
      "mean_token_accuracy": 0.18551670759916306,
      "num_tokens": 14778344.0,
      "step": 6750
    },
    {
      "entropy": 6.340229177474976,
      "epoch": 7.209289909236519,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014954010286988935,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18180549144744873,
      "num_tokens": 14790518.0,
      "step": 6755
    },
    {
      "entropy": 6.292685031890869,
      "epoch": 7.2146289375333685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014918938540037018,
      "loss": 5.4807,
      "mean_token_accuracy": 0.18915681540966034,
      "num_tokens": 14801792.0,
      "step": 6760
    },
    {
      "entropy": 6.231640481948853,
      "epoch": 7.219967965830219,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014883911209587366,
      "loss": 5.366,
      "mean_token_accuracy": 0.20036828219890596,
      "num_tokens": 14812259.0,
      "step": 6765
    },
    {
      "entropy": 6.258896732330323,
      "epoch": 7.2253069941270684,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001484892841930134,
      "loss": 5.4381,
      "mean_token_accuracy": 0.19050138741731643,
      "num_tokens": 14823858.0,
      "step": 6770
    },
    {
      "entropy": 6.2924394607543945,
      "epoch": 7.230646022423919,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000148139902926831,
      "loss": 5.4068,
      "mean_token_accuracy": 0.19775196760892869,
      "num_tokens": 14834069.0,
      "step": 6775
    },
    {
      "entropy": 6.317218446731568,
      "epoch": 7.235985050720769,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014779096953079096,
      "loss": 5.4495,
      "mean_token_accuracy": 0.19177347272634507,
      "num_tokens": 14844998.0,
      "step": 6780
    },
    {
      "entropy": 6.294413757324219,
      "epoch": 7.241324079017619,
      "grad_norm": 1.078125,
      "learning_rate": 0.00014744248523677658,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18353998363018037,
      "num_tokens": 14855997.0,
      "step": 6785
    },
    {
      "entropy": 6.298984670639038,
      "epoch": 7.246663107314469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014709445127508594,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1885671928524971,
      "num_tokens": 14866813.0,
      "step": 6790
    },
    {
      "entropy": 6.257570743560791,
      "epoch": 7.252002135611319,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014674686887442695,
      "loss": 5.4425,
      "mean_token_accuracy": 0.18660287111997603,
      "num_tokens": 14877681.0,
      "step": 6795
    },
    {
      "entropy": 6.268080139160157,
      "epoch": 7.257341163908169,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014639973926191353,
      "loss": 5.5047,
      "mean_token_accuracy": 0.18286219090223313,
      "num_tokens": 14888896.0,
      "step": 6800
    },
    {
      "entropy": 6.3286150932312015,
      "epoch": 7.262680192205019,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014605306366306093,
      "loss": 5.4721,
      "mean_token_accuracy": 0.18981748819351196,
      "num_tokens": 14899230.0,
      "step": 6805
    },
    {
      "entropy": 6.259786605834961,
      "epoch": 7.268019220501869,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014570684330178152,
      "loss": 5.4336,
      "mean_token_accuracy": 0.1927536189556122,
      "num_tokens": 14910726.0,
      "step": 6810
    },
    {
      "entropy": 6.311066293716431,
      "epoch": 7.2733582487987185,
      "grad_norm": 1.125,
      "learning_rate": 0.00014536107940038063,
      "loss": 5.4831,
      "mean_token_accuracy": 0.18997626900672912,
      "num_tokens": 14922464.0,
      "step": 6815
    },
    {
      "entropy": 6.3254467964172365,
      "epoch": 7.278697277095569,
      "grad_norm": 1.078125,
      "learning_rate": 0.00014501577317955195,
      "loss": 5.5114,
      "mean_token_accuracy": 0.18482305705547333,
      "num_tokens": 14932719.0,
      "step": 6820
    },
    {
      "entropy": 6.237675046920776,
      "epoch": 7.284036305392418,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014467092585837348,
      "loss": 5.3684,
      "mean_token_accuracy": 0.1937697485089302,
      "num_tokens": 14943497.0,
      "step": 6825
    },
    {
      "entropy": 6.2961009502410885,
      "epoch": 7.289375333689269,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000144326538654303,
      "loss": 5.4553,
      "mean_token_accuracy": 0.19217261373996736,
      "num_tokens": 14953887.0,
      "step": 6830
    },
    {
      "entropy": 6.266338968276978,
      "epoch": 7.294714361986118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014398261278317405,
      "loss": 5.4194,
      "mean_token_accuracy": 0.1873815566301346,
      "num_tokens": 14965193.0,
      "step": 6835
    },
    {
      "entropy": 6.252061939239502,
      "epoch": 7.300053390282969,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00014363914945919126,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19088050872087478,
      "num_tokens": 14976266.0,
      "step": 6840
    },
    {
      "entropy": 6.311152935028076,
      "epoch": 7.305392418579818,
      "grad_norm": 1.25,
      "learning_rate": 0.0001432961498949263,
      "loss": 5.4244,
      "mean_token_accuracy": 0.19315261840820314,
      "num_tokens": 14987993.0,
      "step": 6845
    },
    {
      "entropy": 6.3497191905975345,
      "epoch": 7.310731446876669,
      "grad_norm": 1.125,
      "learning_rate": 0.0001429536153013137,
      "loss": 5.4808,
      "mean_token_accuracy": 0.18216035068035125,
      "num_tokens": 14998445.0,
      "step": 6850
    },
    {
      "entropy": 6.271890926361084,
      "epoch": 7.316070475173518,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014261154688764645,
      "loss": 5.4611,
      "mean_token_accuracy": 0.188706374168396,
      "num_tokens": 15008788.0,
      "step": 6855
    },
    {
      "entropy": 6.2674929141998295,
      "epoch": 7.321409503470369,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00014226994586157148,
      "loss": 5.4978,
      "mean_token_accuracy": 0.184069262444973,
      "num_tokens": 15020167.0,
      "step": 6860
    },
    {
      "entropy": 6.334926986694336,
      "epoch": 7.326748531767218,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014192881342908604,
      "loss": 5.4803,
      "mean_token_accuracy": 0.18917828649282456,
      "num_tokens": 15031082.0,
      "step": 6865
    },
    {
      "entropy": 6.286368560791016,
      "epoch": 7.3320875600640685,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00014158815079453268,
      "loss": 5.4412,
      "mean_token_accuracy": 0.19314158707857132,
      "num_tokens": 15041728.0,
      "step": 6870
    },
    {
      "entropy": 6.300330734252929,
      "epoch": 7.337426588360918,
      "grad_norm": 1.125,
      "learning_rate": 0.00014124795916059548,
      "loss": 5.4293,
      "mean_token_accuracy": 0.19236666411161424,
      "num_tokens": 15052131.0,
      "step": 6875
    },
    {
      "entropy": 6.248535871505737,
      "epoch": 7.342765616657768,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014090823972829586,
      "loss": 5.3861,
      "mean_token_accuracy": 0.19000423103570938,
      "num_tokens": 15063377.0,
      "step": 6880
    },
    {
      "entropy": 6.272410345077515,
      "epoch": 7.348104644954618,
      "grad_norm": 1.125,
      "learning_rate": 0.00014056899369698785,
      "loss": 5.387,
      "mean_token_accuracy": 0.19852375388145446,
      "num_tokens": 15073808.0,
      "step": 6885
    },
    {
      "entropy": 6.3073197364807125,
      "epoch": 7.353443673251468,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001402302222643545,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1940150499343872,
      "num_tokens": 15084409.0,
      "step": 6890
    },
    {
      "entropy": 6.310897445678711,
      "epoch": 7.358782701548318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013989192662640306,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18742646425962448,
      "num_tokens": 15094861.0,
      "step": 6895
    },
    {
      "entropy": 6.303181982040405,
      "epoch": 7.364121729845168,
      "grad_norm": 1.125,
      "learning_rate": 0.0001395541079774612,
      "loss": 5.4783,
      "mean_token_accuracy": 0.18808880746364592,
      "num_tokens": 15104939.0,
      "step": 6900
    },
    {
      "entropy": 6.226565361022949,
      "epoch": 7.369460758142019,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013921676751017255,
      "loss": 5.365,
      "mean_token_accuracy": 0.19940806925296783,
      "num_tokens": 15115243.0,
      "step": 6905
    },
    {
      "entropy": 6.267017602920532,
      "epoch": 7.374799786438868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013887990641549245,
      "loss": 5.4636,
      "mean_token_accuracy": 0.18399228155612946,
      "num_tokens": 15127363.0,
      "step": 6910
    },
    {
      "entropy": 6.279464530944824,
      "epoch": 7.380138814735718,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013854352588268408,
      "loss": 5.3497,
      "mean_token_accuracy": 0.19327647387981414,
      "num_tokens": 15138128.0,
      "step": 6915
    },
    {
      "entropy": 6.303591060638428,
      "epoch": 7.385477843032568,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001382076270993138,
      "loss": 5.5058,
      "mean_token_accuracy": 0.1778939798474312,
      "num_tokens": 15148911.0,
      "step": 6920
    },
    {
      "entropy": 6.257078218460083,
      "epoch": 7.3908168713294184,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001378722112512474,
      "loss": 5.4458,
      "mean_token_accuracy": 0.18303258419036866,
      "num_tokens": 15159564.0,
      "step": 6925
    },
    {
      "entropy": 6.266393041610717,
      "epoch": 7.396155899626268,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001375372795226455,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1912732571363449,
      "num_tokens": 15170536.0,
      "step": 6930
    },
    {
      "entropy": 6.302108573913574,
      "epoch": 7.401494927923118,
      "grad_norm": 1.21875,
      "learning_rate": 0.00013720283309595987,
      "loss": 5.463,
      "mean_token_accuracy": 0.19185779988765717,
      "num_tokens": 15181242.0,
      "step": 6935
    },
    {
      "entropy": 6.343116331100464,
      "epoch": 7.406833956219968,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00013686887315192842,
      "loss": 5.4834,
      "mean_token_accuracy": 0.18818345814943313,
      "num_tokens": 15192873.0,
      "step": 6940
    },
    {
      "entropy": 6.30702543258667,
      "epoch": 7.412172984516818,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013653540086957218,
      "loss": 5.4501,
      "mean_token_accuracy": 0.19412994533777236,
      "num_tokens": 15204059.0,
      "step": 6945
    },
    {
      "entropy": 6.3180159568786625,
      "epoch": 7.417512012813668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001362024174261901,
      "loss": 5.5063,
      "mean_token_accuracy": 0.18488460928201675,
      "num_tokens": 15214435.0,
      "step": 6950
    },
    {
      "entropy": 6.308865594863891,
      "epoch": 7.422851041110518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001358699239973556,
      "loss": 5.4797,
      "mean_token_accuracy": 0.1889624461531639,
      "num_tokens": 15225514.0,
      "step": 6955
    },
    {
      "entropy": 6.288485956192017,
      "epoch": 7.428190069407368,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001355379217569119,
      "loss": 5.373,
      "mean_token_accuracy": 0.19830633103847503,
      "num_tokens": 15236329.0,
      "step": 6960
    },
    {
      "entropy": 6.25889835357666,
      "epoch": 7.433529097704218,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013520641187696834,
      "loss": 5.3872,
      "mean_token_accuracy": 0.19201608002185822,
      "num_tokens": 15246239.0,
      "step": 6965
    },
    {
      "entropy": 6.273760986328125,
      "epoch": 7.438868126001068,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013487539552789585,
      "loss": 5.4801,
      "mean_token_accuracy": 0.19855868369340895,
      "num_tokens": 15258107.0,
      "step": 6970
    },
    {
      "entropy": 6.29157280921936,
      "epoch": 7.444207154297918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013454487387832298,
      "loss": 5.4118,
      "mean_token_accuracy": 0.18662227392196656,
      "num_tokens": 15269494.0,
      "step": 6975
    },
    {
      "entropy": 6.361139106750488,
      "epoch": 7.4495461825947675,
      "grad_norm": 1.25,
      "learning_rate": 0.00013421484809513195,
      "loss": 5.5343,
      "mean_token_accuracy": 0.17982688099145888,
      "num_tokens": 15280847.0,
      "step": 6980
    },
    {
      "entropy": 6.348047399520874,
      "epoch": 7.454885210891618,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013388531934345417,
      "loss": 5.4806,
      "mean_token_accuracy": 0.18755375295877458,
      "num_tokens": 15291602.0,
      "step": 6985
    },
    {
      "entropy": 6.304960441589356,
      "epoch": 7.460224239188467,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013355628878666654,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1935595080256462,
      "num_tokens": 15302346.0,
      "step": 6990
    },
    {
      "entropy": 6.280840587615967,
      "epoch": 7.465563267485318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013322775758638675,
      "loss": 5.4369,
      "mean_token_accuracy": 0.19602899849414826,
      "num_tokens": 15313721.0,
      "step": 6995
    },
    {
      "entropy": 6.242779636383057,
      "epoch": 7.470902295782167,
      "grad_norm": 1.25,
      "learning_rate": 0.0001328997269024701,
      "loss": 5.4481,
      "mean_token_accuracy": 0.1883457660675049,
      "num_tokens": 15325088.0,
      "step": 7000
    },
    {
      "epoch": 7.470902295782167,
      "eval_entropy": 6.113379762079808,
      "eval_loss": 6.547084331512451,
      "eval_mean_token_accuracy": 0.13488195277266687,
      "eval_num_tokens": 15325088.0,
      "eval_runtime": 1.8116,
      "eval_samples_per_second": 1700.127,
      "eval_steps_per_second": 212.516,
      "step": 7000
    },
    {
      "entropy": 6.287962102890015,
      "epoch": 7.476241324079018,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013257219789300417,
      "loss": 5.48,
      "mean_token_accuracy": 0.19072551280260086,
      "num_tokens": 15335527.0,
      "step": 7005
    },
    {
      "entropy": 6.238534307479858,
      "epoch": 7.481580352375867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000132245171714306,
      "loss": 5.3978,
      "mean_token_accuracy": 0.19542142897844314,
      "num_tokens": 15347574.0,
      "step": 7010
    },
    {
      "entropy": 6.273657131195068,
      "epoch": 7.486919380672718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00013191864952091727,
      "loss": 5.4766,
      "mean_token_accuracy": 0.1871364861726761,
      "num_tokens": 15358778.0,
      "step": 7015
    },
    {
      "entropy": 6.32083649635315,
      "epoch": 7.492258408969567,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013159263246560008,
      "loss": 5.5285,
      "mean_token_accuracy": 0.18418796807527543,
      "num_tokens": 15369696.0,
      "step": 7020
    },
    {
      "entropy": 6.350054740905762,
      "epoch": 7.497597437266418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013126712169933362,
      "loss": 5.4787,
      "mean_token_accuracy": 0.19281676262617112,
      "num_tokens": 15380848.0,
      "step": 7025
    },
    {
      "entropy": 6.338539266586304,
      "epoch": 7.502936465563267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001309421183713093,
      "loss": 5.489,
      "mean_token_accuracy": 0.18487109690904618,
      "num_tokens": 15390866.0,
      "step": 7030
    },
    {
      "entropy": 6.299791240692139,
      "epoch": 7.5082754938601175,
      "grad_norm": 1.125,
      "learning_rate": 0.00013061762362892742,
      "loss": 5.5355,
      "mean_token_accuracy": 0.18896956443786622,
      "num_tokens": 15402827.0,
      "step": 7035
    },
    {
      "entropy": 6.263814258575439,
      "epoch": 7.513614522156967,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013029363861779224,
      "loss": 5.3697,
      "mean_token_accuracy": 0.19655501693487168,
      "num_tokens": 15413590.0,
      "step": 7040
    },
    {
      "entropy": 6.30505633354187,
      "epoch": 7.518953550453817,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012997016448170907,
      "loss": 5.5179,
      "mean_token_accuracy": 0.1944273218512535,
      "num_tokens": 15424626.0,
      "step": 7045
    },
    {
      "entropy": 6.302839183807373,
      "epoch": 7.524292578750668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000129647202362679,
      "loss": 5.5083,
      "mean_token_accuracy": 0.18509242087602615,
      "num_tokens": 15435369.0,
      "step": 7050
    },
    {
      "entropy": 6.266674900054932,
      "epoch": 7.529631607047517,
      "grad_norm": 1.171875,
      "learning_rate": 0.00012932475340089606,
      "loss": 5.3647,
      "mean_token_accuracy": 0.1904921844601631,
      "num_tokens": 15446093.0,
      "step": 7055
    },
    {
      "entropy": 6.313183116912842,
      "epoch": 7.534970635344367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00012900281873474214,
      "loss": 5.3895,
      "mean_token_accuracy": 0.19382342249155043,
      "num_tokens": 15458535.0,
      "step": 7060
    },
    {
      "entropy": 6.348286533355713,
      "epoch": 7.540309663641217,
      "grad_norm": 1.1875,
      "learning_rate": 0.00012868139950078376,
      "loss": 5.4868,
      "mean_token_accuracy": 0.1908288910984993,
      "num_tokens": 15469288.0,
      "step": 7065
    },
    {
      "entropy": 6.249121379852295,
      "epoch": 7.545648691938068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001283604968337676,
      "loss": 5.4569,
      "mean_token_accuracy": 0.18953004479408264,
      "num_tokens": 15479643.0,
      "step": 7070
    },
    {
      "entropy": 6.264892864227295,
      "epoch": 7.550987720234917,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00012804011186661664,
      "loss": 5.4565,
      "mean_token_accuracy": 0.18462781757116317,
      "num_tokens": 15491012.0,
