eng-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10/checkpoint-9000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.730124391563008,
  "eval_steps": 1000,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.91689715385437,
      "epoch": 0.005408328826392645,
      "grad_norm": 19.875,
      "learning_rate": 2e-06,
      "loss": 14.1759,
      "mean_token_accuracy": 9.372071363031865e-05,
      "num_tokens": 10948.0,
      "step": 5
    },
    {
      "entropy": 4.94269962310791,
      "epoch": 0.01081665765278529,
      "grad_norm": 23.75,
      "learning_rate": 4.5e-06,
      "loss": 14.1132,
      "mean_token_accuracy": 0.0,
      "num_tokens": 21455.0,
      "step": 10
    },
    {
      "entropy": 4.975670337677002,
      "epoch": 0.016224986479177934,
      "grad_norm": 27.625,
      "learning_rate": 7e-06,
      "loss": 14.0429,
      "mean_token_accuracy": 8.873114711605012e-05,
      "num_tokens": 32377.0,
      "step": 15
    },
    {
      "entropy": 5.118604516983032,
      "epoch": 0.02163331530557058,
      "grad_norm": 32.75,
      "learning_rate": 9.5e-06,
      "loss": 13.6272,
      "mean_token_accuracy": 6.706907879561186e-05,
      "num_tokens": 42276.0,
      "step": 20
    },
    {
      "entropy": 5.543013238906861,
      "epoch": 0.02704164413196322,
      "grad_norm": 40.75,
      "learning_rate": 1.2e-05,
      "loss": 13.0208,
      "mean_token_accuracy": 0.00013905229279771448,
      "num_tokens": 55049.0,
      "step": 25
    },
    {
      "entropy": 7.297787666320801,
      "epoch": 0.03244997295835587,
      "grad_norm": 26.375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 11.8384,
      "mean_token_accuracy": 0.0,
      "num_tokens": 66792.0,
      "step": 30
    },
    {
      "entropy": 10.221694660186767,
      "epoch": 0.03785830178474851,
      "grad_norm": 3.640625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.8659,
      "mean_token_accuracy": 0.004666937288129702,
      "num_tokens": 78208.0,
      "step": 35
    },
    {
      "entropy": 10.73062868118286,
      "epoch": 0.04326663061114116,
      "grad_norm": 3.359375,
      "learning_rate": 1.95e-05,
      "loss": 10.6432,
      "mean_token_accuracy": 0.013734118174761533,
      "num_tokens": 88871.0,
      "step": 40
    },
    {
      "entropy": 10.73009204864502,
      "epoch": 0.048674959437533805,
      "grad_norm": 3.140625,
      "learning_rate": 2.2e-05,
      "loss": 10.5305,
      "mean_token_accuracy": 0.0145380774512887,
      "num_tokens": 100118.0,
      "step": 45
    },
    {
      "entropy": 10.722947692871093,
      "epoch": 0.05408328826392644,
      "grad_norm": 2.4375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.362,
      "mean_token_accuracy": 0.012977494951337576,
      "num_tokens": 113120.0,
      "step": 50
    },
    {
      "entropy": 10.69485158920288,
      "epoch": 0.05949161709031909,
      "grad_norm": 2.890625,
      "learning_rate": 2.7e-05,
      "loss": 10.146,
      "mean_token_accuracy": 0.016121995542198418,
      "num_tokens": 123220.0,
      "step": 55
    },
    {
      "entropy": 10.705347919464112,
      "epoch": 0.06489994591671173,
      "grad_norm": 1.984375,
      "learning_rate": 2.95e-05,
      "loss": 10.053,
      "mean_token_accuracy": 0.01566679133102298,
      "num_tokens": 134479.0,
      "step": 60
    },
    {
      "entropy": 10.732111358642578,
      "epoch": 0.07030827474310439,
      "grad_norm": 1.953125,
      "learning_rate": 3.2e-05,
      "loss": 9.9286,
      "mean_token_accuracy": 0.030279947444796562,
      "num_tokens": 145189.0,
      "step": 65
    },
    {
      "entropy": 10.72783441543579,
      "epoch": 0.07571660356949703,
      "grad_norm": 1.96875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.8452,
      "mean_token_accuracy": 0.03374573942273855,
      "num_tokens": 157354.0,
      "step": 70
    },
    {
      "entropy": 10.714603519439697,
      "epoch": 0.08112493239588967,
      "grad_norm": 1.84375,
      "learning_rate": 3.7e-05,
      "loss": 9.7908,
      "mean_token_accuracy": 0.03534814547747374,
      "num_tokens": 168821.0,
      "step": 75
    },
    {
      "entropy": 10.72258472442627,
      "epoch": 0.08653326122228232,
      "grad_norm": 1.8984375,
      "learning_rate": 3.95e-05,
      "loss": 9.7017,
      "mean_token_accuracy": 0.04154699631035328,
      "num_tokens": 180554.0,
      "step": 80
    },
    {
      "entropy": 10.723107147216798,
      "epoch": 0.09194159004867496,
      "grad_norm": 2.09375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.6177,
      "mean_token_accuracy": 0.04423647858202458,
      "num_tokens": 191229.0,
      "step": 85
    },
    {
      "entropy": 10.715191745758057,
      "epoch": 0.09734991887506761,
      "grad_norm": 1.9296875,
      "learning_rate": 4.45e-05,
      "loss": 9.5371,
      "mean_token_accuracy": 0.03949474133551121,
      "num_tokens": 201903.0,
      "step": 90
    },
    {
      "entropy": 10.705009365081787,
      "epoch": 0.10275824770146025,
      "grad_norm": 1.9140625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4731,
      "mean_token_accuracy": 0.03862293735146523,
      "num_tokens": 214255.0,
      "step": 95
    },
    {
      "entropy": 10.693695735931396,
      "epoch": 0.10816657652785289,
      "grad_norm": 1.859375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3346,
      "mean_token_accuracy": 0.04093401432037354,
      "num_tokens": 225423.0,
      "step": 100
    },
    {
      "entropy": 10.675817108154297,
      "epoch": 0.11357490535424554,
      "grad_norm": 1.859375,
      "learning_rate": 5.2e-05,
      "loss": 9.2358,
      "mean_token_accuracy": 0.046506138145923616,
      "num_tokens": 237801.0,
      "step": 105
    },
    {
      "entropy": 10.651765060424804,
      "epoch": 0.11898323418063818,
      "grad_norm": 1.9765625,
      "learning_rate": 5.45e-05,
      "loss": 9.0789,
      "mean_token_accuracy": 0.05033866539597511,
      "num_tokens": 249195.0,
      "step": 110
    },
    {
      "entropy": 10.622098636627197,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.8359375,
      "learning_rate": 5.7e-05,
      "loss": 8.9519,
      "mean_token_accuracy": 0.046295545622706415,
      "num_tokens": 260129.0,
      "step": 115
    },
    {
      "entropy": 10.583237266540527,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.7890625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9331,
      "mean_token_accuracy": 0.044216564670205113,
      "num_tokens": 271220.0,
      "step": 120
    },
    {
      "entropy": 10.535919380187988,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.7890625,
      "learning_rate": 6.2e-05,
      "loss": 8.7308,
      "mean_token_accuracy": 0.05066610909998417,
      "num_tokens": 281560.0,
      "step": 125
    },
    {
      "entropy": 10.480502033233643,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.7109375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.6033,
      "mean_token_accuracy": 0.04720585718750954,
      "num_tokens": 292518.0,
      "step": 130
    },
    {
      "entropy": 10.400840473175048,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.6328125,
      "learning_rate": 6.7e-05,
      "loss": 8.537,
      "mean_token_accuracy": 0.04684793278574943,
      "num_tokens": 304865.0,
      "step": 135
    },
    {
      "entropy": 10.312588691711426,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.65625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.3772,
      "mean_token_accuracy": 0.04892168343067169,
      "num_tokens": 315951.0,
      "step": 140
    },
    {
      "entropy": 10.193289756774902,
      "epoch": 0.1568415359653867,
      "grad_norm": 1.53125,
      "learning_rate": 7.2e-05,
      "loss": 8.2559,
      "mean_token_accuracy": 0.047295660525560376,
      "num_tokens": 327094.0,
      "step": 145
    },
    {
      "entropy": 10.057239818572999,
      "epoch": 0.16224986479177933,
      "grad_norm": 1.625,
      "learning_rate": 7.45e-05,
      "loss": 8.1645,
      "mean_token_accuracy": 0.04913058504462242,
      "num_tokens": 337801.0,
      "step": 150
    },
    {
      "entropy": 9.920266246795654,
      "epoch": 0.167658193618172,
      "grad_norm": 1.2421875,
      "learning_rate": 7.7e-05,
      "loss": 8.0478,
      "mean_token_accuracy": 0.04764632135629654,
      "num_tokens": 349039.0,
      "step": 155
    },
    {
      "entropy": 9.760813426971435,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.265625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9465,
      "mean_token_accuracy": 0.049764279276132584,
      "num_tokens": 360639.0,
      "step": 160
    },
    {
      "entropy": 9.56991138458252,
      "epoch": 0.17847485127095727,
      "grad_norm": 1.03125,
      "learning_rate": 8.2e-05,
      "loss": 7.8804,
      "mean_token_accuracy": 0.05512551851570606,
      "num_tokens": 371005.0,
      "step": 165
    },
    {
      "entropy": 9.318969631195069,
      "epoch": 0.1838831800973499,
      "grad_norm": 1.109375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.7379,
      "mean_token_accuracy": 0.05374485366046429,
      "num_tokens": 383340.0,
      "step": 170
    },
    {
      "entropy": 9.042248249053955,
      "epoch": 0.18929150892374255,
      "grad_norm": 0.875,
      "learning_rate": 8.7e-05,
      "loss": 7.6303,
      "mean_token_accuracy": 0.04950361587107181,
      "num_tokens": 394682.0,
      "step": 175
    },
    {
      "entropy": 8.764017868041993,
      "epoch": 0.19469983775013522,
      "grad_norm": 0.76953125,
      "learning_rate": 8.95e-05,
      "loss": 7.5815,
      "mean_token_accuracy": 0.059978223592042926,
      "num_tokens": 405713.0,
      "step": 180
    },
    {
      "entropy": 8.552699661254882,
      "epoch": 0.20010816657652786,
      "grad_norm": 0.703125,
      "learning_rate": 9.2e-05,
      "loss": 7.5243,
      "mean_token_accuracy": 0.06490898504853249,
      "num_tokens": 416208.0,
      "step": 185
    },
    {
      "entropy": 8.323922729492187,
      "epoch": 0.2055164954029205,
      "grad_norm": 0.73828125,
      "learning_rate": 9.45e-05,
      "loss": 7.4179,
      "mean_token_accuracy": 0.05761545822024346,
      "num_tokens": 427595.0,
      "step": 190
    },
    {
      "entropy": 8.167845249176025,
      "epoch": 0.21092482422931313,
      "grad_norm": 0.8125,
      "learning_rate": 9.7e-05,
      "loss": 7.506,
      "mean_token_accuracy": 0.06468431949615479,
      "num_tokens": 438273.0,
      "step": 195
    },
    {
      "entropy": 8.073835182189942,
      "epoch": 0.21633315305570577,
      "grad_norm": 0.6875,
      "learning_rate": 9.95e-05,
      "loss": 7.4195,
      "mean_token_accuracy": 0.06935816742479801,
      "num_tokens": 448948.0,
      "step": 200
    },
    {
      "entropy": 7.998777389526367,
      "epoch": 0.22174148188209844,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000102,
      "loss": 7.456,
      "mean_token_accuracy": 0.06588310226798058,
      "num_tokens": 460838.0,
      "step": 205
    },
    {
      "entropy": 8.015505456924439,
      "epoch": 0.22714981070849108,
      "grad_norm": 1.375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4476,
      "mean_token_accuracy": 0.06993156149983407,
      "num_tokens": 473555.0,
      "step": 210
    },
    {
      "entropy": 7.928825998306275,
      "epoch": 0.23255813953488372,
      "grad_norm": 0.6875,
      "learning_rate": 0.000107,
      "loss": 7.3803,
      "mean_token_accuracy": 0.07670803219079972,
      "num_tokens": 485221.0,
      "step": 215
    },
    {
      "entropy": 7.916201114654541,
      "epoch": 0.23796646836127636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001095,
      "loss": 7.2914,
      "mean_token_accuracy": 0.07322146371006966,
      "num_tokens": 496552.0,
      "step": 220
    },
    {
      "entropy": 7.902019739151001,
      "epoch": 0.24337479718766902,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000112,
      "loss": 7.3551,
      "mean_token_accuracy": 0.07495715916156769,
      "num_tokens": 509760.0,
      "step": 225
    },
    {
      "entropy": 7.87739930152893,
      "epoch": 0.24878312601406166,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001145,
      "loss": 7.2003,
      "mean_token_accuracy": 0.07767140977084637,
      "num_tokens": 521175.0,
      "step": 230
    },
    {
      "entropy": 7.810295057296753,
      "epoch": 0.2541914548404543,
      "grad_norm": 0.66796875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2954,
      "mean_token_accuracy": 0.07847636789083481,
      "num_tokens": 532597.0,
      "step": 235
    },
    {
      "entropy": 7.815588092803955,
      "epoch": 0.25959978366684694,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3371,
      "mean_token_accuracy": 0.07920736819505692,
      "num_tokens": 545661.0,
      "step": 240
    },
    {
      "entropy": 7.830226469039917,
      "epoch": 0.2650081124932396,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000122,
      "loss": 7.3186,
      "mean_token_accuracy": 0.07971370816230774,
      "num_tokens": 557336.0,
      "step": 245
    },
    {
      "entropy": 7.839285278320313,
      "epoch": 0.2704164413196322,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0001245,
      "loss": 7.1789,
      "mean_token_accuracy": 0.08502659052610398,
      "num_tokens": 567398.0,
      "step": 250
    },
    {
      "entropy": 7.752034616470337,
      "epoch": 0.27582477014602486,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000127,
      "loss": 7.254,
      "mean_token_accuracy": 0.07750963866710663,
      "num_tokens": 579543.0,
      "step": 255
    },
    {
      "entropy": 7.797254037857056,
      "epoch": 0.28123309897241755,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001295,
      "loss": 7.0797,
      "mean_token_accuracy": 0.0938566841185093,
      "num_tokens": 589218.0,
      "step": 260
    },
    {
      "entropy": 7.631022548675537,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000132,
      "loss": 7.1516,
      "mean_token_accuracy": 0.0841096393764019,
      "num_tokens": 600327.0,
      "step": 265
    },
    {
      "entropy": 7.694751644134522,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.0968,
      "mean_token_accuracy": 0.08569389209151268,
      "num_tokens": 610923.0,
      "step": 270
    },
    {
      "entropy": 7.742217016220093,
      "epoch": 0.29745808545159547,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1711,
      "mean_token_accuracy": 0.08022493869066238,
      "num_tokens": 621138.0,
      "step": 275
    },
    {
      "entropy": 7.701586723327637,
      "epoch": 0.3028664142779881,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001395,
      "loss": 7.0946,
      "mean_token_accuracy": 0.0921371579170227,
      "num_tokens": 631440.0,
      "step": 280
    },
    {
      "entropy": 7.630160140991211,
      "epoch": 0.30827474310438074,
      "grad_norm": 1.0625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1312,
      "mean_token_accuracy": 0.08813719153404236,
      "num_tokens": 642328.0,
      "step": 285
    },
    {
      "entropy": 7.665129518508911,
      "epoch": 0.3136830719307734,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0001445,
      "loss": 7.0909,
      "mean_token_accuracy": 0.0894947700202465,
      "num_tokens": 652875.0,
      "step": 290
    },
    {
      "entropy": 7.610320234298706,
      "epoch": 0.319091400757166,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000147,
      "loss": 7.031,
      "mean_token_accuracy": 0.08914472088217736,
      "num_tokens": 663726.0,
      "step": 295
    },
    {
      "entropy": 7.622774505615235,
      "epoch": 0.32449972958355866,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0001495,
      "loss": 7.2158,
      "mean_token_accuracy": 0.08153997585177422,
      "num_tokens": 675638.0,
      "step": 300
    },
    {
      "entropy": 7.5613326072692875,
      "epoch": 0.3299080584099513,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000152,
      "loss": 7.0719,
      "mean_token_accuracy": 0.09202613458037376,
      "num_tokens": 687239.0,
      "step": 305
    },
    {
      "entropy": 7.59527850151062,
      "epoch": 0.335316387236344,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0996,
      "mean_token_accuracy": 0.08664119839668274,
      "num_tokens": 698905.0,
      "step": 310
    },
    {
      "entropy": 7.607007694244385,
      "epoch": 0.34072471606273663,
      "grad_norm": 1.0625,
      "learning_rate": 0.000157,
      "loss": 7.09,
      "mean_token_accuracy": 0.0928929626941681,
      "num_tokens": 710602.0,
      "step": 315
    },
    {
      "entropy": 7.484040689468384,
      "epoch": 0.34613304488912927,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0001595,
      "loss": 7.0519,
      "mean_token_accuracy": 0.08713712617754936,
      "num_tokens": 721326.0,
      "step": 320
    },
    {
      "entropy": 7.593839359283447,
      "epoch": 0.3515413737155219,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000162,
      "loss": 7.0427,
      "mean_token_accuracy": 0.0913507841527462,
      "num_tokens": 732972.0,
      "step": 325
    },
    {
      "entropy": 7.492053413391114,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.0028,
      "mean_token_accuracy": 0.09586582258343697,
      "num_tokens": 743624.0,
      "step": 330
    },
    {
      "entropy": 7.524333477020264,
      "epoch": 0.3623580313683072,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.076,
      "mean_token_accuracy": 0.09233946427702903,
      "num_tokens": 756384.0,
      "step": 335
    },
    {
      "entropy": 7.58307991027832,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.0545,
      "mean_token_accuracy": 0.09635318964719772,
      "num_tokens": 767256.0,
      "step": 340
    },
    {
      "entropy": 7.474336338043213,
      "epoch": 0.37317468902109246,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9979,
      "mean_token_accuracy": 0.09708881303668022,
      "num_tokens": 779059.0,
      "step": 345
    },
    {
      "entropy": 7.505778360366821,
      "epoch": 0.3785830178474851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.0686,
      "mean_token_accuracy": 0.09691090956330299,
      "num_tokens": 790291.0,
      "step": 350
    },
    {
      "entropy": 7.510104370117188,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000177,
      "loss": 6.9605,
      "mean_token_accuracy": 0.09804000407457351,
      "num_tokens": 801450.0,
      "step": 355
    },
    {
      "entropy": 7.427967882156372,
      "epoch": 0.38939967550027044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001795,
      "loss": 6.8869,
      "mean_token_accuracy": 0.10146211236715316,
      "num_tokens": 812428.0,
      "step": 360
    },
    {
      "entropy": 7.357033348083496,
      "epoch": 0.3948080043266631,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000182,
      "loss": 6.9743,
      "mean_token_accuracy": 0.09577520415186883,
      "num_tokens": 823644.0,
      "step": 365
    },
    {
      "entropy": 7.539126396179199,
      "epoch": 0.4002163331530557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001845,
      "loss": 6.967,
      "mean_token_accuracy": 0.10204331800341607,
      "num_tokens": 834064.0,
      "step": 370
    },
    {
      "entropy": 7.464277219772339,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000187,
      "loss": 6.9062,
      "mean_token_accuracy": 0.1017673671245575,
      "num_tokens": 845207.0,
      "step": 375
    },
    {
      "entropy": 7.3919721126556395,
      "epoch": 0.411032990805841,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001895,
      "loss": 6.9524,
      "mean_token_accuracy": 0.09975179508328438,
      "num_tokens": 855564.0,
      "step": 380
    },
    {
      "entropy": 7.475247764587403,
      "epoch": 0.41644131963223363,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000192,
      "loss": 6.9807,
      "mean_token_accuracy": 0.10713295936584473,
      "num_tokens": 866959.0,
      "step": 385
    },
    {
      "entropy": 7.331955623626709,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001945,
      "loss": 6.9058,
      "mean_token_accuracy": 0.10511352196335792,
      "num_tokens": 877319.0,
      "step": 390
    },
    {
      "entropy": 7.372516441345215,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.9939,
      "mean_token_accuracy": 0.10005753636360168,
      "num_tokens": 888037.0,
      "step": 395
    },
    {
      "entropy": 7.404991912841797,
      "epoch": 0.43266630611141155,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7495,
      "mean_token_accuracy": 0.10937805995345115,
      "num_tokens": 898775.0,
      "step": 400
    },
    {
      "entropy": 7.364098739624024,
      "epoch": 0.43807463493780424,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000202,
      "loss": 6.8821,
      "mean_token_accuracy": 0.10221289917826652,