      "step": 7075
    },
    {
      "entropy": 6.2518665313720705,
      "epoch": 7.556326748531768,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00012772024573042625,
      "loss": 5.4577,
      "mean_token_accuracy": 0.1880313828587532,
      "num_tokens": 15501621.0,
      "step": 7080
    },
    {
      "entropy": 6.325869989395142,
      "epoch": 7.561665776828617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00012740089955445993,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19207198768854142,
      "num_tokens": 15512463.0,
      "step": 7085
    },
    {
      "entropy": 6.317035770416259,
      "epoch": 7.5670048051254675,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012708207446614577,
      "loss": 5.417,
      "mean_token_accuracy": 0.19555209428071976,
      "num_tokens": 15522141.0,
      "step": 7090
    },
    {
      "entropy": 6.288141679763794,
      "epoch": 7.572343833422317,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012676377159107194,
      "loss": 5.5003,
      "mean_token_accuracy": 0.18633041232824327,
      "num_tokens": 15532332.0,
      "step": 7095
    },
    {
      "entropy": 6.313205575942993,
      "epoch": 7.577682861719167,
      "grad_norm": 1.171875,
      "learning_rate": 0.00012644599205298323,
      "loss": 5.5188,
      "mean_token_accuracy": 0.19106141477823257,
      "num_tokens": 15542157.0,
      "step": 7100
    },
    {
      "entropy": 6.186431550979615,
      "epoch": 7.583021890016017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00012612873697377645,
      "loss": 5.3858,
      "mean_token_accuracy": 0.19531311988830566,
      "num_tokens": 15554015.0,
      "step": 7105
    },
    {
      "entropy": 6.259077882766723,
      "epoch": 7.588360918312867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00012581200747349713,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19319947063922882,
      "num_tokens": 15565134.0,
      "step": 7110
    },
    {
      "entropy": 6.28724513053894,
      "epoch": 7.593699946609717,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001254958046703354,
      "loss": 5.4835,
      "mean_token_accuracy": 0.18521590232849122,
      "num_tokens": 15575570.0,
      "step": 7115
    },
    {
      "entropy": 6.304519891738892,
      "epoch": 7.599038974906567,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001251801296806215,
      "loss": 5.452,
      "mean_token_accuracy": 0.1922187700867653,
      "num_tokens": 15586350.0,
      "step": 7120
    },
    {
      "entropy": 6.311883592605591,
      "epoch": 7.604378003203417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00012486498361882276,
      "loss": 5.4631,
      "mean_token_accuracy": 0.1864357113838196,
      "num_tokens": 15597064.0,
      "step": 7125
    },
    {
      "entropy": 6.283797788619995,
      "epoch": 7.609717031500267,
      "grad_norm": 1.078125,
      "learning_rate": 0.00012455036759753872,
      "loss": 5.4201,
      "mean_token_accuracy": 0.19565890729427338,
      "num_tokens": 15608424.0,
      "step": 7130
    },
    {
      "entropy": 6.309296417236328,
      "epoch": 7.615056059797117,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001242362827274981,
      "loss": 5.4694,
      "mean_token_accuracy": 0.18753717243671417,
      "num_tokens": 15619193.0,
      "step": 7135
    },
    {
      "entropy": 6.307245063781738,
      "epoch": 7.620395088093967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00012392273011755387,
      "loss": 5.4772,
      "mean_token_accuracy": 0.18487688452005385,
      "num_tokens": 15630391.0,
      "step": 7140
    },
    {
      "entropy": 6.3016722202301025,
      "epoch": 7.6257341163908166,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012360971087468044,
      "loss": 5.4741,
      "mean_token_accuracy": 0.18722636699676515,
      "num_tokens": 15641098.0,
      "step": 7145
    },
    {
      "entropy": 6.349103164672852,
      "epoch": 7.631073144687667,
      "grad_norm": 1.1875,
      "learning_rate": 0.00012329722610396885,
      "loss": 5.5289,
      "mean_token_accuracy": 0.17972193509340287,
      "num_tokens": 15652677.0,
      "step": 7150
    },
    {
      "entropy": 6.281785202026367,
      "epoch": 7.6364121729845165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012298527690862346,
      "loss": 5.3607,
      "mean_token_accuracy": 0.20771307796239852,
      "num_tokens": 15664161.0,
      "step": 7155
    },
    {
      "entropy": 6.28417181968689,
      "epoch": 7.641751201281367,
      "grad_norm": 1.25,
      "learning_rate": 0.00012267386438995762,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1937227040529251,
      "num_tokens": 15674663.0,
      "step": 7160
    },
    {
      "entropy": 6.308889245986938,
      "epoch": 7.647090229578216,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012236298964739018,
      "loss": 5.4676,
      "mean_token_accuracy": 0.18741294145584106,
      "num_tokens": 15684798.0,
      "step": 7165
    },
    {
      "entropy": 6.325117874145508,
      "epoch": 7.652429257875067,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012205265377844127,
      "loss": 5.5007,
      "mean_token_accuracy": 0.1891069769859314,
      "num_tokens": 15694796.0,
      "step": 7170
    },
    {
      "entropy": 6.306446743011475,
      "epoch": 7.657768286171917,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001217428578787285,
      "loss": 5.5164,
      "mean_token_accuracy": 0.1903630942106247,
      "num_tokens": 15705325.0,
      "step": 7175
    },
    {
      "entropy": 6.331416177749634,
      "epoch": 7.663107314468767,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012143360304196345,
      "loss": 5.4456,
      "mean_token_accuracy": 0.19512286633253098,
      "num_tokens": 15716106.0,
      "step": 7180
    },
    {
      "entropy": 6.254957675933838,
      "epoch": 7.668446342765616,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012112489035994715,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1882454499602318,
      "num_tokens": 15726896.0,
      "step": 7185
    },
    {
      "entropy": 6.280194568634033,
      "epoch": 7.673785371062467,
      "grad_norm": 1.15625,
      "learning_rate": 0.00012081672092256693,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1862812101840973,
      "num_tokens": 15737492.0,
      "step": 7190
    },
    {
      "entropy": 6.3121009349823,
      "epoch": 7.679124399359317,
      "grad_norm": 1.171875,
      "learning_rate": 0.00012050909581779194,
      "loss": 5.4492,
      "mean_token_accuracy": 0.18987040221691132,
      "num_tokens": 15748941.0,
      "step": 7195
    },
    {
      "entropy": 6.290960836410522,
      "epoch": 7.6844634276561665,
      "grad_norm": 1.0625,
      "learning_rate": 0.00012020201613166996,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1893820658326149,
      "num_tokens": 15760364.0,
      "step": 7200
    },
    {
      "entropy": 6.273015642166138,
      "epoch": 7.689802455953017,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001198954829483227,
      "loss": 5.4886,
      "mean_token_accuracy": 0.19192273169755936,
      "num_tokens": 15770827.0,
      "step": 7205
    },
    {
      "entropy": 6.291214561462402,
      "epoch": 7.695141484249866,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011958949734994294,
      "loss": 5.4105,
      "mean_token_accuracy": 0.19716935455799103,
      "num_tokens": 15781685.0,
      "step": 7210
    },
    {
      "entropy": 6.3456682682037355,
      "epoch": 7.700480512546717,
      "grad_norm": 1.078125,
      "learning_rate": 0.00011928406041679016,
      "loss": 5.5194,
      "mean_token_accuracy": 0.18710066080093385,
      "num_tokens": 15793467.0,
      "step": 7215
    },
    {
      "entropy": 6.278537702560425,
      "epoch": 7.705819540843566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011897917322718658,
      "loss": 5.4711,
      "mean_token_accuracy": 0.18923619240522385,
      "num_tokens": 15804030.0,
      "step": 7220
    },
    {
      "entropy": 6.326219749450684,
      "epoch": 7.711158569140417,
      "grad_norm": 1.171875,
      "learning_rate": 0.00011867483685751393,
      "loss": 5.4758,
      "mean_token_accuracy": 0.19105219542980195,
      "num_tokens": 15815834.0,
      "step": 7225
    },
    {
      "entropy": 6.304016733169556,
      "epoch": 7.716497597437266,
      "grad_norm": 1.0625,
      "learning_rate": 0.00011837105238220905,
      "loss": 5.4935,
      "mean_token_accuracy": 0.18709275275468826,
      "num_tokens": 15828605.0,
      "step": 7230
    },
    {
      "entropy": 6.295789098739624,
      "epoch": 7.721836625734117,
      "grad_norm": 1.25,
      "learning_rate": 0.00011806782087376044,
      "loss": 5.4023,
      "mean_token_accuracy": 0.1915412038564682,
      "num_tokens": 15838820.0,
      "step": 7235
    },
    {
      "entropy": 6.2806047916412355,
      "epoch": 7.727175654030966,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011776514340270435,
      "loss": 5.471,
      "mean_token_accuracy": 0.19484994560480118,
      "num_tokens": 15850425.0,
      "step": 7240
    },
    {
      "entropy": 6.332225275039673,
      "epoch": 7.732514682327817,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001174630210376212,
      "loss": 5.5353,
      "mean_token_accuracy": 0.18361962884664534,
      "num_tokens": 15861175.0,
      "step": 7245
    },
    {
      "entropy": 6.294142627716065,
      "epoch": 7.737853710624666,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011716145484513142,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18699735403060913,
      "num_tokens": 15871425.0,
      "step": 7250
    },
    {
      "entropy": 6.280290603637695,
      "epoch": 7.7431927389215165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011686044588989222,
      "loss": 5.3874,
      "mean_token_accuracy": 0.20123666226863862,
      "num_tokens": 15881984.0,
      "step": 7255
    },
    {
      "entropy": 6.221923875808716,
      "epoch": 7.748531767218366,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00011655999523459311,
      "loss": 5.4414,
      "mean_token_accuracy": 0.1885385051369667,
      "num_tokens": 15891940.0,
      "step": 7260
    },
    {
      "entropy": 6.3203362941741945,
      "epoch": 7.753870795515216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011626010393995301,
      "loss": 5.5108,
      "mean_token_accuracy": 0.1837375894188881,
      "num_tokens": 15902899.0,
      "step": 7265
    },
    {
      "entropy": 6.2765544891357425,
      "epoch": 7.759209823812066,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011596077306471577,
      "loss": 5.4318,
      "mean_token_accuracy": 0.193292735517025,
      "num_tokens": 15913022.0,
      "step": 7270
    },
    {
      "entropy": 6.257391500473022,
      "epoch": 7.764548852108916,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011566200366564665,
      "loss": 5.4287,
      "mean_token_accuracy": 0.19449929893016815,
      "num_tokens": 15924046.0,
      "step": 7275
    },
    {
      "entropy": 6.305574607849121,
      "epoch": 7.769887880405766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00011536379679752901,
      "loss": 5.5019,
      "mean_token_accuracy": 0.18431514203548433,
      "num_tokens": 15935993.0,
      "step": 7280
    },
    {
      "entropy": 6.27879376411438,
      "epoch": 7.775226908702616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011506615351315989,
      "loss": 5.52,
      "mean_token_accuracy": 0.18554560840129852,
      "num_tokens": 15946183.0,
      "step": 7285
    },
    {
      "entropy": 6.297955513000488,
      "epoch": 7.780565936999466,
      "grad_norm": 1.203125,
      "learning_rate": 0.00011476907486334688,
      "loss": 5.4348,
      "mean_token_accuracy": 0.18957224488258362,
      "num_tokens": 15956010.0,
      "step": 7290
    },
    {
      "entropy": 6.317027902603149,
      "epoch": 7.785904965296316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011447256189690404,
      "loss": 5.4409,
      "mean_token_accuracy": 0.1914788231253624,
      "num_tokens": 15966747.0,
      "step": 7295
    },
    {
      "entropy": 6.245621633529663,
      "epoch": 7.791243993593166,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011417661566064834,
      "loss": 5.4596,
      "mean_token_accuracy": 0.1893852472305298,
      "num_tokens": 15976551.0,
      "step": 7300
    },
    {
      "entropy": 6.285864305496216,
      "epoch": 7.796583021890016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011388123719939586,
      "loss": 5.4493,
      "mean_token_accuracy": 0.1863636240363121,
      "num_tokens": 15987074.0,
      "step": 7305
    },
    {
      "entropy": 6.305343341827393,
      "epoch": 7.801922050186866,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011358642755595853,
      "loss": 5.4769,
      "mean_token_accuracy": 0.1860616162419319,
      "num_tokens": 15997180.0,
      "step": 7310
    },
    {
      "entropy": 6.33305025100708,
      "epoch": 7.807261078483716,
      "grad_norm": 1.109375,
      "learning_rate": 0.00011329218777113965,
      "loss": 5.4329,
      "mean_token_accuracy": 0.1920934110879898,
      "num_tokens": 16008653.0,
      "step": 7315
    },
    {
      "entropy": 6.348929166793823,
      "epoch": 7.812600106780566,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011299851888373102,
      "loss": 5.5976,
      "mean_token_accuracy": 0.18123090267181396,
      "num_tokens": 16020561.0,
      "step": 7320
    },
    {
      "entropy": 6.2745198726654055,
      "epoch": 7.817939135077416,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011270542193050888,
      "loss": 5.4626,
      "mean_token_accuracy": 0.19890002608299256,
      "num_tokens": 16031456.0,
      "step": 7325
    },
    {
      "entropy": 6.341566324234009,
      "epoch": 7.823278163374265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011241289794623018,
      "loss": 5.5555,
      "mean_token_accuracy": 0.1821800947189331,
      "num_tokens": 16042757.0,
      "step": 7330
    },
    {
      "entropy": 6.235800075531006,
      "epoch": 7.828617191671116,
      "grad_norm": 1.234375,
      "learning_rate": 0.00011212094796362907,
      "loss": 5.3304,
      "mean_token_accuracy": 0.19140418320894242,
      "num_tokens": 16053022.0,
      "step": 7335
    },
    {
      "entropy": 6.276721525192261,
      "epoch": 7.833956219967966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011182957301341321,
      "loss": 5.4054,
      "mean_token_accuracy": 0.18919897228479385,
      "num_tokens": 16063959.0,
      "step": 7340
    },
    {
      "entropy": 6.3329778671264645,
      "epoch": 7.839295248264816,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011153877412426034,
      "loss": 5.4952,
      "mean_token_accuracy": 0.18930234760046005,
      "num_tokens": 16074402.0,
      "step": 7345
    },
    {
      "entropy": 6.211783933639526,
      "epoch": 7.844634276561666,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001112485523228142,
      "loss": 5.371,
      "mean_token_accuracy": 0.1979040816426277,
      "num_tokens": 16085596.0,
      "step": 7350
    },
    {
      "entropy": 6.2267842292785645,
      "epoch": 7.849973304858516,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001109589086336814,
      "loss": 5.44,
      "mean_token_accuracy": 0.19745024889707566,
      "num_tokens": 16095625.0,
      "step": 7355
    },
    {
      "entropy": 6.285884141921997,
      "epoch": 7.855312333155366,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011066984407942737,
      "loss": 5.419,
      "mean_token_accuracy": 0.18915786296129228,
      "num_tokens": 16107750.0,
      "step": 7360
    },
    {
      "entropy": 6.269041252136231,
      "epoch": 7.8606513614522155,
      "grad_norm": 1.1875,
      "learning_rate": 0.000110381359680573,
      "loss": 5.4201,
      "mean_token_accuracy": 0.19418328255414963,
      "num_tokens": 16118652.0,
      "step": 7365
    },
    {
      "entropy": 6.27239875793457,
      "epoch": 7.865990389749066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011009345645559113,
      "loss": 5.4206,
      "mean_token_accuracy": 0.1961502403020859,
      "num_tokens": 16129598.0,
      "step": 7370
    },
    {
      "entropy": 6.298056602478027,
      "epoch": 7.8713294180459155,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010980613542090256,
      "loss": 5.4407,
      "mean_token_accuracy": 0.18880777657032013,
      "num_tokens": 16140844.0,
      "step": 7375
    },
    {
      "entropy": 6.235966157913208,
      "epoch": 7.876668446342766,
      "grad_norm": 1.125,
      "learning_rate": 0.00010951939759087298,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1888560727238655,
      "num_tokens": 16150922.0,
      "step": 7380
    },
    {
      "entropy": 6.307231903076172,
      "epoch": 7.882007474639615,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010923324397780887,
      "loss": 5.4756,
      "mean_token_accuracy": 0.19293616265058516,
      "num_tokens": 16162440.0,
      "step": 7385
    },
    {
      "entropy": 6.284047031402588,
      "epoch": 7.887346502936466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001089476755919544,
      "loss": 5.4599,
      "mean_token_accuracy": 0.19319476932287216,
      "num_tokens": 16173475.0,
      "step": 7390
    },
    {
      "entropy": 6.298360204696655,
      "epoch": 7.892685531233315,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010866269344148746,
      "loss": 5.5093,
      "mean_token_accuracy": 0.18962480425834655,
      "num_tokens": 16185668.0,
      "step": 7395
    },
    {
      "entropy": 6.305071783065796,
      "epoch": 7.898024559530166,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010837829853251636,
      "loss": 5.4894,
      "mean_token_accuracy": 0.18193406611680984,
      "num_tokens": 16198027.0,
      "step": 7400
    },
    {
      "entropy": 6.3255843162536625,
      "epoch": 7.903363587827015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010809449186907611,
      "loss": 5.5162,
      "mean_token_accuracy": 0.18655869215726853,
      "num_tokens": 16209601.0,
      "step": 7405
    },
    {
      "entropy": 6.297071886062622,
      "epoch": 7.908702616123866,
      "grad_norm": 1.125,