      "num_tokens": 909513.0,
      "step": 405
    },
    {
      "entropy": 7.383790636062622,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.923,
      "mean_token_accuracy": 0.10452766120433807,
      "num_tokens": 920730.0,
      "step": 410
    },
    {
      "entropy": 7.36325511932373,
      "epoch": 0.4488912925905895,
      "grad_norm": 1.1875,
      "learning_rate": 0.000207,
      "loss": 6.8456,
      "mean_token_accuracy": 0.11156300008296967,
      "num_tokens": 931457.0,
      "step": 415
    },
    {
      "entropy": 7.295261812210083,
      "epoch": 0.45429962141698216,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002095,
      "loss": 6.8372,
      "mean_token_accuracy": 0.10695499628782272,
      "num_tokens": 942572.0,
      "step": 420
    },
    {
      "entropy": 7.35825834274292,
      "epoch": 0.4597079502433748,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000212,
      "loss": 6.8416,
      "mean_token_accuracy": 0.09994131103157997,
      "num_tokens": 953132.0,
      "step": 425
    },
    {
      "entropy": 7.296279668807983,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002145,
      "loss": 6.8596,
      "mean_token_accuracy": 0.10478808656334877,
      "num_tokens": 963551.0,
      "step": 430
    },
    {
      "entropy": 7.320891809463501,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.078125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8273,
      "mean_token_accuracy": 0.10656212195754051,
      "num_tokens": 975297.0,
      "step": 435
    },
    {
      "entropy": 7.322855615615845,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002195,
      "loss": 6.8267,
      "mean_token_accuracy": 0.10611714869737625,
      "num_tokens": 986910.0,
      "step": 440
    },
    {
      "entropy": 7.355953693389893,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.484375,
      "learning_rate": 0.000222,
      "loss": 6.94,
      "mean_token_accuracy": 0.11208969727158546,
      "num_tokens": 997058.0,
      "step": 445
    },
    {
      "entropy": 7.25784215927124,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.125,
      "learning_rate": 0.0002245,
      "loss": 6.8025,
      "mean_token_accuracy": 0.10399467796087265,
      "num_tokens": 1008261.0,
      "step": 450
    },
    {
      "entropy": 7.249159240722657,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7439,
      "mean_token_accuracy": 0.11101565584540367,
      "num_tokens": 1019193.0,
      "step": 455
    },
    {
      "entropy": 7.18016037940979,
      "epoch": 0.4975662520281233,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.853,
      "mean_token_accuracy": 0.10952779948711396,
      "num_tokens": 1031227.0,
      "step": 460
    },
    {
      "entropy": 7.201208257675171,
      "epoch": 0.502974580854516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6978,
      "mean_token_accuracy": 0.10652377605438232,
      "num_tokens": 1042963.0,
      "step": 465
    },
    {
      "entropy": 7.195361566543579,
      "epoch": 0.5083829096809086,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7466,
      "mean_token_accuracy": 0.1075796753168106,
      "num_tokens": 1053709.0,
      "step": 470
    },
    {
      "entropy": 7.22407546043396,
      "epoch": 0.5137912385073012,
      "grad_norm": 1.3125,
      "learning_rate": 0.000237,
      "loss": 6.7614,
      "mean_token_accuracy": 0.11968052312731743,
      "num_tokens": 1064527.0,
      "step": 475
    },
    {
      "entropy": 7.201431703567505,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002395,
      "loss": 6.8083,
      "mean_token_accuracy": 0.11001280769705772,
      "num_tokens": 1075127.0,
      "step": 480
    },
    {
      "entropy": 7.159610843658447,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000242,
      "loss": 6.8036,
      "mean_token_accuracy": 0.11006302908062934,
      "num_tokens": 1086581.0,
      "step": 485
    },
    {
      "entropy": 7.195751047134399,
      "epoch": 0.5300162249864792,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002445,
      "loss": 6.7321,
      "mean_token_accuracy": 0.1197355069220066,
      "num_tokens": 1097818.0,
      "step": 490
    },
    {
      "entropy": 7.115999031066894,
      "epoch": 0.5354245538128718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000247,
      "loss": 6.7213,
      "mean_token_accuracy": 0.11185307502746582,
      "num_tokens": 1108043.0,
      "step": 495
    },
    {
      "entropy": 7.117218828201294,
      "epoch": 0.5408328826392644,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002495,
      "loss": 6.7845,
      "mean_token_accuracy": 0.11339325457811356,
      "num_tokens": 1119875.0,
      "step": 500
    },
    {
      "entropy": 7.161368274688721,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000252,
      "loss": 6.6836,
      "mean_token_accuracy": 0.11546274945139885,
      "num_tokens": 1132406.0,
      "step": 505
    },
    {
      "entropy": 7.2924041748046875,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002545,
      "loss": 6.8578,
      "mean_token_accuracy": 0.10715171545743943,
      "num_tokens": 1143828.0,
      "step": 510
    },
    {
      "entropy": 7.111097812652588,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.125,
      "learning_rate": 0.000257,
      "loss": 6.6629,
      "mean_token_accuracy": 0.1149873249232769,
      "num_tokens": 1155482.0,
      "step": 515
    },
    {
      "entropy": 7.082928657531738,
      "epoch": 0.5624661979448351,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002595,
      "loss": 6.6192,
      "mean_token_accuracy": 0.12244254648685456,
      "num_tokens": 1166847.0,
      "step": 520
    },
    {
      "entropy": 7.048942422866821,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.125,
      "learning_rate": 0.000262,
      "loss": 6.6899,
      "mean_token_accuracy": 0.11682257875800132,
      "num_tokens": 1177419.0,
      "step": 525
    },
    {
      "entropy": 7.181053352355957,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.8558,
      "mean_token_accuracy": 0.10907636061310769,
      "num_tokens": 1188975.0,
      "step": 530
    },
    {
      "entropy": 7.074787044525147,
      "epoch": 0.578691184424013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6899,
      "mean_token_accuracy": 0.11405783966183662,
      "num_tokens": 1200252.0,
      "step": 535
    },
    {
      "entropy": 7.110227489471436,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7279,
      "mean_token_accuracy": 0.12044090032577515,
      "num_tokens": 1211864.0,
      "step": 540
    },
    {
      "entropy": 7.007605695724488,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6089,
      "mean_token_accuracy": 0.11882655844092369,
      "num_tokens": 1222429.0,
      "step": 545
    },
    {
      "entropy": 7.159776020050049,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002745,
      "loss": 6.7062,
      "mean_token_accuracy": 0.11563858911395072,
      "num_tokens": 1233782.0,
      "step": 550
    },
    {
      "entropy": 6.999762058258057,
      "epoch": 0.6003244997295836,
      "grad_norm": 1.40625,
      "learning_rate": 0.000277,
      "loss": 6.6624,
      "mean_token_accuracy": 0.11662511229515075,
      "num_tokens": 1244309.0,
      "step": 555
    },
    {
      "entropy": 7.080610895156861,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002795,
      "loss": 6.6242,
      "mean_token_accuracy": 0.1214139811694622,
      "num_tokens": 1256403.0,
      "step": 560
    },
    {
      "entropy": 7.026603126525879,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7025,
      "mean_token_accuracy": 0.11629552468657493,
      "num_tokens": 1268758.0,
      "step": 565
    },
    {
      "entropy": 7.049094152450562,
      "epoch": 0.6165494862087615,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002845,
      "loss": 6.5898,
      "mean_token_accuracy": 0.12565962076187134,
      "num_tokens": 1279355.0,
      "step": 570
    },
    {
      "entropy": 6.986866807937622,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000287,
      "loss": 6.5681,
      "mean_token_accuracy": 0.118294008821249,
      "num_tokens": 1290619.0,
      "step": 575
    },
    {
      "entropy": 7.035441446304321,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002895,
      "loss": 6.5748,
      "mean_token_accuracy": 0.12222455516457557,
      "num_tokens": 1301851.0,
      "step": 580
    },
    {
      "entropy": 7.028501653671265,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.21875,
      "learning_rate": 0.000292,
      "loss": 6.6876,
      "mean_token_accuracy": 0.12039072066545486,
      "num_tokens": 1313260.0,
      "step": 585
    },
    {
      "entropy": 7.0265075206756595,
      "epoch": 0.638182801514332,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002945,
      "loss": 6.6719,
      "mean_token_accuracy": 0.11919900625944138,
      "num_tokens": 1324189.0,
      "step": 590
    },
    {
      "entropy": 7.09400782585144,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.265625,
      "learning_rate": 0.000297,
      "loss": 6.7163,
      "mean_token_accuracy": 0.11836740374565125,
      "num_tokens": 1335765.0,
      "step": 595
    },
    {
      "entropy": 6.97562804222107,
      "epoch": 0.6489994591671173,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002995,
      "loss": 6.6305,
      "mean_token_accuracy": 0.12355265021324158,
      "num_tokens": 1347653.0,
      "step": 600
    },
    {
      "entropy": 7.013906860351563,
      "epoch": 0.65440778799351,
      "grad_norm": 0.96875,
      "learning_rate": 0.000302,
      "loss": 6.627,
      "mean_token_accuracy": 0.12648534998297692,
      "num_tokens": 1360201.0,
      "step": 605
    },
    {
      "entropy": 6.952250146865845,
      "epoch": 0.6598161168199026,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003045,
      "loss": 6.5073,
      "mean_token_accuracy": 0.12420706525444984,
      "num_tokens": 1370974.0,
      "step": 610
    },
    {
      "entropy": 6.93790454864502,
      "epoch": 0.6652244456462953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000307,
      "loss": 6.5714,
      "mean_token_accuracy": 0.12407453581690789,
      "num_tokens": 1383855.0,
      "step": 615
    },
    {
      "entropy": 6.938852977752686,
      "epoch": 0.670632774472688,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0003095,
      "loss": 6.536,
      "mean_token_accuracy": 0.12525033503770827,
      "num_tokens": 1395041.0,
      "step": 620
    },
    {
      "entropy": 6.976763200759888,
      "epoch": 0.6760411032990806,
      "grad_norm": 1.3125,
      "learning_rate": 0.000312,
      "loss": 6.6361,
      "mean_token_accuracy": 0.12184548899531364,
      "num_tokens": 1406102.0,
      "step": 625
    },
    {
      "entropy": 6.916235494613647,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145,
      "loss": 6.4695,
      "mean_token_accuracy": 0.1252748616039753,
      "num_tokens": 1417713.0,
      "step": 630
    },
    {
      "entropy": 6.897937250137329,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.140625,
      "learning_rate": 0.000317,
      "loss": 6.6022,
      "mean_token_accuracy": 0.1269610993564129,
      "num_tokens": 1428526.0,
      "step": 635
    },
    {
      "entropy": 6.923327493667602,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003195,
      "loss": 6.5567,
      "mean_token_accuracy": 0.12470588013529778,
      "num_tokens": 1438849.0,
      "step": 640
    },
    {
      "entropy": 6.979862880706787,
      "epoch": 0.6976744186046512,
      "grad_norm": 1.03125,
      "learning_rate": 0.000322,
      "loss": 6.6268,
      "mean_token_accuracy": 0.12327171638607978,
      "num_tokens": 1450776.0,
      "step": 645
    },
    {
      "entropy": 6.835441255569458,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4933,
      "mean_token_accuracy": 0.12600184082984925,
      "num_tokens": 1462552.0,
      "step": 650
    },
    {
      "entropy": 6.796398162841797,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4924,
      "mean_token_accuracy": 0.12828927859663963,
      "num_tokens": 1473867.0,
      "step": 655
    },
    {
      "entropy": 6.799835538864135,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.497,
      "mean_token_accuracy": 0.12288376614451409,
      "num_tokens": 1485233.0,
      "step": 660
    },
    {
      "entropy": 6.901881504058838,
      "epoch": 0.7193077339102217,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5643,
      "mean_token_accuracy": 0.12888247296214103,
      "num_tokens": 1496919.0,
      "step": 665
    },
    {
      "entropy": 6.90382194519043,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5039,
      "mean_token_accuracy": 0.12288212552666664,
      "num_tokens": 1508695.0,
      "step": 670
    },
    {
      "entropy": 6.869405269622803,
      "epoch": 0.730124391563007,
      "grad_norm": 1.15625,
      "learning_rate": 0.000337,
      "loss": 6.5593,
      "mean_token_accuracy": 0.1286173418164253,
      "num_tokens": 1519629.0,
      "step": 675
    },
    {
      "entropy": 6.8630880355834964,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003395,
      "loss": 6.4569,
      "mean_token_accuracy": 0.12953948080539704,
      "num_tokens": 1530144.0,
      "step": 680
    },
    {
      "entropy": 6.725225353240967,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.421875,
      "learning_rate": 0.000342,
      "loss": 6.4683,
      "mean_token_accuracy": 0.12884586453437805,
      "num_tokens": 1541013.0,
      "step": 685
    },
    {
      "entropy": 6.818743658065796,
      "epoch": 0.7463493780421849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.54,
      "mean_token_accuracy": 0.1279229611158371,
      "num_tokens": 1553344.0,
      "step": 690
    },
    {
      "entropy": 6.90196681022644,
      "epoch": 0.7517577068685776,
      "grad_norm": 1.5625,
      "learning_rate": 0.000347,
      "loss": 6.5904,
      "mean_token_accuracy": 0.123768300563097,
      "num_tokens": 1565298.0,
      "step": 695
    },
    {
      "entropy": 6.776127195358276,
      "epoch": 0.7571660356949702,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003495,
      "loss": 6.5125,
      "mean_token_accuracy": 0.1363504111766815,
      "num_tokens": 1577645.0,
      "step": 700
    },
    {
      "entropy": 6.815530920028687,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.15625,
      "learning_rate": 0.000352,
      "loss": 6.4674,
      "mean_token_accuracy": 0.13412229791283609,
      "num_tokens": 1588387.0,
      "step": 705
    },
    {
      "entropy": 6.863557958602906,
      "epoch": 0.7679826933477556,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003545,
      "loss": 6.5756,
      "mean_token_accuracy": 0.13169802352786064,
      "num_tokens": 1599837.0,
      "step": 710
    },
    {
      "entropy": 6.80879316329956,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.25,
      "learning_rate": 0.000357,
      "loss": 6.5179,
      "mean_token_accuracy": 0.12488113269209862,
      "num_tokens": 1610401.0,
      "step": 715
    },
    {
      "entropy": 6.792469310760498,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003595,
      "loss": 6.4423,
      "mean_token_accuracy": 0.12702938616275788,
      "num_tokens": 1622688.0,
      "step": 720
    },
    {
      "entropy": 6.785536813735962,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000362,
      "loss": 6.5322,
      "mean_token_accuracy": 0.11961603239178657,
      "num_tokens": 1633594.0,
      "step": 725
    },
    {
      "entropy": 6.796718645095825,
      "epoch": 0.7896160086533262,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003645,
      "loss": 6.431,
      "mean_token_accuracy": 0.1275559477508068,
      "num_tokens": 1645861.0,
      "step": 730
    },
    {
      "entropy": 6.67658257484436,
      "epoch": 0.7950243374797188,
      "grad_norm": 1.375,
      "learning_rate": 0.000367,
      "loss": 6.3343,
      "mean_token_accuracy": 0.13718428909778596,
      "num_tokens": 1656903.0,
      "step": 735
    },
    {
      "entropy": 6.801560592651367,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003695,
      "loss": 6.441,
      "mean_token_accuracy": 0.13460942655801772,
      "num_tokens": 1668130.0,
      "step": 740
    },
    {
      "entropy": 6.625500011444092,
      "epoch": 0.8058409951325041,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000372,
      "loss": 6.3347,
      "mean_token_accuracy": 0.14152220115065575,
      "num_tokens": 1678228.0,
      "step": 745
    },
    {
      "entropy": 6.857194662094116,
      "epoch": 0.8112493239588967,
      "grad_norm": 1.25,
      "learning_rate": 0.0003745,
      "loss": 6.487,
      "mean_token_accuracy": 0.1327496737241745,
      "num_tokens": 1690339.0,
      "step": 750
    },
    {
      "entropy": 6.740129566192627,
      "epoch": 0.8166576527852893,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000377,
      "loss": 6.4986,
      "mean_token_accuracy": 0.12592008486390113,
      "num_tokens": 1701067.0,
      "step": 755
    },
    {
      "entropy": 6.824430894851685,
      "epoch": 0.822065981611682,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003795,
      "loss": 6.4343,
      "mean_token_accuracy": 0.13577242717146873,
      "num_tokens": 1712835.0,
      "step": 760
    },
    {
      "entropy": 6.693302774429322,
      "epoch": 0.8274743104380746,
      "grad_norm": 1.09375,
      "learning_rate": 0.000382,
      "loss": 6.3978,
      "mean_token_accuracy": 0.13282729983329772,
      "num_tokens": 1723432.0,
      "step": 765
    },
    {
      "entropy": 6.772776031494141,
      "epoch": 0.8328826392644673,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003845,
      "loss": 6.4324,
      "mean_token_accuracy": 0.13451724424958228,
      "num_tokens": 1734134.0,
      "step": 770
    },
    {
      "entropy": 6.688350820541382,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3783,
      "mean_token_accuracy": 0.13390962406992912,
      "num_tokens": 1744723.0,
      "step": 775
    },
    {
      "entropy": 6.730376195907593,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.511,
      "mean_token_accuracy": 0.131271693110466,
      "num_tokens": 1755269.0,
      "step": 780
    },
    {
      "entropy": 6.77961688041687,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5187,
      "mean_token_accuracy": 0.13170363679528235,
      "num_tokens": 1766344.0,
      "step": 785
    },
    {
      "entropy": 6.774808263778686,
      "epoch": 0.8545159545700378,
      "grad_norm": 1.46875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4812,
      "mean_token_accuracy": 0.12957761734724044,
      "num_tokens": 1779052.0,
      "step": 790
    },
    {
      "entropy": 6.616818809509278,
      "epoch": 0.8599242833964305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.32,
      "mean_token_accuracy": 0.13625366762280464,
      "num_tokens": 1789754.0,
      "step": 795
    },
    {
      "entropy": 6.559850549697876,
      "epoch": 0.8653326122228231,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003995,
      "loss": 6.2474,
      "mean_token_accuracy": 0.1390355110168457,
      "num_tokens": 1801161.0,
      "step": 800
    },
    {
      "entropy": 6.705098056793213,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.328125,
      "learning_rate": 0.000402,
      "loss": 6.3624,
      "mean_token_accuracy": 0.13253424242138861,
      "num_tokens": 1810892.0,
      "step": 805
    },
    {
      "entropy": 6.548418378829956,
      "epoch": 0.8761492698756085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004045,
      "loss": 6.403,
      "mean_token_accuracy": 0.13478510975837707,
      "num_tokens": 1821574.0,
      "step": 810
    },
    {
      "entropy": 6.720035552978516,
      "epoch": 0.8815575987020011,
      "grad_norm": 0.984375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4552,
      "mean_token_accuracy": 0.13616764321923255,
      "num_tokens": 1833342.0,
      "step": 815
    },
    {
      "entropy": 6.621445178985596,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.25,
      "learning_rate": 0.0004095,
      "loss": 6.3237,
      "mean_token_accuracy": 0.1389230526983738,
      "num_tokens": 1843768.0,
      "step": 820
    },
    {
      "entropy": 6.765406370162964,
      "epoch": 0.8923742563547864,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000412,
      "loss": 6.4073,
      "mean_token_accuracy": 0.1308816060423851,
      "num_tokens": 1855569.0,
      "step": 825
    },
    {
      "entropy": 6.6211669921875,
      "epoch": 0.897782585181179,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004145,
      "loss": 6.3909,
      "mean_token_accuracy": 0.1304692007601261,
      "num_tokens": 1866684.0,
      "step": 830
    },
    {