      "learning_rate": 0.00010781127445312516,
      "loss": 5.3818,
      "mean_token_accuracy": 0.2019501507282257,
      "num_tokens": 16221300.0,
      "step": 7410
    },
    {
      "entropy": 6.289458560943603,
      "epoch": 7.914041644420715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00010752864728454143,
      "loss": 5.4794,
      "mean_token_accuracy": 0.18796940445899962,
      "num_tokens": 16232923.0,
      "step": 7415
    },
    {
      "entropy": 6.289797735214234,
      "epoch": 7.9193806727175655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010724661136111918,
      "loss": 5.4568,
      "mean_token_accuracy": 0.18568747937679292,
      "num_tokens": 16243373.0,
      "step": 7420
    },
    {
      "entropy": 6.255471897125244,
      "epoch": 7.924719701014415,
      "grad_norm": 1.25,
      "learning_rate": 0.00010696516767856545,
      "loss": 5.4383,
      "mean_token_accuracy": 0.186435666680336,
      "num_tokens": 16254716.0,
      "step": 7425
    },
    {
      "entropy": 6.313004779815674,
      "epoch": 7.930058729311265,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001066843172304959,
      "loss": 5.5306,
      "mean_token_accuracy": 0.18111457228660582,
      "num_tokens": 16266174.0,
      "step": 7430
    },
    {
      "entropy": 6.270875453948975,
      "epoch": 7.935397757608115,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010640406100843242,
      "loss": 5.3602,
      "mean_token_accuracy": 0.19346348643302919,
      "num_tokens": 16275927.0,
      "step": 7435
    },
    {
      "entropy": 6.3107868194580075,
      "epoch": 7.940736785904965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010612440000179861,
      "loss": 5.5281,
      "mean_token_accuracy": 0.18930120766162872,
      "num_tokens": 16287518.0,
      "step": 7440
    },
    {
      "entropy": 6.290305042266846,
      "epoch": 7.946075814201816,
      "grad_norm": 1.1875,
      "learning_rate": 0.000105845335197917,
      "loss": 5.4716,
      "mean_token_accuracy": 0.18532466441392897,
      "num_tokens": 16298201.0,
      "step": 7445
    },
    {
      "entropy": 6.315746784210205,
      "epoch": 7.951414842498665,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010556686758200499,
      "loss": 5.4838,
      "mean_token_accuracy": 0.1856360211968422,
      "num_tokens": 16309768.0,
      "step": 7450
    },
    {
      "entropy": 6.258326768875122,
      "epoch": 7.956753870795515,
      "grad_norm": 1.171875,
      "learning_rate": 0.00010528899813717194,
      "loss": 5.5407,
      "mean_token_accuracy": 0.18397447764873504,
      "num_tokens": 16320837.0,
      "step": 7455
    },
    {
      "entropy": 6.266445732116699,
      "epoch": 7.962092899092365,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010501172784441522,
      "loss": 5.4617,
      "mean_token_accuracy": 0.18468159288167954,
      "num_tokens": 16332729.0,
      "step": 7460
    },
    {
      "entropy": 6.273439836502075,
      "epoch": 7.967431927389216,
      "grad_norm": 1.125,
      "learning_rate": 0.00010473505768261691,
      "loss": 5.4136,
      "mean_token_accuracy": 0.18490265756845475,
      "num_tokens": 16343444.0,
      "step": 7465
    },
    {
      "entropy": 6.235618734359742,
      "epoch": 7.972770955686065,
      "grad_norm": 1.25,
      "learning_rate": 0.00010445898862854056,
      "loss": 5.4047,
      "mean_token_accuracy": 0.19600029289722443,
      "num_tokens": 16354452.0,
      "step": 7470
    },
    {
      "entropy": 6.254809045791626,
      "epoch": 7.9781099839829155,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010418352165682737,
      "loss": 5.4354,
      "mean_token_accuracy": 0.19964061975479125,
      "num_tokens": 16365276.0,
      "step": 7475
    },
    {
      "entropy": 6.267963027954101,
      "epoch": 7.983449012279765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00010390865773999305,
      "loss": 5.4557,
      "mean_token_accuracy": 0.19333289116621016,
      "num_tokens": 16376176.0,
      "step": 7480
    },
    {
      "entropy": 6.220429754257202,
      "epoch": 7.988788040576615,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010363439784842419,
      "loss": 5.4219,
      "mean_token_accuracy": 0.18904615938663483,
      "num_tokens": 16386610.0,
      "step": 7485
    },
    {
      "entropy": 6.231282806396484,
      "epoch": 7.994127068873465,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010336074295037506,
      "loss": 5.376,
      "mean_token_accuracy": 0.20314324349164964,
      "num_tokens": 16396838.0,
      "step": 7490
    },
    {
      "entropy": 6.271087694168091,
      "epoch": 7.999466097170315,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010308769401196371,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18455882072448732,
      "num_tokens": 16407344.0,
      "step": 7495
    },
    {
      "entropy": 6.314283529917399,
      "epoch": 8.00427122263748,
      "grad_norm": 1.21875,
      "learning_rate": 0.00010281525199716934,
      "loss": 5.3818,
      "mean_token_accuracy": 0.1875956224070655,
      "num_tokens": 16416439.0,
      "step": 7500
    },
    {
      "epoch": 8.00427122263748,
      "eval_entropy": 6.136151521856134,
      "eval_loss": 6.5423078536987305,
      "eval_mean_token_accuracy": 0.13510648057638824,
      "eval_num_tokens": 16416439.0,
      "eval_runtime": 1.8109,
      "eval_samples_per_second": 1700.848,
      "eval_steps_per_second": 212.606,
      "step": 7500
    },
    {
      "entropy": 6.285388946533203,
      "epoch": 8.00961025093433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00010254341786782806,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19184360802173614,
      "num_tokens": 16427682.0,
      "step": 7505
    },
    {
      "entropy": 6.251141166687011,
      "epoch": 8.01494927923118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00010227219258363021,
      "loss": 5.3525,
      "mean_token_accuracy": 0.1970635950565338,
      "num_tokens": 16438338.0,
      "step": 7510
    },
    {
      "entropy": 6.2690081119537355,
      "epoch": 8.02028830752803,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010200157710211637,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20117098987102508,
      "num_tokens": 16449151.0,
      "step": 7515
    },
    {
      "entropy": 6.241944169998169,
      "epoch": 8.025627335824879,
      "grad_norm": 1.25,
      "learning_rate": 0.00010173157237867444,
      "loss": 5.3415,
      "mean_token_accuracy": 0.19427326321601868,
      "num_tokens": 16460160.0,
      "step": 7520
    },
    {
      "entropy": 6.247269773483277,
      "epoch": 8.03096636412173,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010146217936653614,
      "loss": 5.3315,
      "mean_token_accuracy": 0.20295249819755554,
      "num_tokens": 16470602.0,
      "step": 7525
    },
    {
      "entropy": 6.3116706848144535,
      "epoch": 8.03630539241858,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001011933990167732,
      "loss": 5.3299,
      "mean_token_accuracy": 0.19919252395629883,
      "num_tokens": 16481086.0,
      "step": 7530
    },
    {
      "entropy": 6.287974166870117,
      "epoch": 8.04164442071543,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00010092523227829492,
      "loss": 5.3898,
      "mean_token_accuracy": 0.1971887692809105,
      "num_tokens": 16492012.0,
      "step": 7535
    },
    {
      "entropy": 6.300996112823486,
      "epoch": 8.046983449012279,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010065768009784381,
      "loss": 5.447,
      "mean_token_accuracy": 0.1879270300269127,
      "num_tokens": 16502793.0,
      "step": 7540
    },
    {
      "entropy": 6.330128574371338,
      "epoch": 8.05232247730913,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001003907434199932,
      "loss": 5.4588,
      "mean_token_accuracy": 0.19208128452301027,
      "num_tokens": 16512457.0,
      "step": 7545
    },
    {
      "entropy": 6.296833324432373,
      "epoch": 8.05766150560598,
      "grad_norm": 1.40625,
      "learning_rate": 0.000100124423187143,
      "loss": 5.3995,
      "mean_token_accuracy": 0.19723411351442338,
      "num_tokens": 16522859.0,
      "step": 7550
    },
    {
      "entropy": 6.311189317703247,
      "epoch": 8.06300053390283,
      "grad_norm": 1.1953125,
      "learning_rate": 9.985872033951708e-05,
      "loss": 5.4119,
      "mean_token_accuracy": 0.195047065615654,
      "num_tokens": 16533755.0,
      "step": 7555
    },
    {
      "entropy": 6.288791131973267,
      "epoch": 8.06833956219968,
      "grad_norm": 1.2578125,
      "learning_rate": 9.95936358151596e-05,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19499182999134063,
      "num_tokens": 16544928.0,
      "step": 7560
    },
    {
      "entropy": 6.2984113693237305,
      "epoch": 8.07367859049653,
      "grad_norm": 1.2109375,
      "learning_rate": 9.932917054993168e-05,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20289150178432463,
      "num_tokens": 16554730.0,
      "step": 7565
    },
    {
      "entropy": 6.211780023574829,
      "epoch": 8.07901761879338,
      "grad_norm": 1.2578125,
      "learning_rate": 9.906532547750839e-05,
      "loss": 5.3808,
      "mean_token_accuracy": 0.19330030232667922,
      "num_tokens": 16565374.0,
      "step": 7570
    },
    {
      "entropy": 6.291947889328003,
      "epoch": 8.08435664709023,
      "grad_norm": 1.09375,
      "learning_rate": 9.880210152937502e-05,
      "loss": 5.373,
      "mean_token_accuracy": 0.19923413395881653,
      "num_tokens": 16576477.0,
      "step": 7575
    },
    {
      "entropy": 6.276288414001465,
      "epoch": 8.08969567538708,
      "grad_norm": 1.1171875,
      "learning_rate": 9.85394996348243e-05,
      "loss": 5.3712,
      "mean_token_accuracy": 0.20000862181186677,
      "num_tokens": 16587675.0,
      "step": 7580
    },
    {
      "entropy": 6.247407913208008,
      "epoch": 8.095034703683929,
      "grad_norm": 1.046875,
      "learning_rate": 9.827752072095258e-05,
      "loss": 5.2777,
      "mean_token_accuracy": 0.2046089142560959,
      "num_tokens": 16600895.0,
      "step": 7585
    },
    {
      "entropy": 6.35636796951294,
      "epoch": 8.10037373198078,
      "grad_norm": 1.1484375,
      "learning_rate": 9.801616571265709e-05,
      "loss": 5.4592,
      "mean_token_accuracy": 0.19454747587442398,
      "num_tokens": 16612318.0,
      "step": 7590
    },
    {
      "entropy": 6.331305074691772,
      "epoch": 8.10571276027763,
      "grad_norm": 1.1796875,
      "learning_rate": 9.775543553263203e-05,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18617027550935744,
      "num_tokens": 16623490.0,
      "step": 7595
    },
    {
      "entropy": 6.314173412322998,
      "epoch": 8.11105178857448,
      "grad_norm": 1.140625,
      "learning_rate": 9.74953311013661e-05,
      "loss": 5.4335,
      "mean_token_accuracy": 0.19303737729787826,
      "num_tokens": 16635378.0,
      "step": 7600
    },
    {
      "entropy": 6.272602128982544,
      "epoch": 8.116390816871329,
      "grad_norm": 1.2578125,
      "learning_rate": 9.723585333713853e-05,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1940835028886795,
      "num_tokens": 16645073.0,
      "step": 7605
    },
    {
      "entropy": 6.308458137512207,
      "epoch": 8.12172984516818,
      "grad_norm": 1.171875,
      "learning_rate": 9.697700315601643e-05,
      "loss": 5.403,
      "mean_token_accuracy": 0.19990355968475343,
      "num_tokens": 16655891.0,
      "step": 7610
    },
    {
      "entropy": 6.243864917755127,
      "epoch": 8.12706887346503,
      "grad_norm": 1.203125,
      "learning_rate": 9.671878147185097e-05,
      "loss": 5.3733,
      "mean_token_accuracy": 0.20272928178310395,
      "num_tokens": 16665887.0,
      "step": 7615
    },
    {
      "entropy": 6.275272703170776,
      "epoch": 8.13240790176188,
      "grad_norm": 1.1484375,
      "learning_rate": 9.646118919627466e-05,
      "loss": 5.4103,
      "mean_token_accuracy": 0.19670272767543792,
      "num_tokens": 16677282.0,
      "step": 7620
    },
    {
      "entropy": 6.315062665939331,
      "epoch": 8.137746930058729,
      "grad_norm": 1.09375,
      "learning_rate": 9.620422723869804e-05,
      "loss": 5.4226,
      "mean_token_accuracy": 0.1900483101606369,
      "num_tokens": 16688907.0,
      "step": 7625
    },
    {
      "entropy": 6.273891258239746,
      "epoch": 8.143085958355579,
      "grad_norm": 1.234375,
      "learning_rate": 9.594789650630595e-05,
      "loss": 5.3772,
      "mean_token_accuracy": 0.19321130514144896,
      "num_tokens": 16698767.0,
      "step": 7630
    },
    {
      "entropy": 6.313991069793701,
      "epoch": 8.14842498665243,
      "grad_norm": 1.265625,
      "learning_rate": 9.569219790405514e-05,
      "loss": 5.3572,
      "mean_token_accuracy": 0.1905400574207306,
      "num_tokens": 16708789.0,
      "step": 7635
    },
    {
      "entropy": 6.297112989425659,
      "epoch": 8.15376401494928,
      "grad_norm": 1.2421875,
      "learning_rate": 9.543713233467044e-05,
      "loss": 5.4322,
      "mean_token_accuracy": 0.19112920761108398,
      "num_tokens": 16718923.0,
      "step": 7640
    },
    {
      "entropy": 6.326007986068726,
      "epoch": 8.159103043246128,
      "grad_norm": 1.1875,
      "learning_rate": 9.518270069864194e-05,
      "loss": 5.5019,
      "mean_token_accuracy": 0.18736330568790435,
      "num_tokens": 16730878.0,
      "step": 7645
    },
    {
      "entropy": 6.2732117652893065,
      "epoch": 8.164442071542979,
      "grad_norm": 1.2421875,
      "learning_rate": 9.492890389422157e-05,
      "loss": 5.3433,
      "mean_token_accuracy": 0.1988297149538994,
      "num_tokens": 16740736.0,
      "step": 7650
    },
    {
      "entropy": 6.253875780105591,
      "epoch": 8.16978109983983,
      "grad_norm": 1.359375,
      "learning_rate": 9.467574281742015e-05,
      "loss": 5.3155,
      "mean_token_accuracy": 0.20268877744674682,
      "num_tokens": 16750294.0,
      "step": 7655
    },
    {
      "entropy": 6.25405969619751,
      "epoch": 8.17512012813668,
      "grad_norm": 1.1484375,
      "learning_rate": 9.442321836200399e-05,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19080522060394287,
      "num_tokens": 16761683.0,
      "step": 7660
    },
    {
      "entropy": 6.2844563007354735,
      "epoch": 8.180459156433528,
      "grad_norm": 1.1953125,
      "learning_rate": 9.417133141949186e-05,
      "loss": 5.4373,
      "mean_token_accuracy": 0.1910292387008667,
      "num_tokens": 16772828.0,
      "step": 7665
    },
    {
      "entropy": 6.286375570297241,
      "epoch": 8.185798184730379,
      "grad_norm": 1.34375,
      "learning_rate": 9.392008287915201e-05,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19190025478601455,
      "num_tokens": 16783853.0,
      "step": 7670
    },
    {
      "entropy": 6.296071195602417,
      "epoch": 8.191137213027229,
      "grad_norm": 1.3515625,
      "learning_rate": 9.366947362799858e-05,
      "loss": 5.3816,
      "mean_token_accuracy": 0.19442843794822692,
      "num_tokens": 16794032.0,
      "step": 7675
    },
    {
      "entropy": 6.2660017013549805,
      "epoch": 8.19647624132408,
      "grad_norm": 1.2265625,
      "learning_rate": 9.341950455078907e-05,
      "loss": 5.4141,
      "mean_token_accuracy": 0.19492938816547395,
      "num_tokens": 16805269.0,
      "step": 7680
    },
    {
      "entropy": 6.236077737808228,
      "epoch": 8.20181526962093,
      "grad_norm": 1.1484375,
      "learning_rate": 9.31701765300206e-05,
      "loss": 5.2736,
      "mean_token_accuracy": 0.2090698704123497,
      "num_tokens": 16817116.0,
      "step": 7685
    },
    {
      "entropy": 6.280970525741577,
      "epoch": 8.207154297917779,
      "grad_norm": 1.2109375,
      "learning_rate": 9.292149044592733e-05,
      "loss": 5.3978,
      "mean_token_accuracy": 0.19769419878721237,
      "num_tokens": 16828492.0,
      "step": 7690
    },
    {
      "entropy": 6.312994480133057,
      "epoch": 8.212493326214629,
      "grad_norm": 1.1171875,
      "learning_rate": 9.267344717647682e-05,
      "loss": 5.4436,
      "mean_token_accuracy": 0.19130415469408035,
      "num_tokens": 16839715.0,
      "step": 7695
    },
    {
      "entropy": 6.257885408401489,
      "epoch": 8.21783235451148,
      "grad_norm": 1.1796875,
      "learning_rate": 9.242604759736759e-05,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19316410273313522,
      "num_tokens": 16850451.0,
      "step": 7700
    },
    {
      "entropy": 6.300156831741333,
      "epoch": 8.22317138280833,
      "grad_norm": 1.1796875,
      "learning_rate": 9.217929258202528e-05,
      "loss": 5.3927,
      "mean_token_accuracy": 0.19539438039064408,
      "num_tokens": 16861046.0,
      "step": 7705
    },
    {
      "entropy": 6.311345863342285,
      "epoch": 8.228510411105178,
      "grad_norm": 1.25,
      "learning_rate": 9.193318300160024e-05,
      "loss": 5.358,
      "mean_token_accuracy": 0.19471683353185654,
      "num_tokens": 16871748.0,
      "step": 7710
    },
    {
      "entropy": 6.294392871856689,
      "epoch": 8.233849439402029,
      "grad_norm": 1.1796875,
      "learning_rate": 9.168771972496404e-05,
      "loss": 5.3503,
      "mean_token_accuracy": 0.19816206097602845,
      "num_tokens": 16883126.0,
      "step": 7715
    },
    {
      "entropy": 6.264479064941407,
      "epoch": 8.23918846769888,
      "grad_norm": 1.203125,
      "learning_rate": 9.144290361870652e-05,
      "loss": 5.3621,
      "mean_token_accuracy": 0.19831522554159164,
      "num_tokens": 16894377.0,
      "step": 7720
    },
    {
      "entropy": 6.309286832809448,
      "epoch": 8.24452749599573,
      "grad_norm": 1.1796875,
      "learning_rate": 9.119873554713274e-05,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19227964729070662,
      "num_tokens": 16905234.0,
      "step": 7725
    },
    {
      "entropy": 6.304486989974976,
      "epoch": 8.249866524292578,
      "grad_norm": 1.265625,
      "learning_rate": 9.095521637225988e-05,
      "loss": 5.3834,
      "mean_token_accuracy": 0.1948908343911171,
      "num_tokens": 16915031.0,
      "step": 7730
    },
    {
      "entropy": 6.334994268417359,
      "epoch": 8.255205552589429,
      "grad_norm": 1.21875,
      "learning_rate": 9.071234695381441e-05,
      "loss": 5.441,
      "mean_token_accuracy": 0.18804772496223449,
      "num_tokens": 16926595.0,
      "step": 7735
    },
    {
      "entropy": 6.2513487339019775,
      "epoch": 8.260544580886279,
      "grad_norm": 1.2421875,
      "learning_rate": 9.04701281492286e-05,
      "loss": 5.3925,
      "mean_token_accuracy": 0.20451612323522567,
      "num_tokens": 16937612.0,
      "step": 7740
    },
    {
      "entropy": 6.253056144714355,
      "epoch": 8.26588360918313,
      "grad_norm": 1.1953125,
      "learning_rate": 9.022856081363806e-05,
      "loss": 5.377,
      "mean_token_accuracy": 0.1940850868821144,
      "num_tokens": 16948734.0,
      "step": 7745
    },
    {
      "entropy": 6.265314435958862,
      "epoch": 8.271222637479978,
      "grad_norm": 1.0546875,
      "learning_rate": 8.998764579987823e-05,
      "loss": 5.3621,
      "mean_token_accuracy": 0.1988656148314476,
      "num_tokens": 16961310.0,
      "step": 7750
    },
    {
      "entropy": 6.3020031452178955,
      "epoch": 8.276561665776828,
      "grad_norm": 1.3515625,
      "learning_rate": 8.974738395848171e-05,
      "loss": 5.3774,
      "mean_token_accuracy": 0.19321369230747223,
      "num_tokens": 16971196.0,
      "step": 7755
    },
    {
      "entropy": 6.257359170913697,
      "epoch": 8.281900694073679,
      "grad_norm": 1.0078125,
      "learning_rate": 8.950777613767508e-05,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19390098601579667,
      "num_tokens": 16983424.0,
      "step": 7760
    },
    {
      "entropy": 6.257221746444702,
      "epoch": 8.28723972237053,
      "grad_norm": 1.234375,
      "learning_rate": 8.926882318337586e-05,
      "loss": 5.301,
      "mean_token_accuracy": 0.19741829335689545,
      "num_tokens": 16994389.0,
      "step": 7765
    },
    {
      "entropy": 6.277281904220581,
      "epoch": 8.292578750667378,
      "grad_norm": 1.2109375,
      "learning_rate": 8.903052593918979e-05,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1859165444970131,
      "num_tokens": 17006239.0,
      "step": 7770
    },
    {
      "entropy": 6.239947509765625,
      "epoch": 8.297917778964228,
      "grad_norm": 1.1953125,
      "learning_rate": 8.879288524640747e-05,
      "loss": 5.3851,
      "mean_token_accuracy": 0.18899611085653306,
      "num_tokens": 17018648.0,
      "step": 7775
    },
    {
      "entropy": 6.286568927764892,
      "epoch": 8.303256807261079,
      "grad_norm": 1.2109375,
      "learning_rate": 8.855590194400182e-05,
      "loss": 5.4784,
      "mean_token_accuracy": 0.18599783927202224,
      "num_tokens": 17028965.0,
      "step": 7780
    },
    {
      "entropy": 6.27871527671814,
      "epoch": 8.308595835557929,
      "grad_norm": 1.3515625,
      "learning_rate": 8.83195768686247e-05,
      "loss": 5.4402,
      "mean_token_accuracy": 0.18707661628723143,
      "num_tokens": 17039432.0,
      "step": 7785
    },
    {
      "entropy": 6.333696699142456,
      "epoch": 8.313934863854778,
      "grad_norm": 1.15625,
      "learning_rate": 8.808391085460424e-05,
      "loss": 5.4866,
      "mean_token_accuracy": 0.19333939999341965,
      "num_tokens": 17050931.0,
      "step": 7790
    },
    {
      "entropy": 6.309759092330933,
      "epoch": 8.319273892151628,
      "grad_norm": 1.1953125,
      "learning_rate": 8.784890473394173e-05,
      "loss": 5.3801,
      "mean_token_accuracy": 0.19778697490692138,
      "num_tokens": 17062665.0,
      "step": 7795
    },
    {
      "entropy": 6.358693599700928,
      "epoch": 8.324612920448478,
      "grad_norm": 1.046875,
      "learning_rate": 8.761455933630891e-05,
      "loss": 5.4792,
      "mean_token_accuracy": 0.18731221705675125,
      "num_tokens": 17074008.0,
      "step": 7800
    },
    {
      "entropy": 6.290230798721313,
      "epoch": 8.329951948745329,
      "grad_norm": 1.203125,
      "learning_rate": 8.738087548904467e-05,
      "loss": 5.426,
      "mean_token_accuracy": 0.18245502561330795,
      "num_tokens": 17084541.0,
      "step": 7805
    },
    {
      "entropy": 6.289392805099487,
      "epoch": 8.335290977042177,
      "grad_norm": 1.2265625,
      "learning_rate": 8.714785401715258e-05,
      "loss": 5.4459,
      "mean_token_accuracy": 0.1914304554462433,
      "num_tokens": 17095094.0,
      "step": 7810
    },
    {
      "entropy": 6.31272611618042,
      "epoch": 8.340630005339028,
      "grad_norm": 1.1640625,
      "learning_rate": 8.691549574329752e-05,
      "loss": 5.4214,
      "mean_token_accuracy": 0.19288700073957443,
      "num_tokens": 17105985.0,
      "step": 7815
    },
    {
      "entropy": 6.292475700378418,
      "epoch": 8.345969033635878,
      "grad_norm": 1.234375,
      "learning_rate": 8.668380148780319e-05,
      "loss": 5.4225,
      "mean_token_accuracy": 0.18336446732282638,
      "num_tokens": 17116016.0,
      "step": 7820
    },
    {
      "entropy": 6.247875785827636,
      "epoch": 8.351308061932729,
      "grad_norm": 1.171875,
      "learning_rate": 8.64527720686489e-05,
      "loss": 5.4281,
      "mean_token_accuracy": 0.19529077261686326,
      "num_tokens": 17126484.0,
      "step": 7825
    },
    {
      "entropy": 6.305013608932495,
      "epoch": 8.356647090229579,
      "grad_norm": 1.25,
      "learning_rate": 8.62224083014668e-05,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19113296419382095,
      "num_tokens": 17138108.0,
      "step": 7830
    },
    {
      "entropy": 6.316189479827881,
      "epoch": 8.361986118526428,
      "grad_norm": 1.1796875,
      "learning_rate": 8.599271099953912e-05,
      "loss": 5.4076,
      "mean_token_accuracy": 0.1995265394449234,
      "num_tokens": 17149224.0,
      "step": 7835
    },
    {
      "entropy": 6.284372425079345,
      "epoch": 8.367325146823278,
      "grad_norm": 1.234375,
      "learning_rate": 8.576368097379502e-05,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1908060133457184,
      "num_tokens": 17161156.0,
      "step": 7840
    },
    {
      "entropy": 6.338185119628906,
      "epoch": 8.372664175120129,
      "grad_norm": 1.3359375,
      "learning_rate": 8.553531903280815e-05,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19008296579122544,
      "num_tokens": 17171528.0,
      "step": 7845
    },
    {
      "entropy": 6.336772871017456,
      "epoch": 8.378003203416979,
      "grad_norm": 1.125,
      "learning_rate": 8.530762598279318e-05,
      "loss": 5.5014,
      "mean_token_accuracy": 0.18084470331668853,
      "num_tokens": 17181784.0,
      "step": 7850
    },
    {
      "entropy": 6.286856937408447,
      "epoch": 8.383342231713828,
      "grad_norm": 1.1171875,
      "learning_rate": 8.508060262760353e-05,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20462595373392106,
      "num_tokens": 17192988.0,
      "step": 7855
    },
    {
      "entropy": 6.279626417160034,
      "epoch": 8.388681260010678,
      "grad_norm": 1.1484375,
      "learning_rate": 8.485424976872835e-05,
      "loss": 5.4212,
      "mean_token_accuracy": 0.18943293690681456,
      "num_tokens": 17203759.0,
      "step": 7860
    },
    {
      "entropy": 6.281919240951538,
      "epoch": 8.394020288307528,
      "grad_norm": 1.2421875,
      "learning_rate": 8.46285682052895e-05,
      "loss": 5.377,
      "mean_token_accuracy": 0.20012870728969573,
      "num_tokens": 17213768.0,
      "step": 7865
    },
    {
      "entropy": 6.262228155136109,
      "epoch": 8.399359316604379,
      "grad_norm": 1.15625,
      "learning_rate": 8.44035587340391e-05,
      "loss": 5.4503,
      "mean_token_accuracy": 0.19287367165088654,
      "num_tokens": 17225550.0,
      "step": 7870
    },
    {
      "entropy": 6.332665014266968,
      "epoch": 8.404698344901227,
      "grad_norm": 1.1796875,
      "learning_rate": 8.417922214935621e-05,
      "loss": 5.462,
      "mean_token_accuracy": 0.18738492280244828,
      "num_tokens": 17235969.0,
      "step": 7875
    },
    {
      "entropy": 6.258744144439698,
      "epoch": 8.410037373198078,
      "grad_norm": 1.2109375,
      "learning_rate": 8.395555924324457e-05,
      "loss": 5.3671,
      "mean_token_accuracy": 0.20065730661153794,
      "num_tokens": 17247529.0,
      "step": 7880
    },
    {
      "entropy": 6.278067016601563,
      "epoch": 8.415376401494928,
      "grad_norm": 1.109375,
      "learning_rate": 8.37325708053294e-05,
      "loss": 5.4302,
      "mean_token_accuracy": 0.19352214634418488,
      "num_tokens": 17258971.0,
      "step": 7885
    },
    {
      "entropy": 6.2095043659210205,
      "epoch": 8.420715429791779,
      "grad_norm": 1.421875,
      "learning_rate": 8.35102576228548e-05,
      "loss": 5.3371,
      "mean_token_accuracy": 0.20132233202457428,
      "num_tokens": 17271281.0,
      "step": 7890
    },
    {
      "entropy": 6.227483892440796,
      "epoch": 8.426054458088627,
      "grad_norm": 1.34375,
      "learning_rate": 8.328862048068089e-05,
      "loss": 5.3125,
      "mean_token_accuracy": 0.19799865186214446,
      "num_tokens": 17281419.0,
      "step": 7895
    },
    {
      "entropy": 6.268981170654297,
      "epoch": 8.431393486385478,
      "grad_norm": 1.078125,
      "learning_rate": 8.306766016128125e-05,
      "loss": 5.37,
      "mean_token_accuracy": 0.19525525271892546,
      "num_tokens": 17292867.0,
      "step": 7900
    },
    {
      "entropy": 6.299656820297241,
      "epoch": 8.436732514682328,
      "grad_norm": 1.3125,
      "learning_rate": 8.284737744473972e-05,
      "loss": 5.426,
      "mean_token_accuracy": 0.19533834755420684,
      "num_tokens": 17303565.0,
      "step": 7905
    },
    {
      "entropy": 6.247980737686158,
      "epoch": 8.442071542979178,
      "grad_norm": 1.2265625,
      "learning_rate": 8.26277731087482e-05,
      "loss": 5.3546,
      "mean_token_accuracy": 0.18990732878446578,
      "num_tokens": 17315107.0,
      "step": 7910
    },
    {
      "entropy": 6.323516082763672,
      "epoch": 8.447410571276027,
      "grad_norm": 1.1796875,
      "learning_rate": 8.240884792860336e-05,
      "loss": 5.4517,
      "mean_token_accuracy": 0.18722765147686005,
      "num_tokens": 17326606.0,
      "step": 7915
    },
    {
      "entropy": 6.2622026920318605,
      "epoch": 8.452749599572877,
      "grad_norm": 1.09375,
      "learning_rate": 8.219060267720427e-05,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18907711058855056,
      "num_tokens": 17337503.0,
      "step": 7920
    },
    {
      "entropy": 6.281539106369019,
      "epoch": 8.458088627869728,
      "grad_norm": 1.25,
      "learning_rate": 8.197303812504968e-05,
      "loss": 5.3781,
      "mean_token_accuracy": 0.19827369153499602,
      "num_tokens": 17347697.0,
      "step": 7925
    },
    {
      "entropy": 6.314615678787232,
      "epoch": 8.463427656166578,
      "grad_norm": 1.15625,
      "learning_rate": 8.175615504023494e-05,
      "loss": 5.5319,
      "mean_token_accuracy": 0.17755035161972046,
      "num_tokens": 17359656.0,
      "step": 7930
    },
    {
      "entropy": 6.286178731918335,
      "epoch": 8.468766684463427,
      "grad_norm": 1.265625,
      "learning_rate": 8.153995418844974e-05,
      "loss": 5.4116,
      "mean_token_accuracy": 0.191170571744442,
      "num_tokens": 17370869.0,
      "step": 7935
    },
    {
      "entropy": 6.290740489959717,
      "epoch": 8.474105712760277,
      "grad_norm": 1.2109375,
      "learning_rate": 8.132443633297508e-05,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20106685310602188,
      "num_tokens": 17380716.0,
      "step": 7940
    },
    {
      "entropy": 6.300388383865356,
      "epoch": 8.479444741057128,
      "grad_norm": 1.2265625,
      "learning_rate": 8.110960223468077e-05,
      "loss": 5.4171,
      "mean_token_accuracy": 0.1983788177371025,
      "num_tokens": 17392012.0,
      "step": 7945
    },
    {
      "entropy": 6.256002759933471,
      "epoch": 8.484783769353978,
      "grad_norm": 1.09375,
      "learning_rate": 8.089545265202258e-05,
      "loss": 5.3892,
      "mean_token_accuracy": 0.19373163133859633,
      "num_tokens": 17403199.0,
      "step": 7950
    },
    {
      "entropy": 6.30955057144165,
      "epoch": 8.490122797650827,
      "grad_norm": 1.21875,
      "learning_rate": 8.068198834103967e-05,
      "loss": 5.4173,
      "mean_token_accuracy": 0.19167587906122208,
      "num_tokens": 17414020.0,
      "step": 7955
    },
    {
      "entropy": 6.248732852935791,
      "epoch": 8.495461825947677,
      "grad_norm": 1.3359375,
      "learning_rate": 8.046921005535198e-05,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19576088786125184,
      "num_tokens": 17425460.0,
      "step": 7960
    },
    {
      "entropy": 6.26835789680481,
      "epoch": 8.500800854244527,
      "grad_norm": 1.1015625,
      "learning_rate": 8.02571185461574e-05,
      "loss": 5.3683,
      "mean_token_accuracy": 0.19737352728843688,
      "num_tokens": 17435819.0,
      "step": 7965
    },
    {
      "entropy": 6.298630714416504,
      "epoch": 8.506139882541378,
      "grad_norm": 1.3125,
      "learning_rate": 8.004571456222928e-05,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1900707334280014,
      "num_tokens": 17446000.0,
      "step": 7970
    },
    {
      "entropy": 6.339548397064209,
      "epoch": 8.511478910838228,
      "grad_norm": 1.1171875,
      "learning_rate": 7.983499884991356e-05,
      "loss": 5.4799,
      "mean_token_accuracy": 0.18726177364587784,
      "num_tokens": 17456986.0,
      "step": 7975
    },
    {
      "entropy": 6.317312383651734,
      "epoch": 8.516817939135077,
      "grad_norm": 1.2734375,
      "learning_rate": 7.962497215312662e-05,
      "loss": 5.4299,
      "mean_token_accuracy": 0.19350540190935134,
      "num_tokens": 17467144.0,
      "step": 7980
    },
    {
      "entropy": 6.330316781997681,
      "epoch": 8.522156967431927,
      "grad_norm": 1.21875,
      "learning_rate": 7.94156352133518e-05,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1904843494296074,
      "num_tokens": 17479243.0,
      "step": 7985
    },
    {
      "entropy": 6.309084367752075,
      "epoch": 8.527495995728778,
      "grad_norm": 1.1875,
      "learning_rate": 7.920698876963784e-05,
      "loss": 5.3797,
      "mean_token_accuracy": 0.19125119149684905,
      "num_tokens": 17490447.0,
      "step": 7990
    },
    {
      "entropy": 6.285092973709107,
      "epoch": 8.532835024025628,
      "grad_norm": 1.2109375,
      "learning_rate": 7.899903355859534e-05,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18608825355768205,
      "num_tokens": 17500464.0,
      "step": 7995
    },
    {
      "entropy": 6.267555618286133,
      "epoch": 8.538174052322477,
      "grad_norm": 1.5546875,
      "learning_rate": 7.879177031439487e-05,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1957908973097801,
      "num_tokens": 17510189.0,
      "step": 8000
    },
    {
      "epoch": 8.538174052322477,
      "eval_entropy": 6.129665725262134,
      "eval_loss": 6.549795627593994,
      "eval_mean_token_accuracy": 0.13511817216292604,
      "eval_num_tokens": 17510189.0,
      "eval_runtime": 1.8063,
      "eval_samples_per_second": 1705.136,
      "eval_steps_per_second": 213.142,
      "step": 8000
    },
    {
      "entropy": 6.260827922821045,
      "epoch": 8.543513080619327,
      "grad_norm": 1.2421875,
      "learning_rate": 7.858519976876375e-05,
      "loss": 5.4092,
      "mean_token_accuracy": 0.19452549070119857,
      "num_tokens": 17521622.0,
      "step": 8005
    },
    {
      "entropy": 6.289570903778076,
      "epoch": 8.548852108916178,
      "grad_norm": 1.265625,
      "learning_rate": 7.837932265098407e-05,
      "loss": 5.3852,
      "mean_token_accuracy": 0.1945295363664627,
      "num_tokens": 17532483.0,
      "step": 8010
    },
    {
      "entropy": 6.254086351394653,
      "epoch": 8.554191137213028,
      "grad_norm": 1.171875,
      "learning_rate": 7.817413968788965e-05,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19627286940813066,
      "num_tokens": 17543402.0,
      "step": 8015
    },
    {
      "entropy": 6.224242401123047,
      "epoch": 8.559530165509877,
      "grad_norm": 1.2265625,
      "learning_rate": 7.796965160386362e-05,
      "loss": 5.3644,
      "mean_token_accuracy": 0.20089459866285325,
      "num_tokens": 17554395.0,
      "step": 8020
    },
    {
      "entropy": 6.309196805953979,
      "epoch": 8.564869193806727,
      "grad_norm": 1.1953125,
      "learning_rate": 7.776585912083605e-05,
      "loss": 5.4384,
      "mean_token_accuracy": 0.1892755374312401,
      "num_tokens": 17566481.0,
      "step": 8025
    },
    {
      "entropy": 6.273343563079834,
      "epoch": 8.570208222103577,
      "grad_norm": 1.3046875,
      "learning_rate": 7.756276295828108e-05,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1954508826136589,
      "num_tokens": 17576807.0,
      "step": 8030
    },
    {
      "entropy": 6.303921890258789,
      "epoch": 8.575547250400428,
      "grad_norm": 1.265625,
      "learning_rate": 7.736036383321467e-05,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20346123427152635,
      "num_tokens": 17587176.0,
      "step": 8035
    },
    {
      "entropy": 6.255138206481933,
      "epoch": 8.580886278697276,
      "grad_norm": 1.21875,
      "learning_rate": 7.715866246019181e-05,
      "loss": 5.3759,
      "mean_token_accuracy": 0.1968400612473488,
      "num_tokens": 17598681.0,
      "step": 8040
    },
    {
      "entropy": 6.262888479232788,
      "epoch": 8.586225306994127,
      "grad_norm": 1.2421875,
      "learning_rate": 7.695765955130439e-05,
      "loss": 5.4428,
      "mean_token_accuracy": 0.1876814156770706,
      "num_tokens": 17611782.0,
      "step": 8045
    },
    {
      "entropy": 6.342615556716919,
      "epoch": 8.591564335290977,
      "grad_norm": 1.234375,
      "learning_rate": 7.675735581617799e-05,
      "loss": 5.3968,
      "mean_token_accuracy": 0.1906011000275612,
      "num_tokens": 17623422.0,
      "step": 8050
    },
    {
      "entropy": 6.258514022827148,
      "epoch": 8.596903363587828,
      "grad_norm": 1.015625,
      "learning_rate": 7.655775196197017e-05,
      "loss": 5.3518,
      "mean_token_accuracy": 0.20063089579343796,
      "num_tokens": 17635844.0,
      "step": 8055
    },
    {
      "entropy": 6.334230518341064,
      "epoch": 8.602242391884676,