      "entropy": 6.672160291671753,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.265625,
      "learning_rate": 0.000417,
      "loss": 6.3193,
      "mean_token_accuracy": 0.1392674833536148,
      "num_tokens": 1878700.0,
      "step": 835
    },
    {
      "entropy": 6.6165376663208,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004195,
      "loss": 6.3703,
      "mean_token_accuracy": 0.13358763232827187,
      "num_tokens": 1889970.0,
      "step": 840
    },
    {
      "entropy": 6.5742271900177,
      "epoch": 0.914007571660357,
      "grad_norm": 1.65625,
      "learning_rate": 0.000422,
      "loss": 6.2577,
      "mean_token_accuracy": 0.13828584030270577,
      "num_tokens": 1902008.0,
      "step": 845
    },
    {
      "entropy": 6.615272140502929,
      "epoch": 0.9194159004867496,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 6.4757,
      "mean_token_accuracy": 0.13929339945316316,
      "num_tokens": 1913253.0,
      "step": 850
    },
    {
      "entropy": 6.614001512527466,
      "epoch": 0.9248242293131422,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000427,
      "loss": 6.3162,
      "mean_token_accuracy": 0.143621264398098,
      "num_tokens": 1923408.0,
      "step": 855
    },
    {
      "entropy": 6.615392589569092,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004295,
      "loss": 6.3774,
      "mean_token_accuracy": 0.13835460394620896,
      "num_tokens": 1935315.0,
      "step": 860
    },
    {
      "entropy": 6.567425060272217,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.015625,
      "learning_rate": 0.000432,
      "loss": 6.2219,
      "mean_token_accuracy": 0.14048045724630356,
      "num_tokens": 1946524.0,
      "step": 865
    },
    {
      "entropy": 6.63021936416626,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004345,
      "loss": 6.3635,
      "mean_token_accuracy": 0.13752378821372985,
      "num_tokens": 1957561.0,
      "step": 870
    },
    {
      "entropy": 6.573499870300293,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 6.2889,
      "mean_token_accuracy": 0.14111032485961914,
      "num_tokens": 1969089.0,
      "step": 875
    },
    {
      "entropy": 6.578218793869018,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004395,
      "loss": 6.3304,
      "mean_token_accuracy": 0.13999469056725503,
      "num_tokens": 1980341.0,
      "step": 880
    },
    {
      "entropy": 6.540131092071533,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.28125,
      "learning_rate": 0.000442,
      "loss": 6.2821,
      "mean_token_accuracy": 0.14466761127114297,
      "num_tokens": 1991220.0,
      "step": 885
    },
    {
      "entropy": 6.561265563964843,
      "epoch": 0.9626825310978907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445,
      "loss": 6.3599,
      "mean_token_accuracy": 0.1384817786514759,
      "num_tokens": 2002650.0,
      "step": 890
    },
    {
      "entropy": 6.561820602416992,
      "epoch": 0.9680908599242833,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000447,
      "loss": 6.274,
      "mean_token_accuracy": 0.14329540133476257,
      "num_tokens": 2012884.0,
      "step": 895
    },
    {
      "entropy": 6.60991587638855,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.354,
      "mean_token_accuracy": 0.1367944747209549,
      "num_tokens": 2023543.0,
      "step": 900
    },
    {
      "entropy": 6.675815534591675,
      "epoch": 0.9789075175770687,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4228,
      "mean_token_accuracy": 0.1348619595170021,
      "num_tokens": 2034789.0,
      "step": 905
    },
    {
      "entropy": 6.4962972640991214,
      "epoch": 0.9843158464034614,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3091,
      "mean_token_accuracy": 0.13746569231152533,
      "num_tokens": 2046719.0,
      "step": 910
    },
    {
      "entropy": 6.470368957519531,
      "epoch": 0.989724175229854,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2586,
      "mean_token_accuracy": 0.13995983302593232,
      "num_tokens": 2057016.0,
      "step": 915
    },
    {
      "entropy": 6.526668930053711,
      "epoch": 0.9951325040562466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2582,
      "mean_token_accuracy": 0.1368989862501621,
      "num_tokens": 2067898.0,
      "step": 920
    },
    {
      "entropy": 6.4760607613457575,
      "epoch": 1.0,
      "grad_norm": 2.078125,
      "learning_rate": 0.000462,
      "loss": 6.2648,
      "mean_token_accuracy": 0.14498859478367698,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.640408563613891,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004645,
      "loss": 6.2126,
      "mean_token_accuracy": 0.13421827852725982,
      "num_tokens": 2089228.0,
      "step": 930
    },
    {
      "entropy": 6.453851366043091,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.3125,
      "learning_rate": 0.000467,
      "loss": 6.2329,
      "mean_token_accuracy": 0.1465141922235489,
      "num_tokens": 2101098.0,
      "step": 935
    },
    {
      "entropy": 6.46987681388855,
      "epoch": 1.016224986479178,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004695,
      "loss": 6.1718,
      "mean_token_accuracy": 0.14940173774957657,
      "num_tokens": 2113081.0,
      "step": 940
    },
    {
      "entropy": 6.516440105438233,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472,
      "loss": 6.1389,
      "mean_token_accuracy": 0.14872755408287047,
      "num_tokens": 2123356.0,
      "step": 945
    },
    {
      "entropy": 6.582454013824463,
      "epoch": 1.0270416441319632,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004745,
      "loss": 6.2387,
      "mean_token_accuracy": 0.14650474786758422,
      "num_tokens": 2134388.0,
      "step": 950
    },
    {
      "entropy": 6.41712703704834,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000477,
      "loss": 6.1166,
      "mean_token_accuracy": 0.14716140180826187,
      "num_tokens": 2145209.0,
      "step": 955
    },
    {
      "entropy": 6.449684047698975,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004795,
      "loss": 6.1591,
      "mean_token_accuracy": 0.14986583292484285,
      "num_tokens": 2157146.0,
      "step": 960
    },
    {
      "entropy": 6.381077623367309,
      "epoch": 1.043266630611141,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 6.1832,
      "mean_token_accuracy": 0.14549879282712935,
      "num_tokens": 2168721.0,
      "step": 965
    },
    {
      "entropy": 6.512086629867554,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004845,
      "loss": 6.2002,
      "mean_token_accuracy": 0.14564864486455917,
      "num_tokens": 2179213.0,
      "step": 970
    },
    {
      "entropy": 6.524026775360108,
      "epoch": 1.0540832882639264,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487,
      "loss": 6.1874,
      "mean_token_accuracy": 0.1458301618695259,
      "num_tokens": 2191224.0,
      "step": 975
    },
    {
      "entropy": 6.455999326705933,
      "epoch": 1.059491617090319,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895,
      "loss": 6.2661,
      "mean_token_accuracy": 0.14348524808883667,
      "num_tokens": 2203665.0,
      "step": 980
    },
    {
      "entropy": 6.5328833103179935,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000492,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1457217141985893,
      "num_tokens": 2215136.0,
      "step": 985
    },
    {
      "entropy": 6.491800928115845,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004945,
      "loss": 6.2022,
      "mean_token_accuracy": 0.14653969779610634,
      "num_tokens": 2226375.0,
      "step": 990
    },
    {
      "entropy": 6.3262214183807375,
      "epoch": 1.075716603569497,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000497,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15307881981134414,
      "num_tokens": 2236097.0,
      "step": 995
    },
    {
      "entropy": 6.592307758331299,
      "epoch": 1.0811249323958896,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995,
      "loss": 6.3171,
      "mean_token_accuracy": 0.13785527274012566,
      "num_tokens": 2247964.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 6.237357024712996,
      "eval_loss": 6.420225620269775,
      "eval_mean_token_accuracy": 0.1424645613455637,
      "eval_num_tokens": 2247964.0,
      "eval_runtime": 2.0378,
      "eval_samples_per_second": 1724.891,
      "eval_steps_per_second": 215.918,
      "step": 1000
    },
    {
      "entropy": 6.435177373886108,
      "epoch": 1.0865332612222822,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999997383518051,
      "loss": 6.2381,
      "mean_token_accuracy": 0.14582235515117645,
      "num_tokens": 2260467.0,
      "step": 1005
    },
    {
      "entropy": 6.423948192596436,
      "epoch": 1.0919415900486749,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999986754070562,
      "loss": 6.1489,
      "mean_token_accuracy": 0.14645124971866608,
      "num_tokens": 2272176.0,
      "step": 1010
    },
    {
      "entropy": 6.449360799789429,
      "epoch": 1.0973499188750675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999967948166009,
      "loss": 6.1928,
      "mean_token_accuracy": 0.15050890892744065,
      "num_tokens": 2283491.0,
      "step": 1015
    },
    {
      "entropy": 6.479728126525879,
      "epoch": 1.1027582477014604,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999940965872734,
      "loss": 6.0795,
      "mean_token_accuracy": 0.14409494027495384,
      "num_tokens": 2295131.0,
      "step": 1020
    },
    {
      "entropy": 6.365811204910278,
      "epoch": 1.1081665765278528,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999905807288788,
      "loss": 6.0943,
      "mean_token_accuracy": 0.151952263712883,
      "num_tokens": 2305832.0,
      "step": 1025
    },
    {
      "entropy": 6.331813812255859,
      "epoch": 1.1135749053542456,
      "grad_norm": 1.25,
      "learning_rate": 0.000499986247254194,
      "loss": 6.0956,
      "mean_token_accuracy": 0.15410929024219513,
      "num_tokens": 2315874.0,
      "step": 1030
    },
    {
      "entropy": 6.534620904922486,
      "epoch": 1.1189832341806383,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999810961789667,
      "loss": 6.3005,
      "mean_token_accuracy": 0.14081365540623664,
      "num_tokens": 2327580.0,
      "step": 1035
    },
    {
      "entropy": 6.438720798492431,
      "epoch": 1.124391563007031,
      "grad_norm": 1.0,
      "learning_rate": 0.000499975127521916,
      "loss": 6.1331,
      "mean_token_accuracy": 0.1487702637910843,
      "num_tokens": 2340426.0,
      "step": 1040
    },
    {
      "entropy": 6.418537998199463,
      "epoch": 1.1297998918334236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999683413047319,
      "loss": 6.2465,
      "mean_token_accuracy": 0.14431385397911073,
      "num_tokens": 2353228.0,
      "step": 1045
    },
    {
      "entropy": 6.436538887023926,
      "epoch": 1.1352082206598162,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999607375520759,
      "loss": 6.0737,
      "mean_token_accuracy": 0.1525431215763092,
      "num_tokens": 2364238.0,
      "step": 1050
    },
    {
      "entropy": 6.368298864364624,
      "epoch": 1.1406165494862088,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999523162915794,
      "loss": 6.2128,
      "mean_token_accuracy": 0.14406751096248627,
      "num_tokens": 2375847.0,
      "step": 1055
    },
    {
      "entropy": 6.355217170715332,
      "epoch": 1.1460248783126015,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999430775538459,
      "loss": 6.0978,
      "mean_token_accuracy": 0.1541084788739681,
      "num_tokens": 2386489.0,
      "step": 1060
    },
    {
      "entropy": 6.399798774719239,
      "epoch": 1.151433207138994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999330213724487,
      "loss": 6.1412,
      "mean_token_accuracy": 0.14941838160157203,
      "num_tokens": 2398839.0,
      "step": 1065
    },
    {
      "entropy": 6.436131286621094,
      "epoch": 1.1568415359653867,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999221477839319,
      "loss": 6.1034,
      "mean_token_accuracy": 0.1503540799021721,
      "num_tokens": 2410016.0,
      "step": 1070
    },
    {
      "entropy": 6.342180395126343,
      "epoch": 1.1622498647917794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999104568278104,
      "loss": 6.1401,
      "mean_token_accuracy": 0.14826380759477614,
      "num_tokens": 2421202.0,
      "step": 1075
    },
    {
      "entropy": 6.498359870910645,
      "epoch": 1.167658193618172,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499897948546569,
      "loss": 6.2367,
      "mean_token_accuracy": 0.13952652588486672,
      "num_tokens": 2431943.0,
      "step": 1080
    },
    {
      "entropy": 6.358457422256469,
      "epoch": 1.1730665224445647,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998846229856629,
      "loss": 6.0432,
      "mean_token_accuracy": 0.14694100618362427,
      "num_tokens": 2443454.0,
      "step": 1085
    },
    {
      "entropy": 6.39977617263794,
      "epoch": 1.1784748512709573,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998704801935171,
      "loss": 6.2352,
      "mean_token_accuracy": 0.1409311354160309,
      "num_tokens": 2455909.0,
      "step": 1090
    },
    {
      "entropy": 6.27590274810791,
      "epoch": 1.18388318009735,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998555202215267,
      "loss": 6.0927,
      "mean_token_accuracy": 0.1536378711462021,
      "num_tokens": 2467486.0,
      "step": 1095
    },
    {
      "entropy": 6.408985948562622,
      "epoch": 1.1892915089237426,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998397431240561,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15721485614776612,
      "num_tokens": 2477560.0,
      "step": 1100
    },
    {
      "entropy": 6.319623804092407,
      "epoch": 1.1946998377501352,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998231489584395,
      "loss": 6.0898,
      "mean_token_accuracy": 0.15928600281476973,
      "num_tokens": 2487222.0,
      "step": 1105
    },
    {
      "entropy": 6.336186408996582,
      "epoch": 1.2001081665765279,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00049980573778498,
      "loss": 6.1226,
      "mean_token_accuracy": 0.1514808341860771,
      "num_tokens": 2497474.0,
      "step": 1110
    },
    {
      "entropy": 6.386902284622193,
      "epoch": 1.2055164954029205,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997875096669498,
      "loss": 6.1757,
      "mean_token_accuracy": 0.1496924042701721,
      "num_tokens": 2508185.0,
      "step": 1115
    },
    {
      "entropy": 6.38016037940979,
      "epoch": 1.2109248242293131,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997684646705901,
      "loss": 6.1276,
      "mean_token_accuracy": 0.14406513720750808,
      "num_tokens": 2518876.0,
      "step": 1120
    },
    {
      "entropy": 6.339206218719482,
      "epoch": 1.2163331530557058,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997486028651103,
      "loss": 6.1181,
      "mean_token_accuracy": 0.1543666183948517,
      "num_tokens": 2530554.0,
      "step": 1125
    },
    {
      "entropy": 6.226878881454468,
      "epoch": 1.2217414818820984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997279243226884,
      "loss": 5.9914,
      "mean_token_accuracy": 0.14919330775737763,
      "num_tokens": 2541331.0,
      "step": 1130
    },
    {
      "entropy": 6.428718852996826,
      "epoch": 1.227149810708491,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997064291184701,
      "loss": 6.2128,
      "mean_token_accuracy": 0.14138779938220977,
      "num_tokens": 2554225.0,
      "step": 1135
    },
    {
      "entropy": 6.325518941879272,
      "epoch": 1.2325581395348837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996841173305691,
      "loss": 6.089,
      "mean_token_accuracy": 0.15215825885534287,
      "num_tokens": 2565050.0,
      "step": 1140
    },
    {
      "entropy": 6.39612340927124,
      "epoch": 1.2379664683612763,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996609890400665,
      "loss": 6.1638,
      "mean_token_accuracy": 0.14989278465509415,
      "num_tokens": 2576926.0,
      "step": 1145
    },
    {
      "entropy": 6.47265625,
      "epoch": 1.243374797187669,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996370443310107,
      "loss": 6.25,
      "mean_token_accuracy": 0.14251136481761933,
      "num_tokens": 2588775.0,
      "step": 1150
    },
    {
      "entropy": 6.229281759262085,
      "epoch": 1.2487831260140616,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996122832904165,
      "loss": 6.0869,
      "mean_token_accuracy": 0.1530088022351265,
      "num_tokens": 2600534.0,
      "step": 1155
    },
    {
      "entropy": 6.241557168960571,
      "epoch": 1.2541914548404542,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995867060082658,
      "loss": 5.9202,
      "mean_token_accuracy": 0.15806380808353424,
      "num_tokens": 2612350.0,
      "step": 1160
    },
    {
      "entropy": 6.271128797531128,
      "epoch": 1.2595997836668469,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995603125775065,
      "loss": 6.1085,
      "mean_token_accuracy": 0.1509605959057808,
      "num_tokens": 2623871.0,
      "step": 1165
    },
    {
      "entropy": 6.281409692764282,
      "epoch": 1.2650081124932395,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004995331030940523,
      "loss": 6.0328,
      "mean_token_accuracy": 0.1470465123653412,
      "num_tokens": 2636371.0,
      "step": 1170
    },
    {
      "entropy": 6.366229057312012,
      "epoch": 1.2704164413196322,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995050776567826,
      "loss": 6.0853,
      "mean_token_accuracy": 0.15386933013796805,
      "num_tokens": 2646794.0,
      "step": 1175
    },
    {
      "entropy": 6.328421592712402,
      "epoch": 1.2758247701460248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994762363675421,
      "loss": 6.101,
      "mean_token_accuracy": 0.1538513943552971,
      "num_tokens": 2657105.0,
      "step": 1180
    },
    {
      "entropy": 6.261244678497315,
      "epoch": 1.2812330989724177,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994465793311396,
      "loss": 6.0358,
      "mean_token_accuracy": 0.1510448545217514,
      "num_tokens": 2668297.0,
      "step": 1185
    },
    {
      "entropy": 6.218826055526733,
      "epoch": 1.28664142779881,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994161066553493,
      "loss": 5.8786,
      "mean_token_accuracy": 0.1550386816263199,
      "num_tokens": 2678834.0,
      "step": 1190
    },
    {
      "entropy": 6.323922491073608,
      "epoch": 1.292049756625203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993848184509086,
      "loss": 6.1272,
      "mean_token_accuracy": 0.14862247109413146,
      "num_tokens": 2690392.0,
      "step": 1195
    },
    {
      "entropy": 6.249981784820557,
      "epoch": 1.2974580854515954,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993527148315193,
      "loss": 5.9429,
      "mean_token_accuracy": 0.1543548472225666,
      "num_tokens": 2701319.0,
      "step": 1200
    },
    {
      "entropy": 6.296136903762817,
      "epoch": 1.3028664142779882,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993197959138458,
      "loss": 6.1028,
      "mean_token_accuracy": 0.14995811581611634,
      "num_tokens": 2713018.0,
      "step": 1205
    },
    {
      "entropy": 6.341501522064209,
      "epoch": 1.3082747431043806,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004992860618175156,
      "loss": 6.1861,
      "mean_token_accuracy": 0.1437143176794052,
      "num_tokens": 2724995.0,
      "step": 1210
    },
    {
      "entropy": 6.296142959594727,
      "epoch": 1.3136830719307735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992515126651184,
      "loss": 6.0388,
      "mean_token_accuracy": 0.15356030091643333,
      "num_tokens": 2736733.0,
      "step": 1215
    },
    {
      "entropy": 6.1639915943145756,
      "epoch": 1.319091400757166,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992161485822059,
      "loss": 6.0109,
      "mean_token_accuracy": 0.15306128412485123,
      "num_tokens": 2748436.0,
      "step": 1220
    },
    {
      "entropy": 6.265667676925659,
      "epoch": 1.3244997295835588,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991799696972913,
      "loss": 6.0276,
      "mean_token_accuracy": 0.1567312479019165,
      "num_tokens": 2758924.0,
      "step": 1225
    },
    {
      "entropy": 6.281225633621216,
      "epoch": 1.3299080584099512,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991429761418488,
      "loss": 6.0892,
      "mean_token_accuracy": 0.14596257656812667,
      "num_tokens": 2769773.0,
      "step": 1230
    },
    {
      "entropy": 6.410810232162476,
      "epoch": 1.335316387236344,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499105168050313,
      "loss": 6.2244,
      "mean_token_accuracy": 0.1527717739343643,
      "num_tokens": 2782414.0,
      "step": 1235
    },
    {
      "entropy": 6.173236465454101,
      "epoch": 1.3407247160627367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990665455600787,
      "loss": 5.9954,
      "mean_token_accuracy": 0.15391278266906738,
      "num_tokens": 2794325.0,
      "step": 1240
    },
    {
      "entropy": 6.122881603240967,
      "epoch": 1.3461330448891293,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990271088114999,
      "loss": 5.8397,
      "mean_token_accuracy": 0.1607421964406967,
      "num_tokens": 2805391.0,
      "step": 1245
    },
    {
      "entropy": 6.106888484954834,
      "epoch": 1.351541373715522,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989868579478901,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15657583773136138,
      "num_tokens": 2816633.0,
      "step": 1250
    },
    {
      "entropy": 6.261765575408935,
      "epoch": 1.3569497025419146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989457931155209,
      "loss": 6.0819,
      "mean_token_accuracy": 0.15534812957048416,
      "num_tokens": 2828395.0,
      "step": 1255
    },
    {
      "entropy": 6.284375190734863,
      "epoch": 1.3623580313683072,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989039144636219,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15927724540233612,
      "num_tokens": 2839592.0,
      "step": 1260
    },
    {
      "entropy": 6.2426642894744875,
      "epoch": 1.3677663601946999,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988612221443804,
      "loss": 6.0407,
      "mean_token_accuracy": 0.16757878810167312,
      "num_tokens": 2849557.0,
      "step": 1265
    },
    {
      "entropy": 6.320202732086182,
      "epoch": 1.3731746890210925,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004988177163129403,
      "loss": 6.0789,
      "mean_token_accuracy": 0.15343549251556396,
      "num_tokens": 2862030.0,
      "step": 1270
    },
    {
      "entropy": 6.190796136856079,
      "epoch": 1.3785830178474852,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987733971274018,
      "loss": 6.0134,
      "mean_token_accuracy": 0.1602019637823105,
      "num_tokens": 2872379.0,
      "step": 1275
    },
    {
      "entropy": 6.323750352859497,
      "epoch": 1.3839913466738778,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987282647488211,
      "loss": 6.0854,
      "mean_token_accuracy": 0.15255604684352875,
      "num_tokens": 2883170.0,
      "step": 1280
    },
    {
      "entropy": 6.206164073944092,
      "epoch": 1.3893996755002704,
      "grad_norm": 1.0,
      "learning_rate": 0.0004986823193412093,
      "loss": 6.0281,
      "mean_token_accuracy": 0.1504069373011589,
      "num_tokens": 2894013.0,
      "step": 1285
    },
    {
      "entropy": 6.265843105316162,
      "epoch": 1.394808004326663,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498635561071532,
      "loss": 6.0965,
      "mean_token_accuracy": 0.15722659826278687,
      "num_tokens": 2905073.0,
      "step": 1290
    },
    {
      "entropy": 6.322227430343628,
      "epoch": 1.4002163331530557,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498587990109709,
      "loss": 6.0611,
      "mean_token_accuracy": 0.15372281968593599,
      "num_tokens": 2916579.0,
      "step": 1295
    },
    {
      "entropy": 6.202920436859131,
      "epoch": 1.4056246619794484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985396066286133,
      "loss": 5.9529,
      "mean_token_accuracy": 0.15599865168333055,
      "num_tokens": 2928522.0,
      "step": 1300
    },
    {
      "entropy": 6.246823740005493,
      "epoch": 1.411032990805841,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984904108040706,
      "loss": 6.0418,
      "mean_token_accuracy": 0.16009896695613862,
      "num_tokens": 2938754.0,
      "step": 1305
    },
    {
      "entropy": 6.096659517288208,
      "epoch": 1.4164413196322336,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984404028148585,
      "loss": 5.8621,
      "mean_token_accuracy": 0.16269333213567733,
      "num_tokens": 2949696.0,
      "step": 1310
    },
    {
      "entropy": 6.224114513397216,
      "epoch": 1.4218496484586263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983895828427062,
      "loss": 6.0548,
      "mean_token_accuracy": 0.14663359969854356,
      "num_tokens": 2961144.0,
      "step": 1315
    },
    {
      "entropy": 6.282317972183227,
      "epoch": 1.427257977285019,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983379510722935,
      "loss": 6.0389,
      "mean_token_accuracy": 0.15705436319112778,
      "num_tokens": 2972368.0,
      "step": 1320
    },
    {
      "entropy": 6.181911420822144,
      "epoch": 1.4326663061114115,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982855076912506,
      "loss": 5.9638,
      "mean_token_accuracy": 0.1542641267180443,
      "num_tokens": 2984853.0,
      "step": 1325
    },
    {
      "entropy": 6.2893272876739506,
      "epoch": 1.4380746349378042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982322528901568,
      "loss": 6.0196,
      "mean_token_accuracy": 0.15734567493200302,
      "num_tokens": 2995095.0,
      "step": 1330
    },
    {
      "entropy": 6.136825799942017,
      "epoch": 1.4434829637641968,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049817818686254,
      "loss": 6.0786,
      "mean_token_accuracy": 0.15213681608438492,
      "num_tokens": 3005926.0,
      "step": 1335
    },
    {
      "entropy": 6.327638864517212,
      "epoch": 1.4488912925905895,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981233098048765,
      "loss": 6.029,
      "mean_token_accuracy": 0.1540302723646164,
      "num_tokens": 3017647.0,
      "step": 1340
    },
    {
      "entropy": 6.090318870544434,
      "epoch": 1.454299621416982,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004980676219165894,
      "loss": 5.9315,
      "mean_token_accuracy": 0.16002180129289628,
      "num_tokens": 3028904.0,
      "step": 1345
    },
    {
      "entropy": 6.2678004741668705,
      "epoch": 1.4597079502433747,
      "grad_norm": 1.21875,
      "learning_rate": 0.000498011123400049,
      "loss": 5.9859,
      "mean_token_accuracy": 0.15976762771606445,
      "num_tokens": 3039347.0,
      "step": 1350
    },
    {
      "entropy": 6.134640789031982,
      "epoch": 1.4651162790697674,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979538144605707,
      "loss": 5.9547,
      "mean_token_accuracy": 0.15985055714845658,
      "num_tokens": 3049951.0,
      "step": 1355
    },
    {
      "entropy": 6.237525272369385,
      "epoch": 1.47052460789616,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004978956953064154,
      "loss": 6.0,
      "mean_token_accuracy": 0.1584839180111885,
      "num_tokens": 3061894.0,
      "step": 1360
    },
    {
      "entropy": 6.22474308013916,
      "epoch": 1.4759329367225527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004978367661487883,
      "loss": 5.9786,
      "mean_token_accuracy": 0.16255034655332565,
      "num_tokens": 3072393.0,
      "step": 1365
    },
    {
      "entropy": 6.14780969619751,
      "epoch": 1.4813412655489453,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004977770272018379,
      "loss": 6.016,
      "mean_token_accuracy": 0.15890694856643678,
      "num_tokens": 3083541.0,
      "step": 1370
    },
    {
      "entropy": 6.193307638168335,
      "epoch": 1.486749594375338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497716478682656,
      "loss": 5.8382,
      "mean_token_accuracy": 0.15965883284807206,
      "num_tokens": 3093851.0,
      "step": 1375
    },
    {
      "entropy": 6.02174940109253,
      "epoch": 1.4921579232017308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004976551208112759,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1565147563815117,
      "num_tokens": 3105703.0,
      "step": 1380
    },
    {
      "entropy": 6.326010656356812,
      "epoch": 1.4975662520281232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004975929538106721,
      "loss": 6.0652,
      "mean_token_accuracy": 0.15564267784357072,
      "num_tokens": 3117063.0,
      "step": 1385
    },
    {
      "entropy": 6.1744105339050295,
      "epoch": 1.502974580854516,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975299779067598,
      "loss": 6.0253,
      "mean_token_accuracy": 0.15938320606946946,
      "num_tokens": 3129020.0,
      "step": 1390
    },
    {
      "entropy": 6.207565689086914,
      "epoch": 1.5083829096809085,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974661933283936,
      "loss": 6.0008,
      "mean_token_accuracy": 0.15763620287179947,
      "num_tokens": 3140627.0,
      "step": 1395
    },
    {
      "entropy": 6.17385630607605,
      "epoch": 1.5137912385073014,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004974016003073667,
      "loss": 6.0328,
      "mean_token_accuracy": 0.15484507232904435,
      "num_tokens": 3152294.0,
      "step": 1400
    },
    {
      "entropy": 6.2234649658203125,
      "epoch": 1.5191995673336938,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004973361990784104,
      "loss": 6.0313,
      "mean_token_accuracy": 0.15227337181568146,
      "num_tokens": 3163770.0,
      "step": 1405
    },
    {
      "entropy": 6.2478063106536865,
      "epoch": 1.5246078961600866,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972699898791927,
      "loss": 5.9855,
      "mean_token_accuracy": 0.1566041514277458,
      "num_tokens": 3175804.0,
      "step": 1410
    },
    {
      "entropy": 6.2702436447143555,
      "epoch": 1.530016224986479,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972029729503182,
      "loss": 5.9948,
      "mean_token_accuracy": 0.15855323225259782,
      "num_tokens": 3187233.0,
      "step": 1415
    },
    {
      "entropy": 6.095852327346802,
      "epoch": 1.535424553812872,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004971351485353265,
      "loss": 5.9551,
      "mean_token_accuracy": 0.16486137360334396,
      "num_tokens": 3198884.0,
      "step": 1420
    },
    {
      "entropy": 6.124053001403809,
      "epoch": 1.5408328826392643,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004970665168806917,
      "loss": 5.9263,
      "mean_token_accuracy": 0.15676868557929993,
      "num_tokens": 3209836.0,
      "step": 1425
    },
    {
      "entropy": 6.1572350025177,
      "epoch": 1.5462412114656572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969970782358214,
      "loss": 5.9607,
      "mean_token_accuracy": 0.15960703641176224,
      "num_tokens": 3221295.0,
      "step": 1430
    },
    {
      "entropy": 6.123327684402466,
      "epoch": 1.5516495402920496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004969268328530557,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16821087300777435,
      "num_tokens": 3231832.0,
      "step": 1435
    },
    {
      "entropy": 6.160953330993652,
      "epoch": 1.5570578691184425,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968557809876666,
      "loss": 5.9881,
      "mean_token_accuracy": 0.15660501420497894,
      "num_tokens": 3243483.0,
      "step": 1440
    },
    {
      "entropy": 6.189469814300537,
      "epoch": 1.562466197944835,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004967839228978568,
      "loss": 6.0146,
      "mean_token_accuracy": 0.15640730857849122,
      "num_tokens": 3255785.0,
      "step": 1445
    },
    {
      "entropy": 6.14083776473999,
      "epoch": 1.5678745267712277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967112588447585,
      "loss": 5.9098,
      "mean_token_accuracy": 0.16025495976209642,
      "num_tokens": 3267035.0,
      "step": 1450
    },
    {
      "entropy": 6.290738296508789,
      "epoch": 1.5732828555976204,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966377890924332,
      "loss": 5.9932,
      "mean_token_accuracy": 0.16156057864427567,
      "num_tokens": 3277848.0,
      "step": 1455
    },
    {
      "entropy": 6.070386457443237,
      "epoch": 1.578691184424013,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004965635139078704,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16141002327203752,
      "num_tokens": 3289864.0,
      "step": 1460
    },
    {
      "entropy": 6.239854574203491,
      "epoch": 1.5840995132504057,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964884335609857,
      "loss": 6.0348,
      "mean_token_accuracy": 0.16052431836724282,
      "num_tokens": 3301441.0,
      "step": 1465
    },
    {
      "entropy": 6.124419498443603,
      "epoch": 1.5895078420767983,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964125483246218,
      "loss": 5.9979,
      "mean_token_accuracy": 0.15076059475541115,
      "num_tokens": 3313216.0,
      "step": 1470
    },
    {
      "entropy": 6.070195484161377,
      "epoch": 1.594916170903191,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004963358584745457,
      "loss": 5.8015,
      "mean_token_accuracy": 0.16063344478607178,
      "num_tokens": 3323760.0,
      "step": 1475
    },
    {
      "entropy": 6.0324774265289305,
      "epoch": 1.6003244997295836,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962583642894482,
      "loss": 5.8812,
      "mean_token_accuracy": 0.16131513714790344,
      "num_tokens": 3334785.0,
      "step": 1480
    },
    {
      "entropy": 6.173135423660279,
      "epoch": 1.6057328285559762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961800660509437,
      "loss": 5.9541,
      "mean_token_accuracy": 0.16318012028932571,
      "num_tokens": 3345741.0,
      "step": 1485
    },
    {
      "entropy": 6.118630647659302,
      "epoch": 1.6111411573823688,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496100964043568,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16484943330287932,
      "num_tokens": 3356118.0,
      "step": 1490
    },
    {
      "entropy": 6.00065393447876,
      "epoch": 1.6165494862087615,
      "grad_norm": 0.96875,
      "learning_rate": 0.000496021058554778,
      "loss": 5.8363,
      "mean_token_accuracy": 0.16194700598716735,
      "num_tokens": 3368107.0,
      "step": 1495
    },
    {
      "entropy": 6.143065738677978,
      "epoch": 1.6219578150351541,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959403498749505,
      "loss": 5.8521,
      "mean_token_accuracy": 0.16149836629629136,
      "num_tokens": 3379367.0,
      "step": 1500
    },
    {
      "entropy": 6.014459323883057,
      "epoch": 1.6273661438615468,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958588382973809,
      "loss": 5.9791,
      "mean_token_accuracy": 0.1566162794828415,
      "num_tokens": 3390597.0,
      "step": 1505
    },
    {
      "entropy": 6.243663692474366,
      "epoch": 1.6327744726879394,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004957765241182826,
      "loss": 5.8886,
      "mean_token_accuracy": 0.1644733265042305,
      "num_tokens": 3402618.0,
      "step": 1510
    },
    {
      "entropy": 5.919700384140015,
      "epoch": 1.638182801514332,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004956934076367854,
      "loss": 5.8038,
      "mean_token_accuracy": 0.1675911948084831,
      "num_tokens": 3413162.0,
      "step": 1515
    },
    {
      "entropy": 6.054862642288208,
      "epoch": 1.6435911303407247,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495609489154935,
      "loss": 5.8933,
      "mean_token_accuracy": 0.1644514724612236,
      "num_tokens": 3424331.0,
      "step": 1520
    },
    {
      "entropy": 6.142803049087524,
      "epoch": 1.6489994591671173,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004955247689776911,
      "loss": 5.9979,
      "mean_token_accuracy": 0.15501589924097062,
      "num_tokens": 3434739.0,
      "step": 1525
    },
    {
      "entropy": 6.229400968551635,
      "epoch": 1.65440778799351,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000495439247412927,
      "loss": 6.0683,
      "mean_token_accuracy": 0.14968138113617896,
      "num_tokens": 3445984.0,
      "step": 1530
    },
    {
      "entropy": 5.973660135269165,
      "epoch": 1.6598161168199026,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004953529247714285,
      "loss": 5.8081,
      "mean_token_accuracy": 0.16547674387693406,
      "num_tokens": 3456955.0,
      "step": 1535
    },
    {
      "entropy": 6.215045833587647,
      "epoch": 1.6652244456462952,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004952658013668922,
      "loss": 5.9572,
      "mean_token_accuracy": 0.16140533089637757,
      "num_tokens": 3468895.0,
      "step": 1540
    },
    {
      "entropy": 6.016715955734253,
      "epoch": 1.670632774472688,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004951778775159246,
      "loss": 5.8493,
      "mean_token_accuracy": 0.16565765887498857,
      "num_tokens": 3480083.0,
      "step": 1545
    },
    {
      "entropy": 6.1815954685211185,
      "epoch": 1.6760411032990805,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004950891535380414,
      "loss": 5.9857,
      "mean_token_accuracy": 0.1545367419719696,
      "num_tokens": 3492242.0,
      "step": 1550
    },
    {
      "entropy": 6.1077052593231205,
      "epoch": 1.6814494321254734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004949996297556656,
      "loss": 5.8076,
      "mean_token_accuracy": 0.16661108732223512,
      "num_tokens": 3501926.0,
      "step": 1555
    },
    {
      "entropy": 5.967837524414063,
      "epoch": 1.6868577609518658,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004949093064941268,
      "loss": 5.8846,
      "mean_token_accuracy": 0.16590456813573837,
      "num_tokens": 3511421.0,
      "step": 1560
    },
    {
      "entropy": 6.1532501697540285,
      "epoch": 1.6922660897782587,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004948181840816601,
      "loss": 5.8242,
      "mean_token_accuracy": 0.16331643760204315,
      "num_tokens": 3522033.0,
      "step": 1565
    },
    {
      "entropy": 6.050137948989868,
      "epoch": 1.697674418604651,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004947262628494045,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16303865164518355,
      "num_tokens": 3533232.0,
      "step": 1570
    },
    {
      "entropy": 6.181738662719726,
      "epoch": 1.703082747431044,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004946335431314018,
      "loss": 5.8914,
      "mean_token_accuracy": 0.16298120319843293,
      "num_tokens": 3544904.0,
      "step": 1575
    },
    {
      "entropy": 6.067026996612549,
      "epoch": 1.7084910762574363,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004945400252645959,
      "loss": 5.9643,
      "mean_token_accuracy": 0.16130709350109101,
      "num_tokens": 3556545.0,
      "step": 1580
    },
    {
      "entropy": 6.050175046920776,
      "epoch": 1.7138994050838292,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004944457095888308,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16689251065254213,
      "num_tokens": 3567361.0,
      "step": 1585
    },
    {
      "entropy": 6.096579885482788,
      "epoch": 1.7193077339102216,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049435059644685,
      "loss": 5.8759,
      "mean_token_accuracy": 0.16191123872995378,
      "num_tokens": 3578943.0,
      "step": 1590
    },
    {
      "entropy": 6.071146059036255,
      "epoch": 1.7247160627366145,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004942546861842947,
      "loss": 5.8558,
      "mean_token_accuracy": 0.16385841220617295,
      "num_tokens": 3589868.0,
      "step": 1595
    },
    {
      "entropy": 6.076867294311524,
      "epoch": 1.730124391563007,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941579791497032,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1693348065018654,
      "num_tokens": 3601245.0,
      "step": 1600
    },
    {
      "entropy": 6.083900833129883,
      "epoch": 1.7355327203893998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004940604756945091,
      "loss": 5.9021,
      "mean_token_accuracy": 0.16264761984348297,
      "num_tokens": 3612168.0,
      "step": 1605
    },
    {
      "entropy": 6.028712034225464,
      "epoch": 1.7409410492157922,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004939621761730399,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16627179235219955,
      "num_tokens": 3624699.0,
      "step": 1610
    },
    {