      "grad_norm": 1.234375,
      "learning_rate": 7.635884869336753e-05,
      "loss": 5.4009,
      "mean_token_accuracy": 0.19301522970199586,
      "num_tokens": 17646476.0,
      "step": 8060
    },
    {
      "entropy": 6.269136762619018,
      "epoch": 8.607581420181527,
      "grad_norm": 1.109375,
      "learning_rate": 7.616064671258323e-05,
      "loss": 5.3929,
      "mean_token_accuracy": 0.19156181663274766,
      "num_tokens": 17658601.0,
      "step": 8065
    },
    {
      "entropy": 6.231473970413208,
      "epoch": 8.612920448478377,
      "grad_norm": 1.1171875,
      "learning_rate": 7.596314671935467e-05,
      "loss": 5.32,
      "mean_token_accuracy": 0.1986356794834137,
      "num_tokens": 17669287.0,
      "step": 8070
    },
    {
      "entropy": 6.267216491699219,
      "epoch": 8.618259476775227,
      "grad_norm": 1.28125,
      "learning_rate": 7.576634941094082e-05,
      "loss": 5.3243,
      "mean_token_accuracy": 0.19997939467430115,
      "num_tokens": 17679545.0,
      "step": 8075
    },
    {
      "entropy": 6.307940292358398,
      "epoch": 8.623598505072078,
      "grad_norm": 1.3203125,
      "learning_rate": 7.557025548212012e-05,
      "loss": 5.4149,
      "mean_token_accuracy": 0.1952625036239624,
      "num_tokens": 17689779.0,
      "step": 8080
    },
    {
      "entropy": 6.272298049926758,
      "epoch": 8.628937533368926,
      "grad_norm": 1.1796875,
      "learning_rate": 7.537486562518735e-05,
      "loss": 5.4555,
      "mean_token_accuracy": 0.1893584817647934,
      "num_tokens": 17702643.0,
      "step": 8085
    },
    {
      "entropy": 6.264082765579223,
      "epoch": 8.634276561665777,
      "grad_norm": 1.265625,
      "learning_rate": 7.518018052995208e-05,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19213759452104567,
      "num_tokens": 17713166.0,
      "step": 8090
    },
    {
      "entropy": 6.306223630905151,
      "epoch": 8.639615589962627,
      "grad_norm": 1.0546875,
      "learning_rate": 7.498620088373544e-05,
      "loss": 5.452,
      "mean_token_accuracy": 0.1918992280960083,
      "num_tokens": 17724822.0,
      "step": 8095
    },
    {
      "entropy": 6.263330554962158,
      "epoch": 8.644954618259476,
      "grad_norm": 1.171875,
      "learning_rate": 7.479292737136828e-05,
      "loss": 5.306,
      "mean_token_accuracy": 0.1935967832803726,
      "num_tokens": 17737078.0,
      "step": 8100
    },
    {
      "entropy": 6.295151472091675,
      "epoch": 8.650293646556326,
      "grad_norm": 1.2109375,
      "learning_rate": 7.460036067518823e-05,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18652090430259705,
      "num_tokens": 17747670.0,
      "step": 8105
    },
    {
      "entropy": 6.289189004898072,
      "epoch": 8.655632674853177,
      "grad_norm": 1.203125,
      "learning_rate": 7.44085014750378e-05,
      "loss": 5.4625,
      "mean_token_accuracy": 0.19083506464958191,
      "num_tokens": 17758797.0,
      "step": 8110
    },
    {
      "entropy": 6.2896851062774655,
      "epoch": 8.660971703150027,
      "grad_norm": 1.2109375,
      "learning_rate": 7.421735044826161e-05,
      "loss": 5.4579,
      "mean_token_accuracy": 0.19278580099344253,
      "num_tokens": 17768713.0,
      "step": 8115
    },
    {
      "entropy": 6.365323209762574,
      "epoch": 8.666310731446877,
      "grad_norm": 1.2265625,
      "learning_rate": 7.40269082697041e-05,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19228217154741287,
      "num_tokens": 17778964.0,
      "step": 8120
    },
    {
      "entropy": 6.255847215652466,
      "epoch": 8.671649759743726,
      "grad_norm": 1.1796875,
      "learning_rate": 7.383717561170728e-05,
      "loss": 5.3435,
      "mean_token_accuracy": 0.20172299146652223,
      "num_tokens": 17790108.0,
      "step": 8125
    },
    {
      "entropy": 6.313925886154175,
      "epoch": 8.676988788040576,
      "grad_norm": 1.1171875,
      "learning_rate": 7.364815314410816e-05,
      "loss": 5.4923,
      "mean_token_accuracy": 0.19610096514225006,
      "num_tokens": 17801626.0,
      "step": 8130
    },
    {
      "entropy": 6.339280319213867,
      "epoch": 8.682327816337427,
      "grad_norm": 1.140625,
      "learning_rate": 7.345984153423656e-05,
      "loss": 5.4846,
      "mean_token_accuracy": 0.18328692018985748,
      "num_tokens": 17812355.0,
      "step": 8135
    },
    {
      "entropy": 6.288567733764649,
      "epoch": 8.687666844634277,
      "grad_norm": 1.2890625,
      "learning_rate": 7.327224144691255e-05,
      "loss": 5.4559,
      "mean_token_accuracy": 0.18972910493612288,
      "num_tokens": 17822986.0,
      "step": 8140
    },
    {
      "entropy": 6.30386004447937,
      "epoch": 8.693005872931126,
      "grad_norm": 1.3828125,
      "learning_rate": 7.30853535444444e-05,
      "loss": 5.3695,
      "mean_token_accuracy": 0.19032205194234847,
      "num_tokens": 17833656.0,
      "step": 8145
    },
    {
      "entropy": 6.334438371658325,
      "epoch": 8.698344901227976,
      "grad_norm": 1.1484375,
      "learning_rate": 7.289917848662582e-05,
      "loss": 5.4056,
      "mean_token_accuracy": 0.19864920377731324,
      "num_tokens": 17844124.0,
      "step": 8150
    },
    {
      "entropy": 6.333331680297851,
      "epoch": 8.703683929524827,
      "grad_norm": 1.3046875,
      "learning_rate": 7.271371693073408e-05,
      "loss": 5.3834,
      "mean_token_accuracy": 0.19299320131540298,
      "num_tokens": 17853528.0,
      "step": 8155
    },
    {
      "entropy": 6.215679454803467,
      "epoch": 8.709022957821677,
      "grad_norm": 1.25,
      "learning_rate": 7.252896953152744e-05,
      "loss": 5.3577,
      "mean_token_accuracy": 0.1953199803829193,
      "num_tokens": 17864599.0,
      "step": 8160
    },
    {
      "entropy": 6.3130731105804445,
      "epoch": 8.714361986118526,
      "grad_norm": 1.1875,
      "learning_rate": 7.23449369412428e-05,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19741948544979096,
      "num_tokens": 17874829.0,
      "step": 8165
    },
    {
      "entropy": 6.300188207626343,
      "epoch": 8.719701014415376,
      "grad_norm": 1.265625,
      "learning_rate": 7.216161980959363e-05,
      "loss": 5.3857,
      "mean_token_accuracy": 0.19758072048425673,
      "num_tokens": 17884914.0,
      "step": 8170
    },
    {
      "entropy": 6.2661253929138185,
      "epoch": 8.725040042712227,
      "grad_norm": 1.2578125,
      "learning_rate": 7.197901878376732e-05,
      "loss": 5.3757,
      "mean_token_accuracy": 0.1984149172902107,
      "num_tokens": 17896156.0,
      "step": 8175
    },
    {
      "entropy": 6.25427360534668,
      "epoch": 8.730379071009077,
      "grad_norm": 1.265625,
      "learning_rate": 7.179713450842338e-05,
      "loss": 5.4116,
      "mean_token_accuracy": 0.18684110045433044,
      "num_tokens": 17906446.0,
      "step": 8180
    },
    {
      "entropy": 6.279653835296631,
      "epoch": 8.735718099305926,
      "grad_norm": 1.125,
      "learning_rate": 7.161596762569051e-05,
      "loss": 5.4131,
      "mean_token_accuracy": 0.18811364769935607,
      "num_tokens": 17918488.0,
      "step": 8185
    },
    {
      "entropy": 6.2619095802307125,
      "epoch": 8.741057127602776,
      "grad_norm": 1.296875,
      "learning_rate": 7.143551877516509e-05,
      "loss": 5.415,
      "mean_token_accuracy": 0.19475423395633698,
      "num_tokens": 17928486.0,
      "step": 8190
    },
    {
      "entropy": 6.161016273498535,
      "epoch": 8.746396155899626,
      "grad_norm": 1.21875,
      "learning_rate": 7.125578859390828e-05,
      "loss": 5.3103,
      "mean_token_accuracy": 0.2026986762881279,
      "num_tokens": 17939036.0,
      "step": 8195
    },
    {
      "entropy": 6.322266101837158,
      "epoch": 8.751735184196477,
      "grad_norm": 1.25,
      "learning_rate": 7.107677771644419e-05,
      "loss": 5.4278,
      "mean_token_accuracy": 0.19225548058748246,
      "num_tokens": 17949610.0,
      "step": 8200
    },
    {
      "entropy": 6.283733224868774,
      "epoch": 8.757074212493325,
      "grad_norm": 1.265625,
      "learning_rate": 7.089848677475735e-05,
      "loss": 5.3987,
      "mean_token_accuracy": 0.19772477447986603,
      "num_tokens": 17960110.0,
      "step": 8205
    },
    {
      "entropy": 6.319091415405273,
      "epoch": 8.762413240790176,
      "grad_norm": 1.328125,
      "learning_rate": 7.072091639829079e-05,
      "loss": 5.4614,
      "mean_token_accuracy": 0.18673837184906006,
      "num_tokens": 17970293.0,
      "step": 8210
    },
    {
      "entropy": 6.276460742950439,
      "epoch": 8.767752269087026,
      "grad_norm": 1.34375,
      "learning_rate": 7.054406721394338e-05,
      "loss": 5.4007,
      "mean_token_accuracy": 0.19787448197603225,
      "num_tokens": 17981615.0,
      "step": 8215
    },
    {
      "entropy": 6.209004163742065,
      "epoch": 8.773091297383877,
      "grad_norm": 1.328125,
      "learning_rate": 7.036793984606808e-05,
      "loss": 5.2191,
      "mean_token_accuracy": 0.20935241430997847,
      "num_tokens": 17992083.0,
      "step": 8220
    },
    {
      "entropy": 6.287459468841552,
      "epoch": 8.778430325680727,
      "grad_norm": 1.2421875,
      "learning_rate": 7.019253491646954e-05,
      "loss": 5.4236,
      "mean_token_accuracy": 0.1912770926952362,
      "num_tokens": 18003374.0,
      "step": 8225
    },
    {
      "entropy": 6.320337247848511,
      "epoch": 8.783769353977576,
      "grad_norm": 1.109375,
      "learning_rate": 7.001785304440168e-05,
      "loss": 5.4061,
      "mean_token_accuracy": 0.19370096325874328,
      "num_tokens": 18015450.0,
      "step": 8230
    },
    {
      "entropy": 6.319497776031494,
      "epoch": 8.789108382274426,
      "grad_norm": 1.15625,
      "learning_rate": 6.9843894846566e-05,
      "loss": 5.4613,
      "mean_token_accuracy": 0.19284655153751373,
      "num_tokens": 18026387.0,
      "step": 8235
    },
    {
      "entropy": 6.297713279724121,
      "epoch": 8.794447410571276,
      "grad_norm": 1.1328125,
      "learning_rate": 6.967066093710892e-05,
      "loss": 5.543,
      "mean_token_accuracy": 0.18906300365924836,
      "num_tokens": 18038387.0,
      "step": 8240
    },
    {
      "entropy": 6.267315292358399,
      "epoch": 8.799786438868125,
      "grad_norm": 1.2421875,
      "learning_rate": 6.949815192761993e-05,
      "loss": 5.3562,
      "mean_token_accuracy": 0.19610127210617065,
      "num_tokens": 18048957.0,
      "step": 8245
    },
    {
      "entropy": 6.1813019752502445,
      "epoch": 8.805125467164975,
      "grad_norm": 1.34375,
      "learning_rate": 6.932636842712914e-05,
      "loss": 5.33,
      "mean_token_accuracy": 0.2018466502428055,
      "num_tokens": 18059750.0,
      "step": 8250
    },
    {
      "entropy": 6.306232166290283,
      "epoch": 8.810464495461826,
      "grad_norm": 1.25,
      "learning_rate": 6.915531104210554e-05,
      "loss": 5.4618,
      "mean_token_accuracy": 0.18518505543470382,
      "num_tokens": 18070321.0,
      "step": 8255
    },
    {
      "entropy": 6.278579187393189,
      "epoch": 8.815803523758676,
      "grad_norm": 1.09375,
      "learning_rate": 6.898498037645451e-05,
      "loss": 5.4655,
      "mean_token_accuracy": 0.18751134872436523,
      "num_tokens": 18082322.0,
      "step": 8260
    },
    {
      "entropy": 6.325710391998291,
      "epoch": 8.821142552055527,
      "grad_norm": 1.171875,
      "learning_rate": 6.881537703151575e-05,
      "loss": 5.3813,
      "mean_token_accuracy": 0.1982166588306427,
      "num_tokens": 18093379.0,
      "step": 8265
    },
    {
      "entropy": 6.189786672592163,
      "epoch": 8.826481580352375,
      "grad_norm": 1.25,
      "learning_rate": 6.864650160606131e-05,
      "loss": 5.2825,
      "mean_token_accuracy": 0.20476205050945281,
      "num_tokens": 18105365.0,
      "step": 8270
    },
    {
      "entropy": 6.2697841167449955,
      "epoch": 8.831820608649226,
      "grad_norm": 1.21875,
      "learning_rate": 6.847835469629332e-05,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2016316384077072,
      "num_tokens": 18116058.0,
      "step": 8275
    },
    {
      "entropy": 6.314205932617187,
      "epoch": 8.837159636946076,
      "grad_norm": 1.265625,
      "learning_rate": 6.831093689584193e-05,
      "loss": 5.4293,
      "mean_token_accuracy": 0.19054114669561387,
      "num_tokens": 18126771.0,
      "step": 8280
    },
    {
      "entropy": 6.288633871078491,
      "epoch": 8.842498665242926,
      "grad_norm": 1.2421875,
      "learning_rate": 6.814424879576321e-05,
      "loss": 5.423,
      "mean_token_accuracy": 0.19208714216947556,
      "num_tokens": 18137384.0,
      "step": 8285
    },
    {
      "entropy": 6.259862089157105,
      "epoch": 8.847837693539775,
      "grad_norm": 1.203125,
      "learning_rate": 6.797829098453717e-05,
      "loss": 5.4301,
      "mean_token_accuracy": 0.19221352338790892,
      "num_tokens": 18148230.0,
      "step": 8290
    },
    {
      "entropy": 6.305414438247681,
      "epoch": 8.853176721836626,
      "grad_norm": 1.2109375,
      "learning_rate": 6.781306404806544e-05,
      "loss": 5.5183,
      "mean_token_accuracy": 0.18625222146511078,
      "num_tokens": 18159184.0,
      "step": 8295
    },
    {
      "entropy": 6.286434888839722,
      "epoch": 8.858515750133476,
      "grad_norm": 1.15625,
      "learning_rate": 6.764856856966952e-05,
      "loss": 5.416,
      "mean_token_accuracy": 0.1962931737303734,
      "num_tokens": 18170336.0,
      "step": 8300
    },
    {
      "entropy": 6.335140943527222,
      "epoch": 8.863854778430326,
      "grad_norm": 1.265625,
      "learning_rate": 6.748480513008843e-05,
      "loss": 5.4185,
      "mean_token_accuracy": 0.19309814125299454,
      "num_tokens": 18181428.0,
      "step": 8305
    },
    {
      "entropy": 6.326192474365234,
      "epoch": 8.869193806727175,
      "grad_norm": 1.203125,
      "learning_rate": 6.732177430747685e-05,
      "loss": 5.4468,
      "mean_token_accuracy": 0.19147571921348572,
      "num_tokens": 18192452.0,
      "step": 8310
    },
    {
      "entropy": 6.315118980407715,
      "epoch": 8.874532835024025,
      "grad_norm": 1.21875,
      "learning_rate": 6.715947667740298e-05,
      "loss": 5.4147,
      "mean_token_accuracy": 0.19402022659778595,
      "num_tokens": 18203374.0,
      "step": 8315
    },
    {
      "entropy": 6.273036479949951,
      "epoch": 8.879871863320876,
      "grad_norm": 1.1875,
      "learning_rate": 6.699791281284645e-05,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19211985021829606,
      "num_tokens": 18214420.0,
      "step": 8320
    },
    {
      "entropy": 6.283283376693726,
      "epoch": 8.885210891617726,
      "grad_norm": 1.1875,
      "learning_rate": 6.68370832841966e-05,
      "loss": 5.4441,
      "mean_token_accuracy": 0.1880252718925476,
      "num_tokens": 18226176.0,
      "step": 8325
    },
    {
      "entropy": 6.352131271362305,
      "epoch": 8.890549919914575,
      "grad_norm": 1.3125,
      "learning_rate": 6.667698865924998e-05,
      "loss": 5.4257,
      "mean_token_accuracy": 0.19260345548391342,
      "num_tokens": 18236792.0,
      "step": 8330
    },
    {
      "entropy": 6.2733378410339355,
      "epoch": 8.895888948211425,
      "grad_norm": 1.1796875,
      "learning_rate": 6.651762950320893e-05,
      "loss": 5.4312,
      "mean_token_accuracy": 0.19283699989318848,
      "num_tokens": 18248047.0,
      "step": 8335
    },
    {
      "entropy": 6.2758337497711185,
      "epoch": 8.901227976508276,
      "grad_norm": 1.1328125,
      "learning_rate": 6.635900637867902e-05,
      "loss": 5.376,
      "mean_token_accuracy": 0.2016781061887741,
      "num_tokens": 18258093.0,
      "step": 8340
    },
    {
      "entropy": 6.266999101638794,
      "epoch": 8.906567004805126,
      "grad_norm": 1.3515625,
      "learning_rate": 6.620111984566739e-05,
      "loss": 5.3901,
      "mean_token_accuracy": 0.19536768645048141,
      "num_tokens": 18268649.0,
      "step": 8345
    },
    {
      "entropy": 6.2864086627960205,
      "epoch": 8.911906033101975,
      "grad_norm": 1.1796875,
      "learning_rate": 6.604397046158066e-05,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19759741872549058,
      "num_tokens": 18278127.0,
      "step": 8350
    },
    {
      "entropy": 6.262919664382935,
      "epoch": 8.917245061398825,
      "grad_norm": 1.21875,
      "learning_rate": 6.588755878122318e-05,
      "loss": 5.3978,
      "mean_token_accuracy": 0.19958573281764985,
      "num_tokens": 18289288.0,
      "step": 8355
    },
    {
      "entropy": 6.278789710998535,
      "epoch": 8.922584089695675,
      "grad_norm": 1.109375,
      "learning_rate": 6.573188535679464e-05,
      "loss": 5.3924,
      "mean_token_accuracy": 0.19610107094049453,
      "num_tokens": 18299482.0,
      "step": 8360
    },
    {
      "entropy": 6.237138795852661,
      "epoch": 8.927923117992526,
      "grad_norm": 1.2421875,
      "learning_rate": 6.557695073788853e-05,
      "loss": 5.2958,
      "mean_token_accuracy": 0.20035430788993835,
      "num_tokens": 18309499.0,
      "step": 8365
    },
    {
      "entropy": 6.274085283279419,
      "epoch": 8.933262146289376,
      "grad_norm": 1.1796875,
      "learning_rate": 6.542275547149005e-05,
      "loss": 5.3748,
      "mean_token_accuracy": 0.19981057792901993,
      "num_tokens": 18320106.0,
      "step": 8370
    },
    {
      "entropy": 6.286875486373901,
      "epoch": 8.938601174586225,
      "grad_norm": 1.203125,
      "learning_rate": 6.526930010197412e-05,
      "loss": 5.4601,
      "mean_token_accuracy": 0.1885668605566025,
      "num_tokens": 18331657.0,
      "step": 8375
    },
    {
      "entropy": 6.303912401199341,
      "epoch": 8.943940202883075,
      "grad_norm": 1.1875,
      "learning_rate": 6.511658517110346e-05,
      "loss": 5.5049,
      "mean_token_accuracy": 0.18762866109609605,
      "num_tokens": 18342951.0,
      "step": 8380
    },
    {
      "entropy": 6.258378219604492,
      "epoch": 8.949279231179926,
      "grad_norm": 1.3203125,
      "learning_rate": 6.496461121802679e-05,
      "loss": 5.3829,
      "mean_token_accuracy": 0.1956523537635803,
      "num_tokens": 18353515.0,
      "step": 8385
    },
    {
      "entropy": 6.277213478088379,
      "epoch": 8.954618259476776,