      "entropy": 6.148156976699829,
      "epoch": 1.746349378042185,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004938630809425166,
      "loss": 5.9773,
      "mean_token_accuracy": 0.16007167398929595,
      "num_tokens": 3635554.0,
      "step": 1615
    },
    {
      "entropy": 5.953314685821534,
      "epoch": 1.7517577068685775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004937631903630513,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16080379039049147,
      "num_tokens": 3646776.0,
      "step": 1620
    },
    {
      "entropy": 6.064091157913208,
      "epoch": 1.7571660356949703,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004936625047976468,
      "loss": 5.8028,
      "mean_token_accuracy": 0.1684928059577942,
      "num_tokens": 3657976.0,
      "step": 1625
    },
    {
      "entropy": 6.161013555526734,
      "epoch": 1.7625743645213627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004935610246121944,
      "loss": 5.9253,
      "mean_token_accuracy": 0.15793002545833587,
      "num_tokens": 3668242.0,
      "step": 1630
    },
    {
      "entropy": 6.1408998489379885,
      "epoch": 1.7679826933477556,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004934587501754736,
      "loss": 5.9519,
      "mean_token_accuracy": 0.16056527346372604,
      "num_tokens": 3679200.0,
      "step": 1635
    },
    {
      "entropy": 6.017309331893921,
      "epoch": 1.7733910221741482,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004933556818591498,
      "loss": 5.8448,
      "mean_token_accuracy": 0.16986124962568283,
      "num_tokens": 3690227.0,
      "step": 1640
    },
    {
      "entropy": 6.095331621170044,
      "epoch": 1.7787993510005409,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004932518200377736,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16581207513809204,
      "num_tokens": 3701821.0,
      "step": 1645
    },
    {
      "entropy": 5.968796920776367,
      "epoch": 1.7842076798269335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004931471650887789,
      "loss": 5.8202,
      "mean_token_accuracy": 0.160505910217762,
      "num_tokens": 3712420.0,
      "step": 1650
    },
    {
      "entropy": 6.156483602523804,
      "epoch": 1.7896160086533262,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004930417173924822,
      "loss": 5.8718,
      "mean_token_accuracy": 0.1640415072441101,
      "num_tokens": 3722211.0,
      "step": 1655
    },
    {
      "entropy": 6.002517366409302,
      "epoch": 1.7950243374797188,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004929354773320806,
      "loss": 5.8544,
      "mean_token_accuracy": 0.159525004029274,
      "num_tokens": 3732822.0,
      "step": 1660
    },
    {
      "entropy": 6.117061996459961,
      "epoch": 1.8004326663061114,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004928284452936506,
      "loss": 5.9121,
      "mean_token_accuracy": 0.16279875338077546,
      "num_tokens": 3744262.0,
      "step": 1665
    },
    {
      "entropy": 5.985845041275025,
      "epoch": 1.805840995132504,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004927206216661473,
      "loss": 5.811,
      "mean_token_accuracy": 0.16297463476657867,
      "num_tokens": 3755713.0,
      "step": 1670
    },
    {
      "entropy": 6.010738372802734,
      "epoch": 1.8112493239588967,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926120068414015,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16570561081171037,
      "num_tokens": 3767179.0,
      "step": 1675
    },
    {
      "entropy": 6.005726718902588,
      "epoch": 1.8166576527852893,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004925026012141201,
      "loss": 5.8377,
      "mean_token_accuracy": 0.1643834114074707,
      "num_tokens": 3777714.0,
      "step": 1680
    },
    {
      "entropy": 6.100140476226807,
      "epoch": 1.822065981611682,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923924051818832,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16618706285953522,
      "num_tokens": 3788092.0,
      "step": 1685
    },
    {
      "entropy": 6.018685340881348,
      "epoch": 1.8274743104380746,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004922814191451435,
      "loss": 5.7929,
      "mean_token_accuracy": 0.16196684390306473,
      "num_tokens": 3798080.0,
      "step": 1690
    },
    {
      "entropy": 6.0061867237091064,
      "epoch": 1.8328826392644673,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004921696435072245,
      "loss": 5.8373,
      "mean_token_accuracy": 0.16848887354135514,
      "num_tokens": 3809547.0,
      "step": 1695
    },
    {
      "entropy": 6.0457484245300295,
      "epoch": 1.83829096809086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004920570786743191,
      "loss": 5.8173,
      "mean_token_accuracy": 0.17119550853967666,
      "num_tokens": 3820681.0,
      "step": 1700
    },
    {
      "entropy": 6.0401458740234375,
      "epoch": 1.8436992969172525,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004919437250554882,
      "loss": 5.8286,
      "mean_token_accuracy": 0.16929454803466798,
      "num_tokens": 3831598.0,
      "step": 1705
    },
    {
      "entropy": 6.023636150360107,
      "epoch": 1.8491076257436452,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004918295830626592,
      "loss": 5.8708,
      "mean_token_accuracy": 0.16968689113855362,
      "num_tokens": 3843437.0,
      "step": 1710
    },
    {
      "entropy": 5.988752841949463,
      "epoch": 1.8545159545700378,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004917146531106242,
      "loss": 5.7428,
      "mean_token_accuracy": 0.1648212730884552,
      "num_tokens": 3855387.0,
      "step": 1715
    },
    {
      "entropy": 6.023560953140259,
      "epoch": 1.8599242833964305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004915989356170389,
      "loss": 5.8914,
      "mean_token_accuracy": 0.1611024931073189,
      "num_tokens": 3866999.0,
      "step": 1720
    },
    {
      "entropy": 6.036889457702637,
      "epoch": 1.865332612222823,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914824310024213,
      "loss": 5.8236,
      "mean_token_accuracy": 0.16607335805892945,
      "num_tokens": 3877721.0,
      "step": 1725
    },
    {
      "entropy": 6.054341411590576,
      "epoch": 1.870740941049216,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004913651396901489,
      "loss": 5.9087,
      "mean_token_accuracy": 0.16150028258562088,
      "num_tokens": 3888831.0,
      "step": 1730
    },
    {
      "entropy": 6.0500908374786375,
      "epoch": 1.8761492698756084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491247062106459,
      "loss": 5.8339,
      "mean_token_accuracy": 0.16762833893299103,
      "num_tokens": 3900389.0,
      "step": 1735
    },
    {
      "entropy": 5.942811632156372,
      "epoch": 1.8815575987020012,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004911281986804455,
      "loss": 5.8029,
      "mean_token_accuracy": 0.16817400902509688,
      "num_tokens": 3911641.0,
      "step": 1740
    },
    {
      "entropy": 5.86643352508545,
      "epoch": 1.8869659275283936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910085498440587,
      "loss": 5.7026,
      "mean_token_accuracy": 0.16996440291404724,
      "num_tokens": 3922877.0,
      "step": 1745
    },
    {
      "entropy": 6.069116258621216,
      "epoch": 1.8923742563547865,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004908881160321025,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16565157771110534,
      "num_tokens": 3933573.0,
      "step": 1750
    },
    {
      "entropy": 5.921838235855103,
      "epoch": 1.897782585181179,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004907668976822338,
      "loss": 5.7278,
      "mean_token_accuracy": 0.17627133280038834,
      "num_tokens": 3944339.0,
      "step": 1755
    },
    {
      "entropy": 6.119243955612182,
      "epoch": 1.9031909140075718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004906448952349603,
      "loss": 5.8294,
      "mean_token_accuracy": 0.17736046612262726,
      "num_tokens": 3955459.0,
      "step": 1760
    },
    {
      "entropy": 5.896986055374145,
      "epoch": 1.9085992428339642,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004905221091336392,
      "loss": 5.7274,
      "mean_token_accuracy": 0.1731481984257698,
      "num_tokens": 3966782.0,
      "step": 1765
    },
    {
      "entropy": 5.9977803230285645,
      "epoch": 1.914007571660357,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004903985398244753,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16875911355018616,
      "num_tokens": 3978794.0,
      "step": 1770
    },
    {
      "entropy": 6.09336519241333,
      "epoch": 1.9194159004867495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902741877565201,
      "loss": 5.8626,
      "mean_token_accuracy": 0.16668010503053665,
      "num_tokens": 3990838.0,
      "step": 1775
    },
    {
      "entropy": 6.055723190307617,
      "epoch": 1.9248242293131423,
      "grad_norm": 1.09375,
      "learning_rate": 0.000490149053381669,
      "loss": 5.8951,
      "mean_token_accuracy": 0.1674518883228302,
      "num_tokens": 4000630.0,
      "step": 1780
    },
    {
      "entropy": 6.027546167373657,
      "epoch": 1.9302325581395348,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004900231371546608,
      "loss": 5.7717,
      "mean_token_accuracy": 0.17280880063772203,
      "num_tokens": 4011433.0,
      "step": 1785
    },
    {
      "entropy": 5.9891197681427,
      "epoch": 1.9356408869659276,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004898964395330753,
      "loss": 5.9187,
      "mean_token_accuracy": 0.166441847383976,
      "num_tokens": 4023671.0,
      "step": 1790
    },
    {
      "entropy": 5.98225588798523,
      "epoch": 1.94104921579232,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004897689609773318,
      "loss": 5.705,
      "mean_token_accuracy": 0.16938872784376144,
      "num_tokens": 4034686.0,
      "step": 1795
    },
    {
      "entropy": 6.053580093383789,
      "epoch": 1.946457544618713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004896407019506879,
      "loss": 5.8215,
      "mean_token_accuracy": 0.17409656196832657,
      "num_tokens": 4046387.0,
      "step": 1800
    },
    {
      "entropy": 5.924436092376709,
      "epoch": 1.9518658734451053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895116629192371,
      "loss": 5.7349,
      "mean_token_accuracy": 0.17091045677661895,
      "num_tokens": 4056103.0,
      "step": 1805
    },
    {
      "entropy": 5.991981029510498,
      "epoch": 1.9572742022714982,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004893818443519074,
      "loss": 5.7,
      "mean_token_accuracy": 0.18084933310747148,
      "num_tokens": 4066940.0,
      "step": 1810
    },
    {
      "entropy": 5.908279466629028,
      "epoch": 1.9626825310978906,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048925124672046,
      "loss": 5.8338,
      "mean_token_accuracy": 0.17757519632577895,
      "num_tokens": 4077232.0,
      "step": 1815
    },
    {
      "entropy": 6.0143650531768795,
      "epoch": 1.9680908599242835,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004891198704994868,
      "loss": 5.8963,
      "mean_token_accuracy": 0.16193630546331406,
      "num_tokens": 4089322.0,
      "step": 1820
    },
    {
      "entropy": 5.993407154083252,
      "epoch": 1.973499188750676,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004889877161664096,
      "loss": 5.7359,
      "mean_token_accuracy": 0.1731564626097679,
      "num_tokens": 4099737.0,
      "step": 1825
    },
    {
      "entropy": 5.926445531845093,
      "epoch": 1.9789075175770687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004888547842014771,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16823544651269912,
      "num_tokens": 4110793.0,
      "step": 1830
    },
    {
      "entropy": 6.006844568252563,
      "epoch": 1.9843158464034614,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004887210750877648,
      "loss": 5.781,
      "mean_token_accuracy": 0.17114681750535965,
      "num_tokens": 4121481.0,
      "step": 1835
    },
    {
      "entropy": 5.902514410018921,
      "epoch": 1.989724175229854,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885865893111716,
      "loss": 5.7411,
      "mean_token_accuracy": 0.17603601664304733,
      "num_tokens": 4132701.0,
      "step": 1840
    },
    {
      "entropy": 5.985061120986939,
      "epoch": 1.9951325040562466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004884513273604195,
      "loss": 5.8062,
      "mean_token_accuracy": 0.17002770006656648,
      "num_tokens": 4143764.0,
      "step": 1845
    },
    {
      "entropy": 6.0021616617838545,
      "epoch": 2.0,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004883152897270502,
      "loss": 5.7771,
      "mean_token_accuracy": 0.170842116077741,
      "num_tokens": 4153778.0,
      "step": 1850
    },
    {
      "entropy": 5.96401014328003,
      "epoch": 2.005408328826393,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004881784769054253,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18372500836849212,
      "num_tokens": 4165099.0,
      "step": 1855
    },
    {
      "entropy": 5.953923511505127,
      "epoch": 2.0108166576527853,
      "grad_norm": 0.875,
      "learning_rate": 0.0004880408893927226,
      "loss": 5.6148,
      "mean_token_accuracy": 0.1810345396399498,
      "num_tokens": 4178051.0,
      "step": 1860
    },
    {
      "entropy": 5.986582851409912,
      "epoch": 2.016224986479178,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004879025276889355,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18339935392141343,
      "num_tokens": 4189987.0,
      "step": 1865
    },
    {
      "entropy": 6.020156669616699,
      "epoch": 2.0216333153055706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048776339229687077,
      "loss": 5.5614,
      "mean_token_accuracy": 0.18038318008184434,
      "num_tokens": 4200999.0,
      "step": 1870
    },
    {
      "entropy": 5.838561725616455,
      "epoch": 2.0270416441319634,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048762348372214683,
      "loss": 5.488,
      "mean_token_accuracy": 0.17765122652053833,
      "num_tokens": 4212122.0,
      "step": 1875
    },
    {
      "entropy": 5.974667358398437,
      "epoch": 2.032449972958356,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048748280247319165,
      "loss": 5.5611,
      "mean_token_accuracy": 0.18277785331010818,
      "num_tokens": 4222580.0,
      "step": 1880
    },
    {
      "entropy": 5.839652013778687,
      "epoch": 2.0378583017847487,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004873413490612412,
      "loss": 5.4009,
      "mean_token_accuracy": 0.18563185334205629,
      "num_tokens": 4233756.0,
      "step": 1885
    },
    {
      "entropy": 5.89198784828186,
      "epoch": 2.043266630611141,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048719912400033757,
      "loss": 5.53,
      "mean_token_accuracy": 0.17903512865304946,
      "num_tokens": 4245625.0,
      "step": 1890
    },
    {
      "entropy": 6.023007154464722,
      "epoch": 2.048674959437534,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000487056127807327,
      "loss": 5.7544,
      "mean_token_accuracy": 0.176625756919384,
      "num_tokens": 4257788.0,
      "step": 1895
    },
    {
      "entropy": 5.884478378295898,
      "epoch": 2.0540832882639264,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048691236100185795,
      "loss": 5.4863,
      "mean_token_accuracy": 0.18627301156520842,
      "num_tokens": 4269641.0,
      "step": 1900
    },
    {
      "entropy": 5.923391056060791,
      "epoch": 2.0594916170903192,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004867678241063792,
      "loss": 5.5542,
      "mean_token_accuracy": 0.18891489654779434,
      "num_tokens": 4280952.0,
      "step": 1905
    },
    {
      "entropy": 5.90982141494751,
      "epoch": 2.0648999459167117,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048662251764613833,
      "loss": 5.4983,
      "mean_token_accuracy": 0.18720743656158448,
      "num_tokens": 4294371.0,
      "step": 1910
    },
    {
      "entropy": 5.810564851760864,
      "epoch": 2.0703082747431045,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004864764421491793,
      "loss": 5.5078,
      "mean_token_accuracy": 0.18103248476982117,
      "num_tokens": 4307112.0,
      "step": 1915
    },
    {
      "entropy": 5.906817483901977,
      "epoch": 2.075716603569497,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004863295981463408,
      "loss": 5.4927,
      "mean_token_accuracy": 0.18527511060237883,
      "num_tokens": 4317199.0,
      "step": 1920
    },
    {
      "entropy": 5.839090013504029,
      "epoch": 2.08112493239589,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004861819861712543,
      "loss": 5.492,
      "mean_token_accuracy": 0.1836896702647209,
      "num_tokens": 4328546.0,
      "step": 1925
    },
    {
      "entropy": 5.978621196746826,
      "epoch": 2.086533261222282,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048603360676034203,
      "loss": 5.545,
      "mean_token_accuracy": 0.18170068711042403,
      "num_tokens": 4338795.0,
      "step": 1930
    },
    {
      "entropy": 5.801554346084595,
      "epoch": 2.091941590048675,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048588446045281505,
      "loss": 5.535,
      "mean_token_accuracy": 0.17918962985277176,
      "num_tokens": 4350373.0,
      "step": 1935
    },
    {
      "entropy": 5.860919666290283,
      "epoch": 2.0973499188750675,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004857345477906714,
      "loss": 5.5141,
      "mean_token_accuracy": 0.17788466066122055,
      "num_tokens": 4362801.0,
      "step": 1940
    },
    {
      "entropy": 5.988518285751343,
      "epoch": 2.1027582477014604,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048558386931869417,
      "loss": 5.5982,
      "mean_token_accuracy": 0.18577716499567032,
      "num_tokens": 4375782.0,
      "step": 1945
    },
    {
      "entropy": 5.7955482006073,
      "epoch": 2.1081665765278528,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004854324255844492,
      "loss": 5.5003,
      "mean_token_accuracy": 0.18517474532127381,
      "num_tokens": 4385580.0,
      "step": 1950
    },
    {
      "entropy": 5.883208179473877,
      "epoch": 2.1135749053542456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048528021713828336,
      "loss": 5.5169,
      "mean_token_accuracy": 0.17969103306531906,
      "num_tokens": 4397434.0,
      "step": 1955
    },
    {
      "entropy": 5.881035614013672,
      "epoch": 2.118983234180638,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004851272445333225,
      "loss": 5.5583,
      "mean_token_accuracy": 0.17854398339986802,
      "num_tokens": 4407888.0,
      "step": 1960
    },
    {
      "entropy": 5.829453849792481,
      "epoch": 2.124391563007031,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048497350832546937,
      "loss": 5.4845,
      "mean_token_accuracy": 0.1851338788866997,
      "num_tokens": 4420010.0,
      "step": 1965
    },
    {
      "entropy": 5.817228889465332,
      "epoch": 2.1297998918334233,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004848190090734018,
      "loss": 5.5211,
      "mean_token_accuracy": 0.17922184765338897,
      "num_tokens": 4430730.0,
      "step": 1970
    },
    {
      "entropy": 5.825629997253418,
      "epoch": 2.135208220659816,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048466374733857035,
      "loss": 5.4969,
      "mean_token_accuracy": 0.18972273021936417,
      "num_tokens": 4442296.0,
      "step": 1975
    },
    {
      "entropy": 5.90300760269165,
      "epoch": 2.1406165494862086,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004845077236851966,
      "loss": 5.5284,
      "mean_token_accuracy": 0.18410815596580504,
      "num_tokens": 4453996.0,
      "step": 1980
    },
    {
      "entropy": 5.782866811752319,
      "epoch": 2.1460248783126015,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048435093868027076,
      "loss": 5.5212,
      "mean_token_accuracy": 0.18278288543224336,
      "num_tokens": 4465779.0,
      "step": 1985
    },
    {
      "entropy": 5.725567245483399,
      "epoch": 2.151433207138994,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048419339289354995,
      "loss": 5.4559,
      "mean_token_accuracy": 0.1890066847205162,
      "num_tokens": 4477113.0,
      "step": 1990
    },
    {
      "entropy": 5.857697677612305,
      "epoch": 2.1568415359653867,
      "grad_norm": 1.046875,
      "learning_rate": 0.000484035086897556,
      "loss": 5.4231,
      "mean_token_accuracy": 0.1865213468670845,
      "num_tokens": 4487920.0,
      "step": 1995
    },
    {
      "entropy": 5.944146585464478,
      "epoch": 2.162249864791779,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004838760212675732,
      "loss": 5.6188,
      "mean_token_accuracy": 0.17914327830076218,
      "num_tokens": 4499049.0,
      "step": 2000
    },
    {
      "epoch": 2.162249864791779,
      "eval_entropy": 5.714854424650019,
      "eval_loss": 6.0214762687683105,
      "eval_mean_token_accuracy": 0.1667367118156769,
      "eval_num_tokens": 4499049.0,
      "eval_runtime": 1.9484,
      "eval_samples_per_second": 1804.032,
      "eval_steps_per_second": 225.825,
      "step": 2000
    },
    {
      "entropy": 5.827992010116577,
      "epoch": 2.167658193618172,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004837161965816464,
      "loss": 5.4443,
      "mean_token_accuracy": 0.18775694668293,
      "num_tokens": 4508944.0,
      "step": 2005
    },
    {
      "entropy": 5.858482933044433,
      "epoch": 2.1730665224445644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048355561342057886,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18398282378911973,
      "num_tokens": 4518845.0,
      "step": 2010
    },
    {
      "entropy": 5.702045488357544,
      "epoch": 2.1784748512709573,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048339427236793024,
      "loss": 5.5246,
      "mean_token_accuracy": 0.18108364194631577,
      "num_tokens": 4530416.0,
      "step": 2015
    },
    {
      "entropy": 5.896171426773071,
      "epoch": 2.1838831800973497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004832321740100143,
      "loss": 5.507,
      "mean_token_accuracy": 0.18198190331459047,
      "num_tokens": 4542630.0,
      "step": 2020
    },
    {
      "entropy": 5.734747648239136,
      "epoch": 2.1892915089237426,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048306931893589666,
      "loss": 5.4876,
      "mean_token_accuracy": 0.18120697140693665,
      "num_tokens": 4553270.0,
      "step": 2025
    },
    {
      "entropy": 5.722292709350586,
      "epoch": 2.194699837750135,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048290570773739327,
      "loss": 5.3589,
      "mean_token_accuracy": 0.18821881115436553,
      "num_tokens": 4564647.0,
      "step": 2030
    },
    {
      "entropy": 5.8666078567504885,
      "epoch": 2.200108166576528,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048274134100906747,
      "loss": 5.5461,
      "mean_token_accuracy": 0.17590499073266982,
      "num_tokens": 4575507.0,
      "step": 2035
    },
    {
      "entropy": 5.806071996688843,
      "epoch": 2.2055164954029207,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048257621934822835,
      "loss": 5.5291,
      "mean_token_accuracy": 0.18711154013872147,
      "num_tokens": 4586761.0,
      "step": 2040
    },
    {
      "entropy": 5.773828077316284,
      "epoch": 2.210924824229313,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004824103433549285,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1893257662653923,
      "num_tokens": 4597689.0,
      "step": 2045
    },
    {
      "entropy": 5.8871923923492435,
      "epoch": 2.2163331530557056,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004822437136319616,
      "loss": 5.4918,
      "mean_token_accuracy": 0.18168212175369264,
      "num_tokens": 4608886.0,
      "step": 2050
    },
    {
      "entropy": 5.820661115646362,
      "epoch": 2.2217414818820984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004820763307848604,
      "loss": 5.5406,
      "mean_token_accuracy": 0.17569175064563752,
      "num_tokens": 4619989.0,
      "step": 2055
    },
    {
      "entropy": 5.758163261413574,
      "epoch": 2.2271498107084913,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048190819542189456,
      "loss": 5.4214,
      "mean_token_accuracy": 0.1874105900526047,
      "num_tokens": 4630218.0,
      "step": 2060
    },
    {
      "entropy": 5.7546144962310795,
      "epoch": 2.2325581395348837,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048173930815406847,
      "loss": 5.509,
      "mean_token_accuracy": 0.1754484325647354,
      "num_tokens": 4642024.0,
      "step": 2065
    },
    {
      "entropy": 5.887257528305054,
      "epoch": 2.2379664683612766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004815696695951187,
      "loss": 5.5216,
      "mean_token_accuracy": 0.18524898290634156,
      "num_tokens": 4653370.0,
      "step": 2070
    },
    {
      "entropy": 5.798015069961548,
      "epoch": 2.243374797187669,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004813992803615123,
      "loss": 5.5252,
      "mean_token_accuracy": 0.18299013376235962,
      "num_tokens": 4664522.0,
      "step": 2075
    },
    {
      "entropy": 5.810057783126831,
      "epoch": 2.248783126014062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048122814107244414,
      "loss": 5.5501,
      "mean_token_accuracy": 0.1842383101582527,
      "num_tokens": 4676542.0,
      "step": 2080
    },
    {
      "entropy": 5.778118324279785,
      "epoch": 2.2541914548404542,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048105625234983466,
      "loss": 5.4775,
      "mean_token_accuracy": 0.18344338834285737,
      "num_tokens": 4687810.0,
      "step": 2085
    },
    {
      "entropy": 5.856587934494018,
      "epoch": 2.259599783666847,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048088361481832807,
      "loss": 5.4774,
      "mean_token_accuracy": 0.18670579940080642,
      "num_tokens": 4697661.0,
      "step": 2090
    },
    {
      "entropy": 5.744546270370483,
      "epoch": 2.2650081124932395,
      "grad_norm": 1.0,
      "learning_rate": 0.00048071022910528934,
      "loss": 5.4055,
      "mean_token_accuracy": 0.18543411344289779,
      "num_tokens": 4708585.0,
      "step": 2095
    },
    {
      "entropy": 5.860214710235596,
      "epoch": 2.2704164413196324,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048053609584080274,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18319359570741653,
      "num_tokens": 4719848.0,
      "step": 2100
    },
    {
      "entropy": 5.85261378288269,
      "epoch": 2.275824770146025,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048036121565766876,
      "loss": 5.5494,
      "mean_token_accuracy": 0.17277902811765672,
      "num_tokens": 4731474.0,
      "step": 2105
    },
    {
      "entropy": 5.786352729797363,
      "epoch": 2.2812330989724177,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004801855891914026,
      "loss": 5.5191,
      "mean_token_accuracy": 0.18504545837640762,
      "num_tokens": 4743081.0,
      "step": 2110
    },
    {
      "entropy": 5.839174079895019,
      "epoch": 2.28664142779881,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048000921708023117,
      "loss": 5.5099,
      "mean_token_accuracy": 0.1780516564846039,
      "num_tokens": 4754829.0,
      "step": 2115
    },
    {
      "entropy": 5.780178022384644,
      "epoch": 2.292049756625203,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004798320999650911,
      "loss": 5.4635,
      "mean_token_accuracy": 0.1831834301352501,
      "num_tokens": 4765522.0,
      "step": 2120
    },
    {
      "entropy": 5.7943672180175785,
      "epoch": 2.2974580854515954,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047965423848962636,
      "loss": 5.4492,
      "mean_token_accuracy": 0.18973132371902465,
      "num_tokens": 4776246.0,
      "step": 2125
    },
    {
      "entropy": 5.773667573928833,
      "epoch": 2.302866414277988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479475633300186,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18006015866994857,
      "num_tokens": 4787595.0,
      "step": 2130
    },
    {
      "entropy": 5.869189023971558,
      "epoch": 2.3082747431043806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047929628504582173,
      "loss": 5.5689,
      "mean_token_accuracy": 0.17607534378767015,
      "num_tokens": 4798877.0,
      "step": 2135
    },
    {
      "entropy": 5.800918674468994,
      "epoch": 2.3136830719307735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004791161943782856,
      "loss": 5.5703,
      "mean_token_accuracy": 0.1844577670097351,
      "num_tokens": 4809542.0,
      "step": 2140
    },
    {
      "entropy": 5.903112983703613,
      "epoch": 2.319091400757166,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047893536195202726,
      "loss": 5.6694,
      "mean_token_accuracy": 0.1777377888560295,
      "num_tokens": 4820982.0,
      "step": 2145
    },
    {
      "entropy": 5.758823585510254,
      "epoch": 2.3244997295835588,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004787537884241923,
      "loss": 5.5097,
      "mean_token_accuracy": 0.1798255279660225,
      "num_tokens": 4831543.0,
      "step": 2150
    },
    {
      "entropy": 6.009968376159668,
      "epoch": 2.329908058409951,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047857147445461943,
      "loss": 5.5692,
      "mean_token_accuracy": 0.179595747590065,
      "num_tokens": 4843221.0,
      "step": 2155
    },
    {
      "entropy": 5.667755794525147,
      "epoch": 2.335316387236344,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004783884207058379,
      "loss": 5.4319,
      "mean_token_accuracy": 0.18061627000570296,
      "num_tokens": 4855201.0,
      "step": 2160
    },
    {
      "entropy": 5.849899578094482,
      "epoch": 2.3407247160627365,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047820462784306555,
      "loss": 5.5254,
      "mean_token_accuracy": 0.18278025984764099,
      "num_tokens": 4866387.0,
      "step": 2165
    },
    {
      "entropy": 5.765425682067871,
      "epoch": 2.3461330448891293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000478020096534206,
      "loss": 5.4552,
      "mean_token_accuracy": 0.18595564514398574,
      "num_tokens": 4877702.0,
      "step": 2170
    },
    {
      "entropy": 5.825589847564697,
      "epoch": 2.3515413737155217,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047783482744984644,
      "loss": 5.5257,
      "mean_token_accuracy": 0.18141573518514634,
      "num_tokens": 4888652.0,
      "step": 2175
    },
    {
      "entropy": 5.823223686218261,
      "epoch": 2.3569497025419146,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047764882126325527,
      "loss": 5.6121,
      "mean_token_accuracy": 0.17831413000822066,
      "num_tokens": 4900995.0,
      "step": 2180
    },
    {
      "entropy": 5.771738338470459,
      "epoch": 2.362358031368307,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004774620786503793,
      "loss": 5.3801,
      "mean_token_accuracy": 0.18677951544523239,
      "num_tokens": 4912173.0,
      "step": 2185
    },
    {
      "entropy": 5.687531471252441,
      "epoch": 2.3677663601947,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004772746002898417,
      "loss": 5.4263,
      "mean_token_accuracy": 0.18884409070014954,
      "num_tokens": 4923149.0,
      "step": 2190
    },
    {
      "entropy": 5.761385726928711,
      "epoch": 2.3731746890210923,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047708638686293903,
      "loss": 5.4345,
      "mean_token_accuracy": 0.18475202023983,
      "num_tokens": 4934156.0,
      "step": 2195
    },
    {
      "entropy": 5.86796875,
      "epoch": 2.378583017847485,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047689743905363963,
      "loss": 5.6353,
      "mean_token_accuracy": 0.1786823108792305,
      "num_tokens": 4946690.0,
      "step": 2200
    },
    {
      "entropy": 5.854101324081421,
      "epoch": 2.383991346673878,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004767077575485801,
      "loss": 5.5328,
      "mean_token_accuracy": 0.19112526029348373,
      "num_tokens": 4958208.0,
      "step": 2205
    },
    {
      "entropy": 5.713196659088135,
      "epoch": 2.3893996755002704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004765173430370635,
      "loss": 5.4818,
      "mean_token_accuracy": 0.19157323986291885,
      "num_tokens": 4968815.0,
      "step": 2210
    },
    {
      "entropy": 5.8433257102966305,
      "epoch": 2.394808004326663,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004763261962110567,
      "loss": 5.4744,
      "mean_token_accuracy": 0.19104113727807998,
      "num_tokens": 4979468.0,
      "step": 2215
    },
    {
      "entropy": 5.799380111694336,
      "epoch": 2.4002163331530557,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004761343177651875,
      "loss": 5.5455,
      "mean_token_accuracy": 0.1798015207052231,
      "num_tokens": 4990841.0,
      "step": 2220
    },
    {
      "entropy": 5.90300669670105,
      "epoch": 2.4056246619794486,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004759417083967428,
      "loss": 5.5932,
      "mean_token_accuracy": 0.18243164122104644,
      "num_tokens": 5001567.0,
      "step": 2225
    },
    {
      "entropy": 5.7606593608856205,
      "epoch": 2.411032990805841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047574836880566544,
      "loss": 5.452,
      "mean_token_accuracy": 0.18293080925941468,
      "num_tokens": 5013219.0,
      "step": 2230
    },
    {
      "entropy": 5.777562761306763,
      "epoch": 2.4164413196322334,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004755542996945519,
      "loss": 5.5853,
      "mean_token_accuracy": 0.18079853951931,
      "num_tokens": 5025771.0,
      "step": 2235
    },
    {
      "entropy": 5.819476509094239,
      "epoch": 2.4218496484586263,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047535950176864976,
      "loss": 5.5355,
      "mean_token_accuracy": 0.18410034775733947,
      "num_tokens": 5038205.0,
      "step": 2240
    },
    {
      "entropy": 5.818848896026611,
      "epoch": 2.427257977285019,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047516397573585525,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1823202833533287,
      "num_tokens": 5048428.0,
      "step": 2245
    },
    {
      "entropy": 5.757763004302978,
      "epoch": 2.4326663061114115,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047496772230671034,
      "loss": 5.474,
      "mean_token_accuracy": 0.18468060344457626,
      "num_tokens": 5058391.0,
      "step": 2250
    },
    {
      "entropy": 5.7085260391235355,
      "epoch": 2.4380746349378044,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004747707421944004,
      "loss": 5.5076,
      "mean_token_accuracy": 0.17939960956573486,
      "num_tokens": 5069832.0,
      "step": 2255
    },
    {
      "entropy": 5.826162624359131,
      "epoch": 2.443482963764197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004745730361147517,
      "loss": 5.4675,
      "mean_token_accuracy": 0.19406913667917253,
      "num_tokens": 5080198.0,
      "step": 2260
    },
    {
      "entropy": 5.757597923278809,
      "epoch": 2.4488912925905897,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004743746047862286,
      "loss": 5.5505,
      "mean_token_accuracy": 0.18233709186315536,
      "num_tokens": 5090817.0,
      "step": 2265
    },
    {
      "entropy": 5.749329280853272,
      "epoch": 2.454299621416982,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004741754489299309,
      "loss": 5.4435,
      "mean_token_accuracy": 0.1810994803905487,
      "num_tokens": 5102699.0,
      "step": 2270
    },
    {
      "entropy": 5.751941967010498,
      "epoch": 2.459707950243375,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047397556926959155,
      "loss": 5.4318,
      "mean_token_accuracy": 0.18434558510780336,
      "num_tokens": 5113922.0,
      "step": 2275
    },
    {
      "entropy": 5.78816180229187,
      "epoch": 2.4651162790697674,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047377496653157374,
      "loss": 5.55,
      "mean_token_accuracy": 0.18299137353897094,
      "num_tokens": 5124920.0,
      "step": 2280
    },
    {
      "entropy": 5.737819004058838,
      "epoch": 2.4705246078961602,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047357364144486806,
      "loss": 5.5297,
      "mean_token_accuracy": 0.1725105255842209,
      "num_tokens": 5136304.0,
      "step": 2285
    },
    {
      "entropy": 5.873019361495972,
      "epoch": 2.4759329367225527,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047337159474109057,
      "loss": 5.6245,
      "mean_token_accuracy": 0.1747826248407364,
      "num_tokens": 5148101.0,
      "step": 2290
    },
    {
      "entropy": 5.8381751537322994,
      "epoch": 2.4813412655489455,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047316882715447934,
      "loss": 5.5139,
      "mean_token_accuracy": 0.18714123517274855,
      "num_tokens": 5159186.0,
      "step": 2295
    },
    {
      "entropy": 5.7168231964111325,
      "epoch": 2.486749594375338,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004729653394218922,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18206719160079957,
      "num_tokens": 5170542.0,
      "step": 2300
    },
    {
      "entropy": 5.775171995162964,
      "epoch": 2.492157923201731,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047276113228280427,
      "loss": 5.5797,
      "mean_token_accuracy": 0.175926311314106,
      "num_tokens": 5182685.0,
      "step": 2305
    },
    {
      "entropy": 5.788300895690918,
      "epoch": 2.497566252028123,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004725562064793045,
      "loss": 5.4108,
      "mean_token_accuracy": 0.18709418922662735,
      "num_tokens": 5192622.0,
      "step": 2310
    },
    {
      "entropy": 5.8062903881073,
      "epoch": 2.502974580854516,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047235056275609397,
      "loss": 5.5926,
      "mean_token_accuracy": 0.17485432624816893,
      "num_tokens": 5205068.0,
      "step": 2315
    },
    {
      "entropy": 5.720317649841308,
      "epoch": 2.5083829096809085,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004721442018604822,
      "loss": 5.459,
      "mean_token_accuracy": 0.18698694854974746,
      "num_tokens": 5216049.0,
      "step": 2320
    },
    {
      "entropy": 5.68174901008606,
      "epoch": 2.5137912385073014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004719371245423853,
      "loss": 5.4598,
      "mean_token_accuracy": 0.18682386875152587,
      "num_tokens": 5227024.0,
      "step": 2325
    },
    {
      "entropy": 5.831164598464966,
      "epoch": 2.5191995673336938,
      "grad_norm": 1.125,
      "learning_rate": 0.00047172933155432273,
      "loss": 5.5655,
      "mean_token_accuracy": 0.1825097918510437,
      "num_tokens": 5238862.0,
      "step": 2330
    },
    {
      "entropy": 5.86603012084961,
      "epoch": 2.5246078961600866,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047152082365141466,
      "loss": 5.6009,
      "mean_token_accuracy": 0.1768431156873703,
      "num_tokens": 5250020.0,
      "step": 2335
    },
    {
      "entropy": 5.766449117660523,
      "epoch": 2.530016224986479,
      "grad_norm": 1.0,
      "learning_rate": 0.0004713116015913794,
      "loss": 5.5203,
      "mean_token_accuracy": 0.17895119786262512,
      "num_tokens": 5260281.0,
      "step": 2340
    },
    {
      "entropy": 5.876428651809692,
      "epoch": 2.535424553812872,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047110166613453016,
      "loss": 5.6226,
      "mean_token_accuracy": 0.1823361337184906,
      "num_tokens": 5272909.0,
      "step": 2345
    },
    {
      "entropy": 5.728987216949463,
      "epoch": 2.5408328826392643,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047089101804377305,
      "loss": 5.4534,
      "mean_token_accuracy": 0.1881086751818657,
      "num_tokens": 5285629.0,
      "step": 2350
    },
    {
      "entropy": 5.841614198684693,
      "epoch": 2.546241211465657,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004706796580846037,
      "loss": 5.5125,
      "mean_token_accuracy": 0.18212157040834426,
      "num_tokens": 5296104.0,
      "step": 2355
    },
    {
      "entropy": 5.869363403320312,
      "epoch": 2.5516495402920496,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047046758702510465,
      "loss": 5.6386,
      "mean_token_accuracy": 0.17351340651512145,
      "num_tokens": 5307148.0,
      "step": 2360
    },
    {
      "entropy": 5.71482105255127,
      "epoch": 2.5570578691184425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047025480563594264,
      "loss": 5.461,
      "mean_token_accuracy": 0.1860464632511139,
      "num_tokens": 5317878.0,
      "step": 2365
    },
    {
      "entropy": 5.838516664505005,