      "grad_norm": 1.2890625,
      "learning_rate": 6.48133787792769e-05,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19595738202333451,
      "num_tokens": 18365139.0,
      "step": 8390
    },
    {
      "entropy": 6.267573308944702,
      "epoch": 8.959957287773625,
      "grad_norm": 1.1328125,
      "learning_rate": 6.46628883887686e-05,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19399414509534835,
      "num_tokens": 18376426.0,
      "step": 8395
    },
    {
      "entropy": 6.318247222900391,
      "epoch": 8.965296316070475,
      "grad_norm": 1.0546875,
      "learning_rate": 6.451314057779708e-05,
      "loss": 5.4031,
      "mean_token_accuracy": 0.1949224889278412,
      "num_tokens": 18387643.0,
      "step": 8400
    },
    {
      "entropy": 6.2453080177307125,
      "epoch": 8.970635344367325,
      "grad_norm": 1.34375,
      "learning_rate": 6.436413587503588e-05,
      "loss": 5.3801,
      "mean_token_accuracy": 0.19522539228200914,
      "num_tokens": 18397625.0,
      "step": 8405
    },
    {
      "entropy": 6.24573826789856,
      "epoch": 8.975974372664176,
      "grad_norm": 1.125,
      "learning_rate": 6.421587480653497e-05,
      "loss": 5.4012,
      "mean_token_accuracy": 0.19939807951450347,
      "num_tokens": 18410033.0,
      "step": 8410
    },
    {
      "entropy": 6.2669624328613285,
      "epoch": 8.981313400961024,
      "grad_norm": 1.140625,
      "learning_rate": 6.406835789571905e-05,
      "loss": 5.3825,
      "mean_token_accuracy": 0.19407907128334045,
      "num_tokens": 18421020.0,
      "step": 8415
    },
    {
      "entropy": 6.294377422332763,
      "epoch": 8.986652429257875,
      "grad_norm": 1.0703125,
      "learning_rate": 6.392158566338561e-05,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19481519758701324,
      "num_tokens": 18432884.0,
      "step": 8420
    },
    {
      "entropy": 6.332372999191284,
      "epoch": 8.991991457554725,
      "grad_norm": 1.1640625,
      "learning_rate": 6.377555862770311e-05,
      "loss": 5.4752,
      "mean_token_accuracy": 0.17780984342098236,
      "num_tokens": 18443478.0,
      "step": 8425
    },
    {
      "entropy": 6.326786518096924,
      "epoch": 8.997330485851576,
      "grad_norm": 1.1953125,
      "learning_rate": 6.363027730420915e-05,
      "loss": 5.4363,
      "mean_token_accuracy": 0.19262375235557555,
      "num_tokens": 18453943.0,
      "step": 8430
    },
    {
      "entropy": 6.3219667010837135,
      "epoch": 9.00213561131874,
      "grad_norm": 1.1171875,
      "learning_rate": 6.348574220580863e-05,
      "loss": 5.3917,
      "mean_token_accuracy": 0.19410793317688835,
      "num_tokens": 18464880.0,
      "step": 8435
    },
    {
      "entropy": 6.291064596176147,
      "epoch": 9.00747463961559,
      "grad_norm": 1.109375,
      "learning_rate": 6.334195384277198e-05,
      "loss": 5.3365,
      "mean_token_accuracy": 0.19902778118848802,
      "num_tokens": 18475884.0,
      "step": 8440
    },
    {
      "entropy": 6.322910499572754,
      "epoch": 9.01281366791244,
      "grad_norm": 1.328125,
      "learning_rate": 6.319891272273333e-05,
      "loss": 5.397,
      "mean_token_accuracy": 0.19973329156637193,
      "num_tokens": 18486900.0,
      "step": 8445
    },
    {
      "entropy": 6.2828192710876465,
      "epoch": 9.01815269620929,
      "grad_norm": 1.265625,
      "learning_rate": 6.30566193506887e-05,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1961638391017914,
      "num_tokens": 18497070.0,
      "step": 8450
    },
    {
      "entropy": 6.302260255813598,
      "epoch": 9.02349172450614,
      "grad_norm": 1.1640625,
      "learning_rate": 6.291507422899425e-05,
      "loss": 5.4474,
      "mean_token_accuracy": 0.19043018370866777,
      "num_tokens": 18508778.0,
      "step": 8455
    },
    {
      "entropy": 6.300608444213867,
      "epoch": 9.028830752802989,
      "grad_norm": 1.1953125,
      "learning_rate": 6.277427785736455e-05,
      "loss": 5.4072,
      "mean_token_accuracy": 0.1943545952439308,
      "num_tokens": 18521076.0,
      "step": 8460
    },
    {
      "entropy": 6.291013383865357,
      "epoch": 9.03416978109984,
      "grad_norm": 1.2109375,
      "learning_rate": 6.26342307328707e-05,
      "loss": 5.3299,
      "mean_token_accuracy": 0.19972898364067077,
      "num_tokens": 18532582.0,
      "step": 8465
    },
    {
      "entropy": 6.308874654769897,
      "epoch": 9.03950880939669,
      "grad_norm": 1.25,
      "learning_rate": 6.249493334993873e-05,
      "loss": 5.3802,
      "mean_token_accuracy": 0.19778040051460266,
      "num_tokens": 18543263.0,
      "step": 8470
    },
    {
      "entropy": 6.294857692718506,
      "epoch": 9.04484783769354,
      "grad_norm": 1.375,
      "learning_rate": 6.235638620034756e-05,
      "loss": 5.4241,
      "mean_token_accuracy": 0.1885809302330017,
      "num_tokens": 18554367.0,
      "step": 8475
    },
    {
      "entropy": 6.267254829406738,
      "epoch": 9.05018686599039,
      "grad_norm": 1.2421875,
      "learning_rate": 6.221858977322774e-05,
      "loss": 5.352,
      "mean_token_accuracy": 0.20446752309799193,
      "num_tokens": 18565075.0,
      "step": 8480
    },
    {
      "entropy": 6.256584358215332,
      "epoch": 9.05552589428724,
      "grad_norm": 1.203125,
      "learning_rate": 6.208154455505918e-05,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20374154597520827,
      "num_tokens": 18575166.0,
      "step": 8485
    },
    {
      "entropy": 6.303496646881103,
      "epoch": 9.06086492258409,
      "grad_norm": 1.125,
      "learning_rate": 6.194525102966997e-05,
      "loss": 5.4337,
      "mean_token_accuracy": 0.19414404034614563,
      "num_tokens": 18586975.0,
      "step": 8490
    },
    {
      "entropy": 6.254971933364868,
      "epoch": 9.06620395088094,
      "grad_norm": 1.1953125,
      "learning_rate": 6.180970967823415e-05,
      "loss": 5.2621,
      "mean_token_accuracy": 0.21001349538564681,
      "num_tokens": 18598524.0,
      "step": 8495
    },
    {
      "entropy": 6.264356327056885,
      "epoch": 9.07154297917779,
      "grad_norm": 1.203125,
      "learning_rate": 6.167492097927051e-05,
      "loss": 5.3619,
      "mean_token_accuracy": 0.19953151643276215,
      "num_tokens": 18609099.0,
      "step": 8500
    },
    {
      "epoch": 9.07154297917779,
      "eval_entropy": 6.114785921418822,
      "eval_loss": 6.549992561340332,
      "eval_mean_token_accuracy": 0.13512441571076195,
      "eval_num_tokens": 18609099.0,
      "eval_runtime": 1.8163,
      "eval_samples_per_second": 1695.776,
      "eval_steps_per_second": 211.972,
      "step": 8500
    },
    {
      "entropy": 6.229560041427613,
      "epoch": 9.076882007474639,
      "grad_norm": 1.328125,
      "learning_rate": 6.154088540864045e-05,
      "loss": 5.338,
      "mean_token_accuracy": 0.19794839322566987,
      "num_tokens": 18619901.0,
      "step": 8505
    },
    {
      "entropy": 6.291391324996948,
      "epoch": 9.08222103577149,
      "grad_norm": 1.484375,
      "learning_rate": 6.140760343954654e-05,
      "loss": 5.3933,
      "mean_token_accuracy": 0.2033597096800804,
      "num_tokens": 18629556.0,
      "step": 8510
    },
    {
      "entropy": 6.246613550186157,
      "epoch": 9.08756006406834,
      "grad_norm": 1.15625,
      "learning_rate": 6.127507554253101e-05,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20000605434179305,
      "num_tokens": 18640487.0,
      "step": 8515
    },
    {
      "entropy": 6.2668591976165775,
      "epoch": 9.09289909236519,
      "grad_norm": 1.0625,
      "learning_rate": 6.11433021854736e-05,
      "loss": 5.4005,
      "mean_token_accuracy": 0.19509114921092988,
      "num_tokens": 18651828.0,
      "step": 8520
    },
    {
      "entropy": 6.2748137474060055,
      "epoch": 9.098238120662039,
      "grad_norm": 1.2578125,
      "learning_rate": 6.1012283833590465e-05,
      "loss": 5.4124,
      "mean_token_accuracy": 0.19893601536750793,
      "num_tokens": 18662673.0,
      "step": 8525
    },
    {
      "entropy": 6.260831069946289,
      "epoch": 9.10357714895889,
      "grad_norm": 1.265625,
      "learning_rate": 6.088202094943207e-05,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19352519661188125,
      "num_tokens": 18672345.0,
      "step": 8530
    },
    {
      "entropy": 6.315216684341431,
      "epoch": 9.10891617725574,
      "grad_norm": 1.1796875,
      "learning_rate": 6.075251399288193e-05,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1841621220111847,
      "num_tokens": 18684514.0,
      "step": 8535
    },
    {
      "entropy": 6.250639724731445,
      "epoch": 9.11425520555259,
      "grad_norm": 1.2109375,
      "learning_rate": 6.062376342115465e-05,
      "loss": 5.3459,
      "mean_token_accuracy": 0.20057544261217117,
      "num_tokens": 18694986.0,
      "step": 8540
    },
    {
      "entropy": 6.322076272964478,
      "epoch": 9.119594233849439,
      "grad_norm": 1.0703125,
      "learning_rate": 6.049576968879459e-05,
      "loss": 5.4228,
      "mean_token_accuracy": 0.19311343878507614,
      "num_tokens": 18706585.0,
      "step": 8545
    },
    {
      "entropy": 6.305493640899658,
      "epoch": 9.124933262146289,
      "grad_norm": 1.171875,
      "learning_rate": 6.0368533247674074e-05,
      "loss": 5.4126,
      "mean_token_accuracy": 0.19299951195716858,
      "num_tokens": 18717745.0,
      "step": 8550
    },
    {
      "entropy": 6.256345796585083,
      "epoch": 9.13027229044314,
      "grad_norm": 1.21875,
      "learning_rate": 6.0242054546991954e-05,
      "loss": 5.3487,
      "mean_token_accuracy": 0.20222388803958893,
      "num_tokens": 18727875.0,
      "step": 8555
    },
    {
      "entropy": 6.275346279144287,
      "epoch": 9.13561131873999,
      "grad_norm": 1.2109375,
      "learning_rate": 6.0116334033271826e-05,
      "loss": 5.3777,
      "mean_token_accuracy": 0.19150869846343993,
      "num_tokens": 18738290.0,
      "step": 8560
    },
    {
      "entropy": 6.28104887008667,
      "epoch": 9.140950347036839,
      "grad_norm": 1.203125,
      "learning_rate": 5.999137215036063e-05,
      "loss": 5.3961,
      "mean_token_accuracy": 0.19925588071346284,
      "num_tokens": 18749462.0,
      "step": 8565
    },
    {
      "entropy": 6.292618322372436,
      "epoch": 9.146289375333689,
      "grad_norm": 1.296875,
      "learning_rate": 5.9867169339427105e-05,
      "loss": 5.3386,
      "mean_token_accuracy": 0.19864167869091034,
      "num_tokens": 18759999.0,
      "step": 8570
    },
    {
      "entropy": 6.2744917392730715,
      "epoch": 9.15162840363054,
      "grad_norm": 1.2109375,
      "learning_rate": 5.974372603895989e-05,
      "loss": 5.3664,
      "mean_token_accuracy": 0.20330148190259933,
      "num_tokens": 18771421.0,
      "step": 8575
    },
    {
      "entropy": 6.231125020980835,
      "epoch": 9.15696743192739,
      "grad_norm": 1.2734375,
      "learning_rate": 5.962104268476648e-05,
      "loss": 5.3475,
      "mean_token_accuracy": 0.19947514533996583,
      "num_tokens": 18781423.0,
      "step": 8580
    },
    {
      "entropy": 6.301869916915893,
      "epoch": 9.162306460224238,
      "grad_norm": 1.3984375,
      "learning_rate": 5.9499119709971266e-05,
      "loss": 5.3937,
      "mean_token_accuracy": 0.19459829330444336,
      "num_tokens": 18790824.0,
      "step": 8585
    },
    {
      "entropy": 6.3307005882263185,
      "epoch": 9.167645488521089,
      "grad_norm": 1.328125,
      "learning_rate": 5.937795754501431e-05,
      "loss": 5.4154,
      "mean_token_accuracy": 0.1953270897269249,
      "num_tokens": 18801694.0,
      "step": 8590
    },
    {
      "entropy": 6.30227746963501,
      "epoch": 9.17298451681794,
      "grad_norm": 1.4765625,
      "learning_rate": 5.9257556617649594e-05,
      "loss": 5.3517,
      "mean_token_accuracy": 0.1940465450286865,
      "num_tokens": 18812180.0,
      "step": 8595
    },
    {
      "entropy": 6.290414094924927,
      "epoch": 9.17832354511479,
      "grad_norm": 1.421875,
      "learning_rate": 5.913791735294367e-05,
      "loss": 5.4369,
      "mean_token_accuracy": 0.19648680686950684,
      "num_tokens": 18823764.0,
      "step": 8600
    },
    {
      "entropy": 6.329561948776245,
      "epoch": 9.18366257341164,
      "grad_norm": 1.2578125,
      "learning_rate": 5.901904017327404e-05,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19430489838123322,
      "num_tokens": 18834592.0,
      "step": 8605
    },
    {
      "entropy": 6.305727863311768,
      "epoch": 9.189001601708489,
      "grad_norm": 1.296875,
      "learning_rate": 5.890092549832776e-05,
      "loss": 5.3353,
      "mean_token_accuracy": 0.19843092262744905,
      "num_tokens": 18844714.0,
      "step": 8610
    },
    {
      "entropy": 6.302178955078125,
      "epoch": 9.194340630005339,
      "grad_norm": 1.234375,
      "learning_rate": 5.8783573745099895e-05,
      "loss": 5.3965,
      "mean_token_accuracy": 0.18645181506872177,
      "num_tokens": 18856680.0,
      "step": 8615
    },
    {
      "entropy": 6.312046670913697,
      "epoch": 9.19967965830219,
      "grad_norm": 1.2734375,
      "learning_rate": 5.8666985327892103e-05,
      "loss": 5.4183,
      "mean_token_accuracy": 0.1923958107829094,
      "num_tokens": 18867146.0,
      "step": 8620
    },
    {
      "entropy": 6.18888897895813,
      "epoch": 9.20501868659904,
      "grad_norm": 1.3203125,
      "learning_rate": 5.855116065831115e-05,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20159239172935486,
      "num_tokens": 18878678.0,
      "step": 8625
    },
    {
      "entropy": 6.295993375778198,
      "epoch": 9.210357714895888,
      "grad_norm": 1.2578125,
      "learning_rate": 5.8436100145267365e-05,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19432373344898224,
      "num_tokens": 18889431.0,
      "step": 8630
    },
    {
      "entropy": 6.289379692077636,
      "epoch": 9.215696743192739,
      "grad_norm": 1.2734375,
      "learning_rate": 5.83218041949734e-05,
      "loss": 5.3944,
      "mean_token_accuracy": 0.19059503376483916,
      "num_tokens": 18899860.0,
      "step": 8635
    },
    {
      "entropy": 6.275321006774902,
      "epoch": 9.22103577148959,
      "grad_norm": 1.1953125,
      "learning_rate": 5.820827321094254e-05,
      "loss": 5.3884,
      "mean_token_accuracy": 0.18925346434116364,
      "num_tokens": 18910589.0,
      "step": 8640
    },
    {
      "entropy": 6.344746208190918,
      "epoch": 9.22637479978644,
      "grad_norm": 1.15625,
      "learning_rate": 5.8095507593987545e-05,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19206939339637757,
      "num_tokens": 18921145.0,
      "step": 8645
    },
    {
      "entropy": 6.2736283302307125,
      "epoch": 9.231713828083288,
      "grad_norm": 1.2890625,
      "learning_rate": 5.798350774221899e-05,
      "loss": 5.3704,
      "mean_token_accuracy": 0.19876059740781785,
      "num_tokens": 18932600.0,
      "step": 8650
    },
    {
      "entropy": 6.267943048477173,
      "epoch": 9.237052856380139,
      "grad_norm": 1.171875,
      "learning_rate": 5.78722740510441e-05,
      "loss": 5.3859,
      "mean_token_accuracy": 0.1947265863418579,
      "num_tokens": 18944712.0,
      "step": 8655
    },
    {
      "entropy": 6.2824924945831295,
      "epoch": 9.242391884676989,
      "grad_norm": 1.140625,
      "learning_rate": 5.776180691316505e-05,
      "loss": 5.405,
      "mean_token_accuracy": 0.19813400655984878,
      "num_tokens": 18956462.0,
      "step": 8660
    },
    {
      "entropy": 6.27752366065979,
      "epoch": 9.24773091297384,
      "grad_norm": 1.34375,
      "learning_rate": 5.765210671857795e-05,
      "loss": 5.3838,
      "mean_token_accuracy": 0.19867743998765947,
      "num_tokens": 18967882.0,
      "step": 8665
    },
    {
      "entropy": 6.274408292770386,
      "epoch": 9.253069941270688,
      "grad_norm": 1.1796875,
      "learning_rate": 5.7543173854571176e-05,
      "loss": 5.3646,
      "mean_token_accuracy": 0.19525448232889175,
      "num_tokens": 18978797.0,
      "step": 8670
    },
    {
      "entropy": 6.322760105133057,
      "epoch": 9.258408969567538,
      "grad_norm": 1.125,
      "learning_rate": 5.743500870572404e-05,
      "loss": 5.3745,
      "mean_token_accuracy": 0.1901499330997467,
      "num_tokens": 18989506.0,
      "step": 8675
    },
    {
      "entropy": 6.30593957901001,
      "epoch": 9.263747997864389,
      "grad_norm": 1.234375,
      "learning_rate": 5.732761165390562e-05,
      "loss": 5.4193,
      "mean_token_accuracy": 0.1921993911266327,
      "num_tokens": 19000072.0,
      "step": 8680
    },
    {
      "entropy": 6.322987842559814,
      "epoch": 9.26908702616124,
      "grad_norm": 1.1328125,
      "learning_rate": 5.7220983078273176e-05,
      "loss": 5.4454,
      "mean_token_accuracy": 0.2010173797607422,
      "num_tokens": 19010656.0,
      "step": 8685
    },
    {
      "entropy": 6.306516361236572,
      "epoch": 9.274426054458088,
      "grad_norm": 1.359375,
      "learning_rate": 5.7115123355270996e-05,
      "loss": 5.3828,
      "mean_token_accuracy": 0.19560512900352478,
      "num_tokens": 19021811.0,
      "step": 8690
    },
    {
      "entropy": 6.246599578857422,
      "epoch": 9.279765082754938,
      "grad_norm": 1.1875,
      "learning_rate": 5.701003285862898e-05,
      "loss": 5.3498,
      "mean_token_accuracy": 0.1951931819319725,
      "num_tokens": 19032673.0,
      "step": 8695
    },
    {
      "entropy": 6.254972982406616,
      "epoch": 9.285104111051789,
      "grad_norm": 1.2421875,
      "learning_rate": 5.690571195936129e-05,
      "loss": 5.294,
      "mean_token_accuracy": 0.20297341346740722,
      "num_tokens": 19043421.0,
      "step": 8700
    },
    {
      "entropy": 6.254808473587036,
      "epoch": 9.290443139348639,
      "grad_norm": 1.3046875,
      "learning_rate": 5.6802161025765126e-05,
      "loss": 5.3124,
      "mean_token_accuracy": 0.20978861898183823,
      "num_tokens": 19053972.0,
      "step": 8705
    },
    {
      "entropy": 6.257140016555786,
      "epoch": 9.295782167645488,
      "grad_norm": 1.125,
      "learning_rate": 5.6699380423419315e-05,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19529064148664474,
      "num_tokens": 19064996.0,
      "step": 8710
    },
    {
      "entropy": 6.293576097488403,
      "epoch": 9.301121195942338,
      "grad_norm": 1.28125,
      "learning_rate": 5.659737051518319e-05,
      "loss": 5.405,
      "mean_token_accuracy": 0.19285243451595308,
      "num_tokens": 19077163.0,
      "step": 8715
    },
    {
      "entropy": 6.281959867477417,
      "epoch": 9.306460224239189,
      "grad_norm": 1.3515625,
      "learning_rate": 5.649613166119512e-05,
      "loss": 5.3772,
      "mean_token_accuracy": 0.19006721973419188,
      "num_tokens": 19087616.0,
      "step": 8720
    },
    {
      "entropy": 6.308144426345825,
      "epoch": 9.311799252536039,
      "grad_norm": 1.0625,
      "learning_rate": 5.639566421887136e-05,
      "loss": 5.4082,
      "mean_token_accuracy": 0.1936774879693985,
      "num_tokens": 19100267.0,
      "step": 8725
    },
    {
      "entropy": 6.217197275161743,
      "epoch": 9.31713828083289,
      "grad_norm": 1.2890625,
      "learning_rate": 5.6295968542904734e-05,
      "loss": 5.2625,
      "mean_token_accuracy": 0.21082474589347838,
      "num_tokens": 19109969.0,
      "step": 8730
    },
    {
      "entropy": 6.270903539657593,
      "epoch": 9.322477309129738,
      "grad_norm": 1.09375,
      "learning_rate": 5.619704498526345e-05,
      "loss": 5.3542,
      "mean_token_accuracy": 0.19501845836639403,
      "num_tokens": 19121033.0,
      "step": 8735
    },
    {
      "entropy": 6.305626344680786,
      "epoch": 9.327816337426588,
      "grad_norm": 1.2578125,
      "learning_rate": 5.6098893895189704e-05,
      "loss": 5.423,
      "mean_token_accuracy": 0.19037313908338546,
      "num_tokens": 19131451.0,
      "step": 8740
    },
    {
      "entropy": 6.2973391056060795,
      "epoch": 9.333155365723439,
      "grad_norm": 1.28125,
      "learning_rate": 5.6001515619198706e-05,
      "loss": 5.413,
      "mean_token_accuracy": 0.19424313753843309,
      "num_tokens": 19142187.0,
      "step": 8745
    },
    {
      "entropy": 6.3233716011047365,
      "epoch": 9.33849439402029,
      "grad_norm": 1.3515625,
      "learning_rate": 5.590491050107717e-05,
      "loss": 5.4345,
      "mean_token_accuracy": 0.19144673198461531,
      "num_tokens": 19152401.0,
      "step": 8750
    },
    {
      "entropy": 6.321363544464111,
      "epoch": 9.343833422317138,
      "grad_norm": 1.140625,
      "learning_rate": 5.580907888188229e-05,
      "loss": 5.4545,
      "mean_token_accuracy": 0.1866098314523697,
      "num_tokens": 19164042.0,
      "step": 8755
    },
    {
      "entropy": 6.327611064910888,
      "epoch": 9.349172450613988,
      "grad_norm": 1.2890625,
      "learning_rate": 5.571402109994049e-05,
      "loss": 5.3791,
      "mean_token_accuracy": 0.1929052382707596,
      "num_tokens": 19174840.0,
      "step": 8760
    },
    {
      "entropy": 6.285183763504028,
      "epoch": 9.354511478910839,
      "grad_norm": 1.2421875,
      "learning_rate": 5.5619737490846215e-05,
      "loss": 5.4256,
      "mean_token_accuracy": 0.18573228418827056,
      "num_tokens": 19186553.0,
      "step": 8765
    },
    {
      "entropy": 6.2528282642364506,
      "epoch": 9.359850507207689,
      "grad_norm": 1.1484375,
      "learning_rate": 5.55262283874607e-05,
      "loss": 5.3138,
      "mean_token_accuracy": 0.19975178986787795,
      "num_tokens": 19198010.0,
      "step": 8770
    },
    {
      "entropy": 6.268246603012085,
      "epoch": 9.365189535504538,
      "grad_norm": 1.2578125,
      "learning_rate": 5.543349411991089e-05,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20094972252845764,
      "num_tokens": 19207891.0,
      "step": 8775
    },
    {
      "entropy": 6.258787059783936,
      "epoch": 9.370528563801388,
      "grad_norm": 1.171875,
      "learning_rate": 5.534153501558823e-05,
      "loss": 5.38,
      "mean_token_accuracy": 0.19716913849115372,
      "num_tokens": 19219140.0,
      "step": 8780
    },
    {
      "entropy": 6.302610540390015,
      "epoch": 9.375867592098238,
      "grad_norm": 1.2578125,
      "learning_rate": 5.5250351399147507e-05,
      "loss": 5.3564,
      "mean_token_accuracy": 0.19000794887542724,
      "num_tokens": 19229642.0,
      "step": 8785
    },
    {
      "entropy": 6.275329351425171,
      "epoch": 9.381206620395089,
      "grad_norm": 1.0859375,
      "learning_rate": 5.515994359250574e-05,
      "loss": 5.4133,
      "mean_token_accuracy": 0.19166371673345567,
      "num_tokens": 19241184.0,
      "step": 8790
    },
    {
      "entropy": 6.325571823120117,
      "epoch": 9.386545648691937,
      "grad_norm": 1.234375,
      "learning_rate": 5.507031191484087e-05,
      "loss": 5.4533,
      "mean_token_accuracy": 0.18626566082239152,
      "num_tokens": 19252201.0,
      "step": 8795
    },
    {
      "entropy": 6.32686185836792,
      "epoch": 9.391884676988788,
      "grad_norm": 1.375,
      "learning_rate": 5.4981456682590955e-05,
      "loss": 5.4565,
      "mean_token_accuracy": 0.18756528198719025,
      "num_tokens": 19261683.0,
      "step": 8800
    },
    {
      "entropy": 6.2155046463012695,
      "epoch": 9.397223705285638,
      "grad_norm": 1.2421875,
      "learning_rate": 5.489337820945279e-05,
      "loss": 5.4225,
      "mean_token_accuracy": 0.18866721391677857,
      "num_tokens": 19273530.0,
      "step": 8805
    },
    {
      "entropy": 6.303810548782349,
      "epoch": 9.402562733582489,
      "grad_norm": 1.265625,
      "learning_rate": 5.480607680638082e-05,
      "loss": 5.3646,
      "mean_token_accuracy": 0.19822666496038438,
      "num_tokens": 19283543.0,
      "step": 8810
    },
    {
      "entropy": 6.266515731811523,
      "epoch": 9.407901761879337,
      "grad_norm": 1.2265625,
      "learning_rate": 5.47195527815862e-05,
      "loss": 5.359,
      "mean_token_accuracy": 0.1917651429772377,
      "num_tokens": 19294399.0,
      "step": 8815
    },
    {
      "entropy": 6.286578798294068,
      "epoch": 9.413240790176188,
      "grad_norm": 1.1875,
      "learning_rate": 5.4633806440535536e-05,
      "loss": 5.3914,
      "mean_token_accuracy": 0.1982104390859604,
      "num_tokens": 19305971.0,
      "step": 8820
    },
    {
      "entropy": 6.293573045730591,
      "epoch": 9.418579818473038,
      "grad_norm": 1.34375,
      "learning_rate": 5.454883808594996e-05,
      "loss": 5.3873,
      "mean_token_accuracy": 0.19418561160564424,
      "num_tokens": 19315983.0,
      "step": 8825
    },
    {
      "entropy": 6.191365814208984,
      "epoch": 9.423918846769888,
      "grad_norm": 1.2421875,
      "learning_rate": 5.446464801780392e-05,
      "loss": 5.2988,
      "mean_token_accuracy": 0.20409801602363586,
      "num_tokens": 19326967.0,
      "step": 8830
    },
    {
      "entropy": 6.2814757347106935,
      "epoch": 9.429257875066737,
      "grad_norm": 1.109375,
      "learning_rate": 5.438123653332416e-05,
      "loss": 5.3361,
      "mean_token_accuracy": 0.19910025894641875,
      "num_tokens": 19338188.0,
      "step": 8835
    },
    {
      "entropy": 6.30628547668457,
      "epoch": 9.434596903363587,
      "grad_norm": 1.2734375,
      "learning_rate": 5.42986039269887e-05,
      "loss": 5.4314,
      "mean_token_accuracy": 0.19197482466697693,
      "num_tokens": 19348319.0,
      "step": 8840
    },
    {
      "entropy": 6.286158275604248,
      "epoch": 9.439935931660438,
      "grad_norm": 1.1953125,
      "learning_rate": 5.4216750490525896e-05,
      "loss": 5.3494,
      "mean_token_accuracy": 0.1952256977558136,
      "num_tokens": 19358764.0,
      "step": 8845
    },
    {
      "entropy": 6.287650728225708,
      "epoch": 9.445274959957288,
      "grad_norm": 1.234375,
      "learning_rate": 5.413567651291315e-05,
      "loss": 5.3961,
      "mean_token_accuracy": 0.1935838997364044,
      "num_tokens": 19369380.0,
      "step": 8850
    },
    {
      "entropy": 6.200163459777832,
      "epoch": 9.450613988254137,
      "grad_norm": 1.015625,
      "learning_rate": 5.4055382280376195e-05,
      "loss": 5.2627,
      "mean_token_accuracy": 0.20308074206113816,
      "num_tokens": 19381819.0,
      "step": 8855
    },
    {
      "entropy": 6.321351432800293,
      "epoch": 9.455953016550987,
      "grad_norm": 1.328125,
      "learning_rate": 5.397586807638774e-05,
      "loss": 5.4202,
      "mean_token_accuracy": 0.19185348302125932,
      "num_tokens": 19391969.0,
      "step": 8860
    },
    {
      "entropy": 6.30733060836792,
      "epoch": 9.461292044847838,
      "grad_norm": 1.2421875,
      "learning_rate": 5.3897134181666886e-05,
      "loss": 5.3968,
      "mean_token_accuracy": 0.2015586495399475,
      "num_tokens": 19402866.0,
      "step": 8865
    },
    {
      "entropy": 6.307026195526123,
      "epoch": 9.466631073144688,
      "grad_norm": 1.296875,
      "learning_rate": 5.3819180874177704e-05,
      "loss": 5.3862,
      "mean_token_accuracy": 0.19666852056980133,
      "num_tokens": 19412929.0,
      "step": 8870
    },
    {
      "entropy": 6.26625394821167,
      "epoch": 9.471970101441539,
      "grad_norm": 1.203125,
      "learning_rate": 5.374200842912858e-05,
      "loss": 5.4024,
      "mean_token_accuracy": 0.1985020011663437,
      "num_tokens": 19424594.0,
      "step": 8875
    },
    {
      "entropy": 6.25069670677185,
      "epoch": 9.477309129738387,
      "grad_norm": 1.2265625,
      "learning_rate": 5.3665617118971134e-05,
      "loss": 5.4028,
      "mean_token_accuracy": 0.19581619799137115,
      "num_tokens": 19435893.0,
      "step": 8880
    },
    {
      "entropy": 6.308282089233399,
      "epoch": 9.482648158035238,
      "grad_norm": 1.1015625,
      "learning_rate": 5.359000721339914e-05,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1967584028840065,
      "num_tokens": 19447414.0,
      "step": 8885
    },
    {
      "entropy": 6.254154205322266,
      "epoch": 9.487987186332088,
      "grad_norm": 1.0703125,
      "learning_rate": 5.3515178979347814e-05,
      "loss": 5.401,
      "mean_token_accuracy": 0.196736142039299,
      "num_tokens": 19458829.0,
      "step": 8890
    },
    {
      "entropy": 6.310901260375976,
      "epoch": 9.493326214628938,
      "grad_norm": 1.046875,
      "learning_rate": 5.3441132680992656e-05,
      "loss": 5.4267,
      "mean_token_accuracy": 0.1950685203075409,
      "num_tokens": 19471630.0,
      "step": 8895
    },
    {
      "entropy": 6.291153192520142,
      "epoch": 9.498665242925787,
      "grad_norm": 1.2890625,
      "learning_rate": 5.3367868579748584e-05,
      "loss": 5.398,
      "mean_token_accuracy": 0.1950012594461441,
      "num_tokens": 19481740.0,
      "step": 8900
    },
    {
      "entropy": 6.238599157333374,
      "epoch": 9.504004271222637,
      "grad_norm": 1.203125,
      "learning_rate": 5.329538693426908e-05,
      "loss": 5.2768,
      "mean_token_accuracy": 0.20877552032470703,
      "num_tokens": 19492165.0,
      "step": 8905
    },
    {
      "entropy": 6.301410770416259,
      "epoch": 9.509343299519488,
      "grad_norm": 1.296875,
      "learning_rate": 5.32236880004452e-05,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19929478764533998,
      "num_tokens": 19502303.0,
      "step": 8910
    },
    {
      "entropy": 6.261133909225464,
      "epoch": 9.514682327816338,
      "grad_norm": 1.2109375,
      "learning_rate": 5.315277203140473e-05,
      "loss": 5.3188,
      "mean_token_accuracy": 0.2059570699930191,
      "num_tokens": 19513124.0,
      "step": 8915
    },
    {
      "entropy": 6.27139687538147,
      "epoch": 9.520021356113187,
      "grad_norm": 1.28125,
      "learning_rate": 5.308263927751119e-05,
      "loss": 5.3821,
      "mean_token_accuracy": 0.19859732836484909,
      "num_tokens": 19523896.0,
      "step": 8920
    },
    {
      "entropy": 6.256344938278199,
      "epoch": 9.525360384410037,
      "grad_norm": 1.1171875,
      "learning_rate": 5.3013289986363083e-05,
      "loss": 5.3545,
      "mean_token_accuracy": 0.19902739077806472,
      "num_tokens": 19535121.0,
      "step": 8925
    },
    {
      "entropy": 6.313183736801148,
      "epoch": 9.530699412706888,
      "grad_norm": 1.140625,
      "learning_rate": 5.294472440279293e-05,
      "loss": 5.439,
      "mean_token_accuracy": 0.19161880016326904,
      "num_tokens": 19546711.0,
      "step": 8930
    },
    {
      "entropy": 6.276342391967773,
      "epoch": 9.536038441003738,
      "grad_norm": 1.3125,
      "learning_rate": 5.2876942768866385e-05,
      "loss": 5.4662,
      "mean_token_accuracy": 0.19173189252614975,
      "num_tokens": 19558628.0,
      "step": 8935
    },
    {
      "entropy": 6.283579540252686,
      "epoch": 9.541377469300587,
      "grad_norm": 1.2109375,
      "learning_rate": 5.280994532388149e-05,
      "loss": 5.3781,
      "mean_token_accuracy": 0.19563505798578262,
      "num_tokens": 19569152.0,
      "step": 8940
    },
    {
      "entropy": 6.342125988006591,
      "epoch": 9.546716497597437,
      "grad_norm": 1.2890625,
      "learning_rate": 5.274373230436775e-05,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19539266228675842,
      "num_tokens": 19579824.0,
      "step": 8945
    },
    {
      "entropy": 6.3311162948608395,
      "epoch": 9.552055525894287,
      "grad_norm": 1.1640625,
      "learning_rate": 5.267830394408531e-05,
      "loss": 5.44,
      "mean_token_accuracy": 0.18797385543584824,
      "num_tokens": 19590315.0,
      "step": 8950
    },
    {
      "entropy": 6.299485445022583,
      "epoch": 9.557394554191138,
      "grad_norm": 1.125,
      "learning_rate": 5.261366047402413e-05,
      "loss": 5.3556,
      "mean_token_accuracy": 0.20104770362377167,
      "num_tokens": 19601561.0,
      "step": 8955
    },
    {
      "entropy": 6.261062526702881,
      "epoch": 9.562733582487986,
      "grad_norm": 1.2421875,
      "learning_rate": 5.2549802122403195e-05,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19275831133127214,
      "num_tokens": 19613662.0,
      "step": 8960
    },
    {
      "entropy": 6.329975318908692,
      "epoch": 9.568072610784837,
      "grad_norm": 1.1328125,
      "learning_rate": 5.248672911466959e-05,
      "loss": 5.426,
      "mean_token_accuracy": 0.19204928278923034,
      "num_tokens": 19625525.0,
      "step": 8965
    },
    {
      "entropy": 6.266425085067749,
      "epoch": 9.573411639081687,
      "grad_norm": 1.25,
      "learning_rate": 5.2424441673498e-05,
      "loss": 5.409,
      "mean_token_accuracy": 0.19519365727901458,
      "num_tokens": 19636505.0,
      "step": 8970
    },
    {
      "entropy": 6.287335205078125,
      "epoch": 9.578750667378538,
      "grad_norm": 1.2109375,
      "learning_rate": 5.23629400187895e-05,
      "loss": 5.3879,
      "mean_token_accuracy": 0.1991070419549942,
      "num_tokens": 19647314.0,
      "step": 8975
    },
    {
      "entropy": 6.318078947067261,
      "epoch": 9.584089695675386,
      "grad_norm": 1.1640625,
      "learning_rate": 5.2302224367671166e-05,
      "loss": 5.3994,
      "mean_token_accuracy": 0.19404184222221374,
      "num_tokens": 19658052.0,
      "step": 8980
    },
    {
      "entropy": 6.295307159423828,
      "epoch": 9.589428723972237,
      "grad_norm": 1.234375,
      "learning_rate": 5.224229493449505e-05,
      "loss": 5.3536,
      "mean_token_accuracy": 0.20603204965591432,
      "num_tokens": 19668589.0,
      "step": 8985
    },
    {
      "entropy": 6.251272344589234,
      "epoch": 9.594767752269087,
      "grad_norm": 1.328125,
      "learning_rate": 5.2183151930837626e-05,
      "loss": 5.4086,
      "mean_token_accuracy": 0.19251115918159484,
      "num_tokens": 19679759.0,
      "step": 8990
    },
    {
      "entropy": 6.297881317138672,
      "epoch": 9.600106780565937,
      "grad_norm": 1.203125,
      "learning_rate": 5.212479556549883e-05,
      "loss": 5.388,
      "mean_token_accuracy": 0.1927838757634163,
      "num_tokens": 19690785.0,
      "step": 8995
    },
    {
      "entropy": 6.299252557754516,
      "epoch": 9.605445808862786,
      "grad_norm": 1.1953125,
      "learning_rate": 5.206722604450148e-05,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19994686990976335,
      "num_tokens": 19702394.0,
      "step": 9000
    },
    {
      "epoch": 9.605445808862786,
      "eval_entropy": 6.120275040415974,
      "eval_loss": 6.550008296966553,
      "eval_mean_token_accuracy": 0.13517576714421248,
      "eval_num_tokens": 19702394.0,
      "eval_runtime": 1.8103,
      "eval_samples_per_second": 1701.367,
      "eval_steps_per_second": 212.671,
      "step": 9000
    },
    {
      "entropy": 6.290256118774414,
      "epoch": 9.610784837159636,
      "grad_norm": 1.1484375,
      "learning_rate": 5.201044357109047e-05,
      "loss": 5.3599,
      "mean_token_accuracy": 0.2012845516204834,
      "num_tokens": 19713137.0,
      "step": 9005
    },
    {
      "entropy": 6.23082799911499,
      "epoch": 9.616123865456487,
      "grad_norm": 1.21875,
      "learning_rate": 5.1954448345732133e-05,
      "loss": 5.2897,
      "mean_token_accuracy": 0.2032169759273529,
      "num_tokens": 19724302.0,
      "step": 9010
    },
    {
      "entropy": 6.314432287216187,
      "epoch": 9.621462893753337,
      "grad_norm": 1.28125,
      "learning_rate": 5.1899240566113436e-05,
      "loss": 5.438,
      "mean_token_accuracy": 0.1953733667731285,
      "num_tokens": 19735302.0,
      "step": 9015
    },
    {
      "entropy": 6.301742839813232,
      "epoch": 9.626801922050188,
      "grad_norm": 1.390625,
      "learning_rate": 5.184482042714132e-05,
      "loss": 5.4432,
      "mean_token_accuracy": 0.19119120687246322,
      "num_tokens": 19747089.0,
      "step": 9020
    },
    {
      "entropy": 6.253672790527344,
      "epoch": 9.632140950347036,
      "grad_norm": 1.234375,
      "learning_rate": 5.1791188120942075e-05,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20582606345415116,
      "num_tokens": 19756974.0,
      "step": 9025
    },
    {
      "entropy": 6.278188705444336,
      "epoch": 9.637479978643887,
      "grad_norm": 1.21875,
      "learning_rate": 5.173834383686056e-05,
      "loss": 5.355,
      "mean_token_accuracy": 0.19714901596307755,
      "num_tokens": 19767556.0,
      "step": 9030
    },
    {
      "entropy": 6.284007501602173,
      "epoch": 9.642819006940737,
      "grad_norm": 1.28125,
      "learning_rate": 5.1686287761459565e-05,
      "loss": 5.4472,