      "epoch": 2.5624661979448353,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047004131469036576,
      "loss": 5.5592,
      "mean_token_accuracy": 0.17682382017374038,
      "num_tokens": 5329603.0,
      "step": 2370
    },
    {
      "entropy": 5.720700931549072,
      "epoch": 2.5678745267712277,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046982711496420057,
      "loss": 5.5078,
      "mean_token_accuracy": 0.17882330268621444,
      "num_tokens": 5340611.0,
      "step": 2375
    },
    {
      "entropy": 5.791160964965821,
      "epoch": 2.57328285559762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046961220723584925,
      "loss": 5.4301,
      "mean_token_accuracy": 0.18916528671979904,
      "num_tokens": 5352017.0,
      "step": 2380
    },
    {
      "entropy": 5.730231904983521,
      "epoch": 2.578691184424013,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046939659228628725,
      "loss": 5.4991,
      "mean_token_accuracy": 0.18712244629859925,
      "num_tokens": 5362482.0,
      "step": 2385
    },
    {
      "entropy": 5.7271409034729,
      "epoch": 2.584099513250406,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046918027089905956,
      "loss": 5.4872,
      "mean_token_accuracy": 0.18555831909179688,
      "num_tokens": 5374074.0,
      "step": 2390
    },
    {
      "entropy": 5.772935438156128,
      "epoch": 2.5895078420767983,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046896324386027873,
      "loss": 5.3009,
      "mean_token_accuracy": 0.19464293718338013,
      "num_tokens": 5384650.0,
      "step": 2395
    },
    {
      "entropy": 5.599877738952637,
      "epoch": 2.5949161709031907,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046874551195862145,
      "loss": 5.5144,
      "mean_token_accuracy": 0.18187478333711624,
      "num_tokens": 5395184.0,
      "step": 2400
    },
    {
      "entropy": 5.734960556030273,
      "epoch": 2.6003244997295836,
      "grad_norm": 1.0625,
      "learning_rate": 0.000468527075985326,
      "loss": 5.3616,
      "mean_token_accuracy": 0.19382341653108598,
      "num_tokens": 5405947.0,
      "step": 2405
    },
    {
      "entropy": 5.675955486297608,
      "epoch": 2.6057328285559764,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00046830793673418924,
      "loss": 5.4578,
      "mean_token_accuracy": 0.18990024775266648,
      "num_tokens": 5418237.0,
      "step": 2410
    },
    {
      "entropy": 5.731205701828003,
      "epoch": 2.611141157382369,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046808809500156355,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18150573670864106,
      "num_tokens": 5429336.0,
      "step": 2415
    },
    {
      "entropy": 5.783448934555054,
      "epoch": 2.6165494862087613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004678675515863545,
      "loss": 5.4827,
      "mean_token_accuracy": 0.18304021060466766,
      "num_tokens": 5440257.0,
      "step": 2420
    },
    {
      "entropy": 5.867169570922852,
      "epoch": 2.621957815035154,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004676463072900173,
      "loss": 5.6001,
      "mean_token_accuracy": 0.1810256853699684,
      "num_tokens": 5453079.0,
      "step": 2425
    },
    {
      "entropy": 5.670599937438965,
      "epoch": 2.627366143861547,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046742436291655415,
      "loss": 5.4609,
      "mean_token_accuracy": 0.18477327674627303,
      "num_tokens": 5464551.0,
      "step": 2430
    },
    {
      "entropy": 5.80149884223938,
      "epoch": 2.6327744726879394,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046720171927251155,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1902604267001152,
      "num_tokens": 5474797.0,
      "step": 2435
    },
    {
      "entropy": 5.745089530944824,
      "epoch": 2.638182801514332,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004669783771669769,
      "loss": 5.4632,
      "mean_token_accuracy": 0.1883749097585678,
      "num_tokens": 5485760.0,
      "step": 2440
    },
    {
      "entropy": 5.777961730957031,
      "epoch": 2.6435911303407247,
      "grad_norm": 1.0,
      "learning_rate": 0.000466754337411576,
      "loss": 5.4424,
      "mean_token_accuracy": 0.19193070828914643,
      "num_tokens": 5496182.0,
      "step": 2445
    },
    {
      "entropy": 5.700955152511597,
      "epoch": 2.6489994591671175,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046652960082046985,
      "loss": 5.5258,
      "mean_token_accuracy": 0.19253770709037782,
      "num_tokens": 5506848.0,
      "step": 2450
    },
    {
      "entropy": 5.823722076416016,
      "epoch": 2.65440778799351,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046630416821035164,
      "loss": 5.6417,
      "mean_token_accuracy": 0.17784025222063066,
      "num_tokens": 5518430.0,
      "step": 2455
    },
    {
      "entropy": 5.798423957824707,
      "epoch": 2.6598161168199024,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004660780404004441,
      "loss": 5.5336,
      "mean_token_accuracy": 0.17896816432476043,
      "num_tokens": 5529741.0,
      "step": 2460
    },
    {
      "entropy": 5.668118810653686,
      "epoch": 2.6652244456462952,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046585121821249623,
      "loss": 5.4998,
      "mean_token_accuracy": 0.18824873566627504,
      "num_tokens": 5541835.0,
      "step": 2465
    },
    {
      "entropy": 5.769140815734863,
      "epoch": 2.670632774472688,
      "grad_norm": 1.0,
      "learning_rate": 0.00046562370247078025,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1851413369178772,
      "num_tokens": 5552177.0,
      "step": 2470
    },
    {
      "entropy": 5.7882355690002445,
      "epoch": 2.6760411032990805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000465395494002089,
      "loss": 5.6132,
      "mean_token_accuracy": 0.17335510849952698,
      "num_tokens": 5564440.0,
      "step": 2475
    },
    {
      "entropy": 5.7682489395141605,
      "epoch": 2.6814494321254734,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004651665936357325,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19111803621053697,
      "num_tokens": 5574905.0,
      "step": 2480
    },
    {
      "entropy": 5.714561653137207,
      "epoch": 2.686857760951866,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004649370022035354,
      "loss": 5.4829,
      "mean_token_accuracy": 0.17880124896764754,
      "num_tokens": 5585583.0,
      "step": 2485
    },
    {
      "entropy": 5.684116697311401,
      "epoch": 2.6922660897782587,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004647067205398332,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1880973681807518,
      "num_tokens": 5597077.0,
      "step": 2490
    },
    {
      "entropy": 5.741129684448242,
      "epoch": 2.697674418604651,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046447574948147027,
      "loss": 5.3192,
      "mean_token_accuracy": 0.19412106722593309,
      "num_tokens": 5607904.0,
      "step": 2495
    },
    {
      "entropy": 5.691903400421142,
      "epoch": 2.703082747431044,
      "grad_norm": 0.859375,
      "learning_rate": 0.00046424408986779587,
      "loss": 5.5088,
      "mean_token_accuracy": 0.18323247134685516,
      "num_tokens": 5619611.0,
      "step": 2500
    },
    {
      "entropy": 5.832025861740112,
      "epoch": 2.7084910762574363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004640117425406615,
      "loss": 5.525,
      "mean_token_accuracy": 0.1868831470608711,
      "num_tokens": 5630365.0,
      "step": 2505
    },
    {
      "entropy": 5.753492307662964,
      "epoch": 2.713899405083829,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000463778708344418,
      "loss": 5.4781,
      "mean_token_accuracy": 0.1851469397544861,
      "num_tokens": 5641008.0,
      "step": 2510
    },
    {
      "entropy": 5.7252857208251955,
      "epoch": 2.7193077339102216,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046354498812591207,
      "loss": 5.4992,
      "mean_token_accuracy": 0.18573593199253083,
      "num_tokens": 5652633.0,
      "step": 2515
    },
    {
      "entropy": 5.730475807189942,
      "epoch": 2.7247160627366145,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046331058273448364,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1760507583618164,
      "num_tokens": 5664108.0,
      "step": 2520
    },
    {
      "entropy": 5.763299751281738,
      "epoch": 2.730124391563007,
      "grad_norm": 1.0,
      "learning_rate": 0.0004630754930219624,
      "loss": 5.4776,
      "mean_token_accuracy": 0.19018414318561555,
      "num_tokens": 5675701.0,
      "step": 2525
    },
    {
      "entropy": 5.663517713546753,
      "epoch": 2.7355327203893998,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004628397198426648,
      "loss": 5.4077,
      "mean_token_accuracy": 0.1896175116300583,
      "num_tokens": 5687292.0,
      "step": 2530
    },
    {
      "entropy": 5.732327222824097,
      "epoch": 2.740941049215792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046260326405339126,
      "loss": 5.4573,
      "mean_token_accuracy": 0.18407863527536392,
      "num_tokens": 5698905.0,
      "step": 2535
    },
    {
      "entropy": 5.6250590801239015,
      "epoch": 2.746349378042185,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004623661265134226,
      "loss": 5.3561,
      "mean_token_accuracy": 0.18782344758510588,
      "num_tokens": 5709678.0,
      "step": 2540
    },
    {
      "entropy": 5.7452105522155765,
      "epoch": 2.7517577068685775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004621283080845172,
      "loss": 5.4641,
      "mean_token_accuracy": 0.19136441349983216,
      "num_tokens": 5720460.0,
      "step": 2545
    },
    {
      "entropy": 5.672007465362549,
      "epoch": 2.7571660356949703,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046188980963090783,
      "loss": 5.4842,
      "mean_token_accuracy": 0.17894484996795654,
      "num_tokens": 5731701.0,
      "step": 2550
    },
    {
      "entropy": 5.738977909088135,
      "epoch": 2.7625743645213627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046165063201929825,
      "loss": 5.4918,
      "mean_token_accuracy": 0.18582795113325118,
      "num_tokens": 5741512.0,
      "step": 2555
    },
    {
      "entropy": 5.6584460735321045,
      "epoch": 2.7679826933477556,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046141077611886074,
      "loss": 5.4201,
      "mean_token_accuracy": 0.18902526795864105,
      "num_tokens": 5751882.0,
      "step": 2560
    },
    {
      "entropy": 5.708429002761841,
      "epoch": 2.773391022174148,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004611702428012319,
      "loss": 5.5221,
      "mean_token_accuracy": 0.1794414073228836,
      "num_tokens": 5761989.0,
      "step": 2565
    },
    {
      "entropy": 5.7255659103393555,
      "epoch": 2.778799351000541,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046092903294051057,
      "loss": 5.4281,
      "mean_token_accuracy": 0.1794588565826416,
      "num_tokens": 5772532.0,
      "step": 2570
    },
    {
      "entropy": 5.767435359954834,
      "epoch": 2.7842076798269337,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046068714741325385,
      "loss": 5.534,
      "mean_token_accuracy": 0.1859470635652542,
      "num_tokens": 5784152.0,
      "step": 2575
    },
    {
      "entropy": 5.7473695278167725,
      "epoch": 2.789616008653326,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004604445870984743,
      "loss": 5.5113,
      "mean_token_accuracy": 0.184004545211792,
      "num_tokens": 5793748.0,
      "step": 2580
    },
    {
      "entropy": 5.673414611816407,
      "epoch": 2.7950243374797186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004602013528776367,
      "loss": 5.448,
      "mean_token_accuracy": 0.18230961561203002,
      "num_tokens": 5804371.0,
      "step": 2585
    },
    {
      "entropy": 5.710828351974487,
      "epoch": 2.8004326663061114,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004599574456346548,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1878191575407982,
      "num_tokens": 5815470.0,
      "step": 2590
    },
    {
      "entropy": 5.758548545837402,
      "epoch": 2.8058409951325043,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045971286625588796,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1792083963751793,
      "num_tokens": 5826037.0,
      "step": 2595
    },
    {
      "entropy": 5.7658631801605225,
      "epoch": 2.8112493239588967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004594676156301383,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1877788320183754,
      "num_tokens": 5836468.0,
      "step": 2600
    },
    {
      "entropy": 5.608815717697143,
      "epoch": 2.816657652785289,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045922169464864725,
      "loss": 5.3191,
      "mean_token_accuracy": 0.19587965607643126,
      "num_tokens": 5846371.0,
      "step": 2605
    },
    {
      "entropy": 5.559684753417969,
      "epoch": 2.822065981611682,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000458975104205092,
      "loss": 5.4053,
      "mean_token_accuracy": 0.18844964355230331,
      "num_tokens": 5856890.0,
      "step": 2610
    },
    {
      "entropy": 5.739346027374268,
      "epoch": 2.827474310438075,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045872784519558304,
      "loss": 5.4574,
      "mean_token_accuracy": 0.19243008345365525,
      "num_tokens": 5868075.0,
      "step": 2615
    },
    {
      "entropy": 5.757848596572876,
      "epoch": 2.8328826392644673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004584799185186599,
      "loss": 5.5509,
      "mean_token_accuracy": 0.18274885341525077,
      "num_tokens": 5879579.0,
      "step": 2620
    },
    {
      "entropy": 5.6886180400848385,
      "epoch": 2.8382909680908597,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045823132507528903,
      "loss": 5.3955,
      "mean_token_accuracy": 0.1806962788105011,
      "num_tokens": 5890648.0,
      "step": 2625
    },
    {
      "entropy": 5.634205150604248,
      "epoch": 2.8436992969172525,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004579820657688593,
      "loss": 5.4475,
      "mean_token_accuracy": 0.18452479988336562,
      "num_tokens": 5902914.0,
      "step": 2630
    },
    {
      "entropy": 5.7609635353088375,
      "epoch": 2.8491076257436454,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004577321415051798,
      "loss": 5.469,
      "mean_token_accuracy": 0.17856979072093965,
      "num_tokens": 5915801.0,
      "step": 2635
    },
    {
      "entropy": 5.804005575180054,
      "epoch": 2.854515954570038,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00045748155319247564,
      "loss": 5.5522,
      "mean_token_accuracy": 0.17649677842855455,
      "num_tokens": 5928973.0,
      "step": 2640
    },
    {
      "entropy": 5.738051176071167,
      "epoch": 2.8599242833964302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004572303017413856,
      "loss": 5.5591,
      "mean_token_accuracy": 0.180062997341156,
      "num_tokens": 5939720.0,
      "step": 2645
    },
    {
      "entropy": 5.7691789150238035,
      "epoch": 2.865332612222823,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004569783880649578,
      "loss": 5.5188,
      "mean_token_accuracy": 0.18212350308895112,
      "num_tokens": 5951956.0,
      "step": 2650
    },
    {
      "entropy": 5.772093343734741,
      "epoch": 2.870740941049216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004567258130786474,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17411991208791733,
      "num_tokens": 5963735.0,
      "step": 2655
    },
    {
      "entropy": 5.793157625198364,
      "epoch": 2.8761492698756084,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045647257770031215,
      "loss": 5.4036,
      "mean_token_accuracy": 0.19065334349870683,
      "num_tokens": 5974567.0,
      "step": 2660
    },
    {
      "entropy": 5.699380826950073,
      "epoch": 2.8815575987020012,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045621868285021023,
      "loss": 5.5303,
      "mean_token_accuracy": 0.1865275114774704,
      "num_tokens": 5985800.0,
      "step": 2665
    },
    {
      "entropy": 5.6607505798339846,
      "epoch": 2.8869659275283936,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045596412945099615,
      "loss": 5.3719,
      "mean_token_accuracy": 0.1900218203663826,
      "num_tokens": 5997388.0,
      "step": 2670
    },
    {
      "entropy": 5.757824373245239,
      "epoch": 2.8923742563547865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045570891842771727,
      "loss": 5.4743,
      "mean_token_accuracy": 0.19120808243751525,
      "num_tokens": 6008739.0,
      "step": 2675
    },
    {
      "entropy": 5.663566303253174,
      "epoch": 2.897782585181179,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00045545305070781135,
      "loss": 5.3661,
      "mean_token_accuracy": 0.19172464907169343,
      "num_tokens": 6019490.0,
      "step": 2680
    },
    {
      "entropy": 5.591286516189575,
      "epoch": 2.903190914007572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045519652722110206,
      "loss": 5.4864,
      "mean_token_accuracy": 0.18697116523981094,
      "num_tokens": 6030712.0,
      "step": 2685
    },
    {
      "entropy": 5.721374559402466,
      "epoch": 2.908599242833964,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045493934889979655,
      "loss": 5.4036,
      "mean_token_accuracy": 0.1920872375369072,
      "num_tokens": 6042161.0,
      "step": 2690
    },
    {
      "entropy": 5.715954875946045,
      "epoch": 2.914007571660357,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00045468151667848116,
      "loss": 5.4172,
      "mean_token_accuracy": 0.18510787039995194,
      "num_tokens": 6054107.0,
      "step": 2695
    },
    {
      "entropy": 5.660524988174439,
      "epoch": 2.9194159004867495,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000454423031494119,
      "loss": 5.4979,
      "mean_token_accuracy": 0.18941813558340073,
      "num_tokens": 6065232.0,
      "step": 2700
    },
    {
      "entropy": 5.724817562103271,
      "epoch": 2.9248242293131423,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004541638942860458,
      "loss": 5.4539,
      "mean_token_accuracy": 0.18301481753587723,
      "num_tokens": 6076294.0,
      "step": 2705
    },
    {
      "entropy": 5.596493291854858,
      "epoch": 2.9302325581395348,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004539041059959667,
      "loss": 5.3626,
      "mean_token_accuracy": 0.19947786033153533,
      "num_tokens": 6087700.0,
      "step": 2710
    },
    {
      "entropy": 5.690376901626587,
      "epoch": 2.9356408869659276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045364366756795295,
      "loss": 5.4399,
      "mean_token_accuracy": 0.19008135050535202,
      "num_tokens": 6098561.0,
      "step": 2715
    },
    {
      "entropy": 5.772283267974854,
      "epoch": 2.94104921579232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045338257994843845,
      "loss": 5.4778,
      "mean_token_accuracy": 0.184920834004879,
      "num_tokens": 6109729.0,
      "step": 2720
    },
    {
      "entropy": 5.687829399108887,
      "epoch": 2.946457544618713,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004531208440862162,
      "loss": 5.4395,
      "mean_token_accuracy": 0.1881929948925972,
      "num_tokens": 6120468.0,
      "step": 2725
    },
    {
      "entropy": 5.630986547470092,
      "epoch": 2.9518658734451053,
      "grad_norm": 0.90625,
      "learning_rate": 0.00045285846093243485,
      "loss": 5.3743,
      "mean_token_accuracy": 0.1906595289707184,
      "num_tokens": 6132457.0,
      "step": 2730
    },
    {
      "entropy": 5.670713949203491,
      "epoch": 2.957274202271498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045259543144059557,
      "loss": 5.4346,
      "mean_token_accuracy": 0.18389843106269838,
      "num_tokens": 6142289.0,
      "step": 2735
    },
    {
      "entropy": 5.689014720916748,
      "epoch": 2.9626825310978906,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004523317565665479,
      "loss": 5.4383,
      "mean_token_accuracy": 0.18834447860717773,
      "num_tokens": 6153786.0,
      "step": 2740
    },
    {
      "entropy": 5.704989385604859,
      "epoch": 2.9680908599242835,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045206743726848716,
      "loss": 5.4117,
      "mean_token_accuracy": 0.18899297416210176,
      "num_tokens": 6163988.0,
      "step": 2745
    },
    {
      "entropy": 5.651515340805053,