      "mean_token_accuracy": 0.19691016972064973,
      "num_tokens": 19777123.0,
      "step": 9035
    },
    {
      "entropy": 6.324623680114746,
      "epoch": 9.648158035237588,
      "grad_norm": 1.3125,
      "learning_rate": 5.1635020078519227e-05,
      "loss": 5.4835,
      "mean_token_accuracy": 0.18013687282800675,
      "num_tokens": 19788244.0,
      "step": 9040
    },
    {
      "entropy": 6.326438760757446,
      "epoch": 9.653497063534436,
      "grad_norm": 1.078125,
      "learning_rate": 5.158454096903631e-05,
      "loss": 5.426,
      "mean_token_accuracy": 0.1968817189335823,
      "num_tokens": 19799327.0,
      "step": 9045
    },
    {
      "entropy": 6.288557815551758,
      "epoch": 9.658836091831287,
      "grad_norm": 1.203125,
      "learning_rate": 5.153485061122347e-05,
      "loss": 5.3892,
      "mean_token_accuracy": 0.1961186483502388,
      "num_tokens": 19809672.0,
      "step": 9050
    },
    {
      "entropy": 6.319461345672607,
      "epoch": 9.664175120128137,
      "grad_norm": 1.375,
      "learning_rate": 5.148594918050893e-05,
      "loss": 5.4234,
      "mean_token_accuracy": 0.1943323254585266,
      "num_tokens": 19820194.0,
      "step": 9055
    },
    {
      "entropy": 6.23256516456604,
      "epoch": 9.669514148424987,
      "grad_norm": 1.2734375,
      "learning_rate": 5.143783684953549e-05,
      "loss": 5.3734,
      "mean_token_accuracy": 0.2008483424782753,
      "num_tokens": 19830457.0,
      "step": 9060
    },
    {
      "entropy": 6.290764045715332,
      "epoch": 9.674853176721836,
      "grad_norm": 1.2265625,
      "learning_rate": 5.1390513788160164e-05,
      "loss": 5.4343,
      "mean_token_accuracy": 0.1943418025970459,
      "num_tokens": 19840901.0,
      "step": 9065
    },
    {
      "entropy": 6.290750217437744,
      "epoch": 9.680192205018686,
      "grad_norm": 1.0546875,
      "learning_rate": 5.1343980163453513e-05,
      "loss": 5.4292,
      "mean_token_accuracy": 0.1949126899242401,
      "num_tokens": 19852352.0,
      "step": 9070
    },
    {
      "entropy": 6.243222332000732,
      "epoch": 9.685531233315537,
      "grad_norm": 1.1484375,
      "learning_rate": 5.129823613969904e-05,
      "loss": 5.4514,
      "mean_token_accuracy": 0.1888757422566414,
      "num_tokens": 19863351.0,
      "step": 9075
    },
    {
      "entropy": 6.196452045440674,
      "epoch": 9.690870261612387,
      "grad_norm": 1.3125,
      "learning_rate": 5.125328187839262e-05,
      "loss": 5.3105,
      "mean_token_accuracy": 0.20151706486940385,
      "num_tokens": 19875871.0,
      "step": 9080
    },
    {
      "entropy": 6.292422723770142,
      "epoch": 9.696209289909236,
      "grad_norm": 1.0625,
      "learning_rate": 5.120911753824192e-05,
      "loss": 5.4469,
      "mean_token_accuracy": 0.19452553540468215,
      "num_tokens": 19887492.0,
      "step": 9085
    },
    {
      "entropy": 6.342274904251099,
      "epoch": 9.701548318206086,
      "grad_norm": 1.203125,
      "learning_rate": 5.1165743275165885e-05,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1942036047577858,
      "num_tokens": 19899306.0,
      "step": 9090
    },
    {
      "entropy": 6.291375207901001,
      "epoch": 9.706887346502937,
      "grad_norm": 1.1328125,
      "learning_rate": 5.112315924229409e-05,
      "loss": 5.4073,
      "mean_token_accuracy": 0.19083765149116516,
      "num_tokens": 19910412.0,
      "step": 9095
    },
    {
      "entropy": 6.3185607433319095,
      "epoch": 9.712226374799787,
      "grad_norm": 1.2734375,
      "learning_rate": 5.108136558996634e-05,
      "loss": 5.406,
      "mean_token_accuracy": 0.1944574609398842,
      "num_tokens": 19921718.0,
      "step": 9100
    },
    {
      "entropy": 6.285332441329956,
      "epoch": 9.717565403096636,
      "grad_norm": 1.171875,
      "learning_rate": 5.1040362465732e-05,
      "loss": 5.4201,
      "mean_token_accuracy": 0.19195159077644347,
      "num_tokens": 19933192.0,
      "step": 9105
    },
    {
      "entropy": 6.2487945556640625,
      "epoch": 9.722904431393486,
      "grad_norm": 1.1484375,
      "learning_rate": 5.100015001434953e-05,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19561336785554886,
      "num_tokens": 19944398.0,
      "step": 9110
    },
    {
      "entropy": 6.312276983261109,
      "epoch": 9.728243459690336,
      "grad_norm": 1.265625,
      "learning_rate": 5.096072837778601e-05,
      "loss": 5.3652,
      "mean_token_accuracy": 0.19846275001764296,
      "num_tokens": 19954625.0,
      "step": 9115
    },
    {
      "entropy": 6.2715788841247555,
      "epoch": 9.733582487987187,
      "grad_norm": 1.1875,
      "learning_rate": 5.092209769521661e-05,
      "loss": 5.3302,
      "mean_token_accuracy": 0.20155353397130965,
      "num_tokens": 19964960.0,
      "step": 9120
    },
    {
      "entropy": 6.2791149616241455,
      "epoch": 9.738921516284037,
      "grad_norm": 1.359375,
      "learning_rate": 5.0884258103024104e-05,
      "loss": 5.3735,
      "mean_token_accuracy": 0.1931841790676117,
      "num_tokens": 19974947.0,
      "step": 9125
    },
    {
      "entropy": 6.304941177368164,
      "epoch": 9.744260544580886,
      "grad_norm": 1.28125,
      "learning_rate": 5.084720973479831e-05,
      "loss": 5.4302,
      "mean_token_accuracy": 0.19329258054494858,
      "num_tokens": 19985498.0,
      "step": 9130
    },
    {
      "entropy": 6.26310715675354,
      "epoch": 9.749599572877736,
      "grad_norm": 1.0703125,
      "learning_rate": 5.081095272133579e-05,
      "loss": 5.3031,
      "mean_token_accuracy": 0.21339126229286193,
      "num_tokens": 19997663.0,
      "step": 9135
    },
    {
      "entropy": 6.320000219345093,
      "epoch": 9.754938601174587,
      "grad_norm": 1.3046875,
      "learning_rate": 5.0775487190639164e-05,
      "loss": 5.3627,
      "mean_token_accuracy": 0.19529865086078643,
      "num_tokens": 20008580.0,
      "step": 9140
    },
    {
      "entropy": 6.271733522415161,
      "epoch": 9.760277629471435,
      "grad_norm": 1.0078125,
      "learning_rate": 5.074081326791692e-05,
      "loss": 5.4209,
      "mean_token_accuracy": 0.19285557717084884,
      "num_tokens": 20020760.0,
      "step": 9145
    },
    {
      "entropy": 6.282210397720337,
      "epoch": 9.765616657768286,
      "grad_norm": 1.2578125,
      "learning_rate": 5.070693107558274e-05,
      "loss": 5.3671,
      "mean_token_accuracy": 0.19827128648757936,
      "num_tokens": 20031751.0,
      "step": 9150
    },
    {
      "entropy": 6.303864002227783,
      "epoch": 9.770955686065136,
      "grad_norm": 1.171875,
      "learning_rate": 5.067384073325518e-05,
      "loss": 5.4921,
      "mean_token_accuracy": 0.18973946273326875,
      "num_tokens": 20042293.0,
      "step": 9155
    },
    {
      "entropy": 6.274455881118774,
      "epoch": 9.776294714361986,
      "grad_norm": 1.140625,
      "learning_rate": 5.064154235775724e-05,
      "loss": 5.3849,
      "mean_token_accuracy": 0.19063759595155716,
      "num_tokens": 20053110.0,
      "step": 9160
    },
    {
      "entropy": 6.263268613815308,
      "epoch": 9.781633742658837,
      "grad_norm": 1.390625,
      "learning_rate": 5.061003606311592e-05,
      "loss": 5.3148,
      "mean_token_accuracy": 0.1992083564400673,
      "num_tokens": 20063306.0,
      "step": 9165
    },
    {
      "entropy": 6.246716165542603,
      "epoch": 9.786972770955686,
      "grad_norm": 1.1015625,
      "learning_rate": 5.057932196056189e-05,
      "loss": 5.3925,
      "mean_token_accuracy": 0.195957013964653,
      "num_tokens": 20074716.0,
      "step": 9170
    },
    {
      "entropy": 6.354883337020874,
      "epoch": 9.792311799252536,
      "grad_norm": 1.375,
      "learning_rate": 5.0549400158528966e-05,
      "loss": 5.506,
      "mean_token_accuracy": 0.18127121031284332,
      "num_tokens": 20084944.0,
      "step": 9175
    },
    {
      "entropy": 6.325537919998169,
      "epoch": 9.797650827549386,
      "grad_norm": 1.203125,
      "learning_rate": 5.0520270762653866e-05,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19326557219028473,
      "num_tokens": 20095436.0,
      "step": 9180
    },
    {
      "entropy": 6.329968738555908,
      "epoch": 9.802989855846237,
      "grad_norm": 1.1875,
      "learning_rate": 5.049193387577574e-05,
      "loss": 5.461,
      "mean_token_accuracy": 0.18755955547094344,
      "num_tokens": 20106757.0,
      "step": 9185
    },
    {
      "entropy": 6.335523796081543,
      "epoch": 9.808328884143085,
      "grad_norm": 1.265625,
      "learning_rate": 5.046438959793586e-05,
      "loss": 5.4066,
      "mean_token_accuracy": 0.19587553143501282,
      "num_tokens": 20118122.0,
      "step": 9190
    },
    {
      "entropy": 6.2556627750396725,
      "epoch": 9.813667912439936,
      "grad_norm": 1.265625,
      "learning_rate": 5.0437638026377274e-05,
      "loss": 5.3486,
      "mean_token_accuracy": 0.199265056848526,
      "num_tokens": 20129948.0,
      "step": 9195
    },
    {
      "entropy": 6.27778000831604,
      "epoch": 9.819006940736786,
      "grad_norm": 1.21875,
      "learning_rate": 5.0411679255544405e-05,
      "loss": 5.4343,
      "mean_token_accuracy": 0.19429014027118682,
      "num_tokens": 20140421.0,
      "step": 9200
    },
    {
      "entropy": 6.258305788040161,
      "epoch": 9.824345969033637,
      "grad_norm": 1.0546875,
      "learning_rate": 5.0386513377082756e-05,
      "loss": 5.3836,
      "mean_token_accuracy": 0.19233323484659196,
      "num_tokens": 20152399.0,
      "step": 9205
    },
    {
      "entropy": 6.319499349594116,
      "epoch": 9.829684997330485,
      "grad_norm": 1.171875,
      "learning_rate": 5.036214047983859e-05,
      "loss": 5.4427,
      "mean_token_accuracy": 0.19616758078336716,
      "num_tokens": 20162893.0,
      "step": 9210
    },
    {
      "entropy": 6.306321716308593,
      "epoch": 9.835024025627336,
      "grad_norm": 1.3046875,
      "learning_rate": 5.033856064985863e-05,
      "loss": 5.3981,
      "mean_token_accuracy": 0.19886463433504104,
      "num_tokens": 20173695.0,
      "step": 9215
    },
    {
      "entropy": 6.291854619979858,
      "epoch": 9.840363053924186,
      "grad_norm": 1.1640625,
      "learning_rate": 5.031577397038969e-05,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1900296166539192,
      "num_tokens": 20185147.0,
      "step": 9220
    },
    {
      "entropy": 6.295867538452148,
      "epoch": 9.845702082221036,
      "grad_norm": 1.2265625,
      "learning_rate": 5.029378052187848e-05,
      "loss": 5.4641,
      "mean_token_accuracy": 0.1876828297972679,
      "num_tokens": 20196665.0,
      "step": 9225
    },
    {
      "entropy": 6.273441457748413,
      "epoch": 9.851041110517885,
      "grad_norm": 1.2890625,
      "learning_rate": 5.02725803819712e-05,
      "loss": 5.336,
      "mean_token_accuracy": 0.20544504076242448,
      "num_tokens": 20207427.0,
      "step": 9230
    },
    {
      "entropy": 6.351820373535157,
      "epoch": 9.856380138814735,
      "grad_norm": 1.2421875,
      "learning_rate": 5.025217362551337e-05,
      "loss": 5.4591,
      "mean_token_accuracy": 0.1901904195547104,
      "num_tokens": 20220705.0,
      "step": 9235
    },
    {
      "entropy": 6.229749298095703,
      "epoch": 9.861719167111586,
      "grad_norm": 1.3515625,
      "learning_rate": 5.023256032454956e-05,
      "loss": 5.2338,
      "mean_token_accuracy": 0.21689044684171677,
      "num_tokens": 20230241.0,
      "step": 9240
    },
    {
      "entropy": 6.208050775527954,
      "epoch": 9.867058195408436,
      "grad_norm": 1.2265625,
      "learning_rate": 5.021374054832304e-05,
      "loss": 5.3703,
      "mean_token_accuracy": 0.1989450752735138,
      "num_tokens": 20241813.0,
      "step": 9245
    },
    {
      "entropy": 6.225540447235107,
      "epoch": 9.872397223705285,
      "grad_norm": 1.265625,
      "learning_rate": 5.0195714363275615e-05,
      "loss": 5.2901,
      "mean_token_accuracy": 0.19797505140304567,
      "num_tokens": 20253359.0,
      "step": 9250
    },
    {
      "entropy": 6.248586368560791,
      "epoch": 9.877736252002135,
      "grad_norm": 1.25,
      "learning_rate": 5.017848183304743e-05,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20024796575307846,
      "num_tokens": 20263377.0,
      "step": 9255
    },
    {
      "entropy": 6.304415845870972,
      "epoch": 9.883075280298986,
      "grad_norm": 1.1484375,
      "learning_rate": 5.016204301847665e-05,
      "loss": 5.3958,
      "mean_token_accuracy": 0.20311103016138077,
      "num_tokens": 20274734.0,
      "step": 9260
    },
    {
      "entropy": 6.334533166885376,
      "epoch": 9.888414308595836,
      "grad_norm": 1.1015625,
      "learning_rate": 5.0146397977599264e-05,
      "loss": 5.4649,
      "mean_token_accuracy": 0.18885774165391922,
      "num_tokens": 20287517.0,
      "step": 9265
    },
    {
      "entropy": 6.31162919998169,
      "epoch": 9.893753336892686,
      "grad_norm": 1.3203125,
      "learning_rate": 5.013154676564892e-05,
      "loss": 5.372,
      "mean_token_accuracy": 0.19851657450199128,
      "num_tokens": 20297683.0,
      "step": 9270
    },
    {
      "entropy": 6.285402011871338,
      "epoch": 9.899092365189535,
      "grad_norm": 1.2421875,
      "learning_rate": 5.0117489435056754e-05,
      "loss": 5.3569,
      "mean_token_accuracy": 0.19836218655109406,
      "num_tokens": 20307947.0,
      "step": 9275
    },
    {
      "entropy": 6.229357671737671,
      "epoch": 9.904431393486385,
      "grad_norm": 1.1328125,
      "learning_rate": 5.010422603545112e-05,
      "loss": 5.3742,
      "mean_token_accuracy": 0.19662382155656816,
      "num_tokens": 20319838.0,
      "step": 9280
    },
    {
      "entropy": 6.2321373462677006,
      "epoch": 9.909770421783236,
      "grad_norm": 1.203125,
      "learning_rate": 5.009175661365743e-05,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2050779268145561,
      "num_tokens": 20330467.0,
      "step": 9285
    },
    {
      "entropy": 6.306796312332153,
      "epoch": 9.915109450080084,
      "grad_norm": 1.2109375,
      "learning_rate": 5.0080081213698074e-05,
      "loss": 5.4319,
      "mean_token_accuracy": 0.19407758861780167,
      "num_tokens": 20341686.0,
      "step": 9290
    },
    {
      "entropy": 6.304175949096679,
      "epoch": 9.920448478376935,
      "grad_norm": 1.171875,
      "learning_rate": 5.00691998767922e-05,
      "loss": 5.4504,
      "mean_token_accuracy": 0.19805026203393936,
      "num_tokens": 20352798.0,
      "step": 9295
    },
    {
      "entropy": 6.31003246307373,
      "epoch": 9.925787506673785,
      "grad_norm": 1.265625,
      "learning_rate": 5.005911264135551e-05,
      "loss": 5.4038,
      "mean_token_accuracy": 0.19758603423833848,
      "num_tokens": 20363577.0,
      "step": 9300
    },
    {
      "entropy": 6.33183217048645,
      "epoch": 9.931126534970636,
      "grad_norm": 1.203125,
      "learning_rate": 5.0049819543000285e-05,
      "loss": 5.4338,
      "mean_token_accuracy": 0.1869886964559555,
      "num_tokens": 20374028.0,
      "step": 9305
    },
    {
      "entropy": 6.210221481323242,
      "epoch": 9.936465563267486,
      "grad_norm": 1.140625,
      "learning_rate": 5.004132061453512e-05,
      "loss": 5.3039,
      "mean_token_accuracy": 0.20492296367883683,
      "num_tokens": 20384907.0,
      "step": 9310
    },
    {
      "entropy": 6.281497192382813,
      "epoch": 9.941804591564335,
      "grad_norm": 1.171875,
      "learning_rate": 5.003361588596482e-05,
      "loss": 5.3362,
      "mean_token_accuracy": 0.1993270546197891,
      "num_tokens": 20396223.0,
      "step": 9315
    },
    {
      "entropy": 6.27079496383667,
      "epoch": 9.947143619861185,
      "grad_norm": 1.2265625,
      "learning_rate": 5.0026705384490366e-05,
      "loss": 5.3659,
      "mean_token_accuracy": 0.1996488407254219,
      "num_tokens": 20406450.0,
      "step": 9320
    },
    {
      "entropy": 6.283566093444824,
      "epoch": 9.952482648158036,
      "grad_norm": 1.2578125,
      "learning_rate": 5.002058913450876e-05,
      "loss": 5.355,
      "mean_token_accuracy": 0.1983364626765251,
      "num_tokens": 20417407.0,
      "step": 9325
    },
    {
      "entropy": 6.266901111602783,
      "epoch": 9.957821676454886,
      "grad_norm": 1.3359375,
      "learning_rate": 5.001526715761301e-05,
      "loss": 5.3673,
      "mean_token_accuracy": 0.19686105400323867,
      "num_tokens": 20427894.0,
      "step": 9330
    },
    {
      "entropy": 6.256697273254394,
      "epoch": 9.963160704751735,
      "grad_norm": 1.2578125,
      "learning_rate": 5.00107394725919e-05,
      "loss": 5.4222,
      "mean_token_accuracy": 0.1950857698917389,
      "num_tokens": 20437842.0,
      "step": 9335
    },
    {
      "entropy": 6.202639198303222,
      "epoch": 9.968499733048585,
      "grad_norm": 1.234375,
      "learning_rate": 5.0007006095430134e-05,
      "loss": 5.332,
      "mean_token_accuracy": 0.2034511983394623,
      "num_tokens": 20448266.0,
      "step": 9340
    },
    {
      "entropy": 6.286227416992188,
      "epoch": 9.973838761345435,
      "grad_norm": 1.3828125,
      "learning_rate": 5.000406703930811e-05,
      "loss": 5.4168,
      "mean_token_accuracy": 0.1973090097308159,
      "num_tokens": 20458328.0,
      "step": 9345
    },
    {
      "entropy": 6.255052375793457,
      "epoch": 9.979177789642286,
      "grad_norm": 1.2109375,
      "learning_rate": 5.0001922314601955e-05,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20041885673999787,
      "num_tokens": 20468523.0,
      "step": 9350
    },
    {
      "entropy": 6.240312576293945,
      "epoch": 9.984516817939134,
      "grad_norm": 1.140625,
      "learning_rate": 5.000057192888341e-05,
      "loss": 5.3593,
      "mean_token_accuracy": 0.19507021754980086,
      "num_tokens": 20478974.0,
      "step": 9355
    },
    {
      "entropy": 6.268662214279175,
      "epoch": 9.989855846235985,
      "grad_norm": 1.2265625,
      "learning_rate": 5.000001588691998e-05,
      "loss": 5.4356,
      "mean_token_accuracy": 0.19796158224344254,
      "num_tokens": 20490325.0,
      "step": 9360
    }
  ],
  "logging_steps": 5,
  "max_steps": 9360,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4452512564736000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}