      "epoch": 2.973499188750676,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004518024745069502,
      "loss": 5.5085,
      "mean_token_accuracy": 0.18532897382974625,
      "num_tokens": 6175859.0,
      "step": 2750
    },
    {
      "entropy": 5.909746503829956,
      "epoch": 2.9789075175770687,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004515368692448123,
      "loss": 5.685,
      "mean_token_accuracy": 0.17218046486377717,
      "num_tokens": 6188116.0,
      "step": 2755
    },
    {
      "entropy": 5.697541189193726,
      "epoch": 2.9843158464034616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045127062244728353,
      "loss": 5.4962,
      "mean_token_accuracy": 0.18963723182678222,
      "num_tokens": 6198606.0,
      "step": 2760
    },
    {
      "entropy": 5.817046451568603,
      "epoch": 2.989724175229854,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004510037350819055,
      "loss": 5.5319,
      "mean_token_accuracy": 0.18838882595300674,
      "num_tokens": 6210366.0,
      "step": 2765
    },
    {
      "entropy": 5.639420747756958,
      "epoch": 2.9951325040562464,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045073620811854744,
      "loss": 5.4186,
      "mean_token_accuracy": 0.18640679866075516,
      "num_tokens": 6221376.0,
      "step": 2770
    },
    {
      "entropy": 5.700975630018446,
      "epoch": 3.0,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00045046804252940294,
      "loss": 5.3876,
      "mean_token_accuracy": 0.18819021847512987,
      "num_tokens": 6230667.0,
      "step": 2775
    },
    {
      "entropy": 5.669461441040039,
      "epoch": 3.005408328826393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045019923928898644,
      "loss": 5.1027,
      "mean_token_accuracy": 0.20836202949285507,
      "num_tokens": 6241158.0,
      "step": 2780
    },
    {
      "entropy": 5.697578859329224,
      "epoch": 3.0108166576527853,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004499297993741294,
      "loss": 5.1899,
      "mean_token_accuracy": 0.19498405903577803,
      "num_tokens": 6253811.0,
      "step": 2785
    },
    {
      "entropy": 5.675133752822876,
      "epoch": 3.016224986479178,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004496597237639773,
      "loss": 5.1636,
      "mean_token_accuracy": 0.20492047667503357,
      "num_tokens": 6265066.0,
      "step": 2790
    },
    {
      "entropy": 5.604762315750122,
      "epoch": 3.0216333153055706,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044938901343998536,
      "loss": 5.1358,
      "mean_token_accuracy": 0.20102774798870088,
      "num_tokens": 6275889.0,
      "step": 2795
    },
    {
      "entropy": 5.651717805862427,
      "epoch": 3.0270416441319634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004491176693859156,
      "loss": 5.0966,
      "mean_token_accuracy": 0.20704333931207658,
      "num_tokens": 6287576.0,
      "step": 2800
    },
    {
      "entropy": 5.632756233215332,
      "epoch": 3.032449972958356,
      "grad_norm": 1.0,
      "learning_rate": 0.0004488456925878329,
      "loss": 5.167,
      "mean_token_accuracy": 0.20255689918994904,
      "num_tokens": 6299419.0,
      "step": 2805
    },
    {
      "entropy": 5.66122088432312,
      "epoch": 3.0378583017847487,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044857308403410163,
      "loss": 5.1603,
      "mean_token_accuracy": 0.20880026370286942,
      "num_tokens": 6310464.0,
      "step": 2810
    },
    {
      "entropy": 5.618158626556396,
      "epoch": 3.043266630611141,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044829984471538183,
      "loss": 5.1334,
      "mean_token_accuracy": 0.21153104454278945,
      "num_tokens": 6321442.0,
      "step": 2815
    },
    {
      "entropy": 5.615363645553589,
      "epoch": 3.048674959437534,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004480259756246258,
      "loss": 5.1818,
      "mean_token_accuracy": 0.20309753566980362,
      "num_tokens": 6332868.0,
      "step": 2820
    },
    {
      "entropy": 5.743928527832031,
      "epoch": 3.0540832882639264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044775147775707465,
      "loss": 5.2016,
      "mean_token_accuracy": 0.20300523489713668,
      "num_tokens": 6345232.0,
      "step": 2825
    },
    {
      "entropy": 5.6975706100463865,
      "epoch": 3.0594916170903192,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000447476352110254,
      "loss": 5.2291,
      "mean_token_accuracy": 0.19758180379867554,
      "num_tokens": 6356524.0,
      "step": 2830
    },
    {
      "entropy": 5.657529878616333,
      "epoch": 3.0648999459167117,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004472005996839712,
      "loss": 5.1794,
      "mean_token_accuracy": 0.2059034824371338,
      "num_tokens": 6367590.0,
      "step": 2835
    },
    {
      "entropy": 5.638584136962891,
      "epoch": 3.0703082747431045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004469242214803112,
      "loss": 5.1819,
      "mean_token_accuracy": 0.19632202982902527,
      "num_tokens": 6378211.0,
      "step": 2840
    },
    {
      "entropy": 5.6545031547546385,
      "epoch": 3.075716603569497,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004466472185036331,
      "loss": 5.1987,
      "mean_token_accuracy": 0.20859755277633668,
      "num_tokens": 6389595.0,
      "step": 2845
    },
    {
      "entropy": 5.593618106842041,
      "epoch": 3.08112493239589,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004463695917605663,
      "loss": 5.1988,
      "mean_token_accuracy": 0.20035351663827897,
      "num_tokens": 6401286.0,
      "step": 2850
    },
    {
      "entropy": 5.665650415420532,
      "epoch": 3.086533261222282,
      "grad_norm": 1.09375,
      "learning_rate": 0.000446091342260007,
      "loss": 5.1434,
      "mean_token_accuracy": 0.20971960723400115,
      "num_tokens": 6411946.0,
      "step": 2855
    },
    {
      "entropy": 5.611321878433228,
      "epoch": 3.091941590048675,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004458124710131145,
      "loss": 5.1017,
      "mean_token_accuracy": 0.20854540467262267,
      "num_tokens": 6422463.0,
      "step": 2860
    },
    {
      "entropy": 5.533938407897949,
      "epoch": 3.0973499188750675,
      "grad_norm": 1.125,
      "learning_rate": 0.00044553297903330766,
      "loss": 5.0985,
      "mean_token_accuracy": 0.20745838284492493,
      "num_tokens": 6433576.0,
      "step": 2865
    },
    {
      "entropy": 5.593739414215088,
      "epoch": 3.1027582477014604,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044525286733626085,
      "loss": 5.082,
      "mean_token_accuracy": 0.20933179259300233,
      "num_tokens": 6444058.0,
      "step": 2870
    },
    {
      "entropy": 5.673543787002563,
      "epoch": 3.1081665765278528,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004449721369399008,
      "loss": 5.257,
      "mean_token_accuracy": 0.20108112394809724,
      "num_tokens": 6453872.0,
      "step": 2875
    },
    {
      "entropy": 5.561177587509155,
      "epoch": 3.1135749053542456,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044469078886440227,
      "loss": 5.1647,
      "mean_token_accuracy": 0.1990603104233742,
      "num_tokens": 6465367.0,
      "step": 2880
    },
    {
      "entropy": 5.583830499649048,
      "epoch": 3.118983234180638,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044440882413218484,
      "loss": 5.0949,
      "mean_token_accuracy": 0.21778711527585984,
      "num_tokens": 6476655.0,
      "step": 2885
    },
    {
      "entropy": 5.598518037796021,
      "epoch": 3.124391563007031,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044412624376790905,
      "loss": 5.17,
      "mean_token_accuracy": 0.2041608527302742,
      "num_tokens": 6488348.0,
      "step": 2890
    },
    {
      "entropy": 5.596985960006714,
      "epoch": 3.1297998918334233,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004438430487984726,
      "loss": 5.1997,
      "mean_token_accuracy": 0.20207406282424928,
      "num_tokens": 6500127.0,
      "step": 2895
    },
    {
      "entropy": 5.524875497817993,
      "epoch": 3.135208220659816,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004435592402530066,
      "loss": 5.1431,
      "mean_token_accuracy": 0.20308458656072617,
      "num_tokens": 6512772.0,
      "step": 2900
    },
    {
      "entropy": 5.68407039642334,
      "epoch": 3.1406165494862086,
      "grad_norm": 1.0625,
      "learning_rate": 0.000443274819162872,
      "loss": 5.1835,
      "mean_token_accuracy": 0.20361116379499436,
      "num_tokens": 6523359.0,
      "step": 2905
    },
    {
      "entropy": 5.561126232147217,
      "epoch": 3.1460248783126015,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004429897865616557,
      "loss": 5.1549,
      "mean_token_accuracy": 0.20259772390127181,
      "num_tokens": 6533679.0,
      "step": 2910
    },
    {
      "entropy": 5.584951639175415,
      "epoch": 3.151433207138994,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004427041434851668,
      "loss": 5.1074,
      "mean_token_accuracy": 0.2015945389866829,
      "num_tokens": 6544694.0,
      "step": 2915
    },
    {
      "entropy": 5.5911280632019045,
      "epoch": 3.1568415359653867,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044241789097143293,
      "loss": 5.1315,
      "mean_token_accuracy": 0.20420949459075927,
      "num_tokens": 6555858.0,
      "step": 2920
    },
    {
      "entropy": 5.5801766872406,
      "epoch": 3.162249864791779,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044213103006069635,
      "loss": 5.1661,
      "mean_token_accuracy": 0.2034042462706566,
      "num_tokens": 6568214.0,
      "step": 2925
    },
    {
      "entropy": 5.491474771499634,
      "epoch": 3.167658193618172,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044184356179541035,
      "loss": 5.1379,
      "mean_token_accuracy": 0.2117103725671768,
      "num_tokens": 6579646.0,
      "step": 2930
    },
    {
      "entropy": 5.602434206008911,
      "epoch": 3.1730665224445644,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004415554872202352,
      "loss": 5.1007,
      "mean_token_accuracy": 0.2130330041050911,
      "num_tokens": 6590683.0,
      "step": 2935
    },
    {
      "entropy": 5.687664127349853,
      "epoch": 3.1784748512709573,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044126680738203466,
      "loss": 5.2317,
      "mean_token_accuracy": 0.20365397334098817,
      "num_tokens": 6601197.0,
      "step": 2940
    },
    {
      "entropy": 5.571243762969971,
      "epoch": 3.1838831800973497,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004409775233298718,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18942825198173524,
      "num_tokens": 6613564.0,
      "step": 2945
    },
    {
      "entropy": 5.628527975082397,
      "epoch": 3.1892915089237426,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004406876361150056,
      "loss": 5.1397,
      "mean_token_accuracy": 0.20057498961687087,
      "num_tokens": 6624696.0,
      "step": 2950
    },
    {
      "entropy": 5.557544326782226,
      "epoch": 3.194699837750135,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004403971467908869,
      "loss": 5.1524,
      "mean_token_accuracy": 0.21364459693431853,
      "num_tokens": 6636302.0,
      "step": 2955
    },
    {
      "entropy": 5.6587158203125,
      "epoch": 3.200108166576528,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004401060564131545,
      "loss": 5.1859,
      "mean_token_accuracy": 0.20433955639600754,
      "num_tokens": 6646517.0,
      "step": 2960
    },
    {
      "entropy": 5.515757083892822,
      "epoch": 3.2055164954029207,
      "grad_norm": 1.0,
      "learning_rate": 0.0004398143660396315,
      "loss": 5.1507,
      "mean_token_accuracy": 0.20412963926792144,
      "num_tokens": 6657768.0,
      "step": 2965
    },
    {
      "entropy": 5.505477380752564,
      "epoch": 3.210924824229313,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004395220767303215,
      "loss": 5.0794,
      "mean_token_accuracy": 0.2084907740354538,
      "num_tokens": 6669110.0,
      "step": 2970
    },
    {
      "entropy": 5.592361831665039,
      "epoch": 3.2163331530557056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004392291895474045,
      "loss": 5.1644,
      "mean_token_accuracy": 0.19438584744930268,
      "num_tokens": 6680925.0,
      "step": 2975
    },
    {
      "entropy": 5.6557297706604,
      "epoch": 3.2217414818820984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004389357055552331,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1981995016336441,
      "num_tokens": 6692923.0,
      "step": 2980
    },
    {
      "entropy": 5.6401595115661625,
      "epoch": 3.2271498107084913,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043864162582032897,
      "loss": 5.1275,
      "mean_token_accuracy": 0.20761849284172057,
      "num_tokens": 6702692.0,
      "step": 2985
    },
    {
      "entropy": 5.544705629348755,
      "epoch": 3.2325581395348837,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004383469514113784,
      "loss": 5.2131,
      "mean_token_accuracy": 0.20174529403448105,
      "num_tokens": 6713830.0,
      "step": 2990
    },
    {
      "entropy": 5.607528018951416,
      "epoch": 3.2379664683612766,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004380516833992289,
      "loss": 5.1986,
      "mean_token_accuracy": 0.19788535088300704,
      "num_tokens": 6725564.0,
      "step": 2995
    },
    {
      "entropy": 5.510459899902344,
      "epoch": 3.243374797187669,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000437755822856885,
      "loss": 5.1444,
      "mean_token_accuracy": 0.203812113404274,
      "num_tokens": 6736563.0,
      "step": 3000
    },
    {
      "epoch": 3.243374797187669,
      "eval_entropy": 5.434760303930803,
      "eval_loss": 5.901750087738037,
      "eval_mean_token_accuracy": 0.17458110260353846,
      "eval_num_tokens": 6736563.0,
      "eval_runtime": 1.9474,
      "eval_samples_per_second": 1805.0,
      "eval_steps_per_second": 225.946,
      "step": 3000
    },
    {
      "entropy": 5.544770526885986,
      "epoch": 3.248783126014062,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004374593708595049,
      "loss": 5.1344,
      "mean_token_accuracy": 0.20361437201499938,
      "num_tokens": 6745845.0,
      "step": 3005
    },
    {
      "entropy": 5.519393014907837,
      "epoch": 3.2541914548404542,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00043716232848439555,
      "loss": 5.075,
      "mean_token_accuracy": 0.2104331910610199,
      "num_tokens": 6757450.0,
      "step": 3010
    },
    {
      "entropy": 5.664768648147583,
      "epoch": 3.259599783666847,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00043686469681100985,
      "loss": 5.24,
      "mean_token_accuracy": 0.20128876864910125,
      "num_tokens": 6768542.0,
      "step": 3015
    },
    {
      "entropy": 5.538003301620483,
      "epoch": 3.2650081124932395,
      "grad_norm": 0.953125,
      "learning_rate": 0.00043656647692094186,
      "loss": 5.1511,
      "mean_token_accuracy": 0.21601292192935945,
      "num_tokens": 6779873.0,
      "step": 3020
    },
    {
      "entropy": 5.572194576263428,
      "epoch": 3.2704164413196324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004362676698979233,
      "loss": 5.2112,
      "mean_token_accuracy": 0.2038590893149376,
      "num_tokens": 6790898.0,
      "step": 3025
    },
    {
      "entropy": 5.649399328231811,
      "epoch": 3.275824770146025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043596827682781974,
      "loss": 5.169,
      "mean_token_accuracy": 0.2034141629934311,
      "num_tokens": 6802175.0,
      "step": 3030
    },
    {
      "entropy": 5.6116852283477785,
      "epoch": 3.2812330989724177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004356682987986262,
      "loss": 5.2851,
      "mean_token_accuracy": 0.19906653463840485,
      "num_tokens": 6813675.0,
      "step": 3035
    },
    {
      "entropy": 5.504955911636353,
      "epoch": 3.28664142779881,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004353677369004635,
      "loss": 5.1625,
      "mean_token_accuracy": 0.20416872650384904,
      "num_tokens": 6825020.0,
      "step": 3040
    },
    {
      "entropy": 5.5862291812896725,
      "epoch": 3.292049756625203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043506659222557426,
      "loss": 5.0572,
      "mean_token_accuracy": 0.21936341524124145,
      "num_tokens": 6834954.0,
      "step": 3045
    },
    {
      "entropy": 5.622218465805053,
      "epoch": 3.2974580854515954,
      "grad_norm": 1.015625,
      "learning_rate": 0.000434764865868319,
      "loss": 5.2878,
      "mean_token_accuracy": 0.19446102380752564,
      "num_tokens": 6847958.0,
      "step": 3050
    },
    {
      "entropy": 5.5780336380004885,
      "epoch": 3.302866414277988,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000434462558925172,
      "loss": 5.0982,
      "mean_token_accuracy": 0.21777901351451873,
      "num_tokens": 6858782.0,
      "step": 3055
    },
    {
      "entropy": 5.448307895660401,
      "epoch": 3.3082747431043806,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043415967249471734,
      "loss": 5.1252,
      "mean_token_accuracy": 0.20691056698560714,
      "num_tokens": 6869807.0,
      "step": 3060
    },
    {
      "entropy": 5.640656137466431,
      "epoch": 3.3136830719307735,
      "grad_norm": 1.171875,
      "learning_rate": 0.000433856207677645,
      "loss": 5.1618,
      "mean_token_accuracy": 0.20938359051942826,
      "num_tokens": 6880642.0,
      "step": 3065
    },
    {
      "entropy": 5.636711311340332,
      "epoch": 3.319091400757166,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000433552165576747,
      "loss": 5.271,
      "mean_token_accuracy": 0.19023908376693727,
      "num_tokens": 6891677.0,
      "step": 3070
    },
    {
      "entropy": 5.603629112243652,
      "epoch": 3.3244997295835588,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043324754729691275,
      "loss": 5.1504,
      "mean_token_accuracy": 0.20489296317100525,
      "num_tokens": 6903383.0,
      "step": 3075
    },
    {
      "entropy": 5.6365159034729,
      "epoch": 3.329908058409951,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043294235394512594,
      "loss": 5.228,
      "mean_token_accuracy": 0.2118686094880104,
      "num_tokens": 6915874.0,
      "step": 3080
    },
    {
      "entropy": 5.532635498046875,
      "epoch": 3.335316387236344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004326365866304599,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20689944326877593,
      "num_tokens": 6926356.0,
      "step": 3085
    },
    {
      "entropy": 5.597835874557495,
      "epoch": 3.3407247160627365,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004323302464640738,
      "loss": 5.1755,
      "mean_token_accuracy": 0.20826709270477295,
      "num_tokens": 6937678.0,
      "step": 3090
    },
    {
      "entropy": 5.582600355148315,
      "epoch": 3.3461330448891293,
      "grad_norm": 1.0,
      "learning_rate": 0.00043202333455920843,
      "loss": 5.172,
      "mean_token_accuracy": 0.20327077507972718,
      "num_tokens": 6949447.0,
      "step": 3095
    },
    {
      "entropy": 5.571978902816772,
      "epoch": 3.3515413737155217,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004317158520311824,
      "loss": 5.0976,
      "mean_token_accuracy": 0.20037805438041686,
      "num_tokens": 6960347.0,
      "step": 3100
    },
    {
      "entropy": 5.48020601272583,
      "epoch": 3.3569497025419146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004314077999973879,
      "loss": 5.1147,
      "mean_token_accuracy": 0.20114873349666595,
      "num_tokens": 6971353.0,
      "step": 3105
    },
    {
      "entropy": 5.5931471824646,
      "epoch": 3.362358031368307,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004310991795772869,
      "loss": 5.2397,
      "mean_token_accuracy": 0.19606329500675201,
      "num_tokens": 6982519.0,
      "step": 3110
    },
    {
      "entropy": 5.495945882797241,
      "epoch": 3.3677663601947,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004307899918924065,
      "loss": 5.0915,
      "mean_token_accuracy": 0.2138282909989357,
      "num_tokens": 6994021.0,
      "step": 3115
    },
    {
      "entropy": 5.497316789627075,
      "epoch": 3.3731746890210923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004304802380663356,
      "loss": 5.1749,
      "mean_token_accuracy": 0.21007626503705978,
      "num_tokens": 7005505.0,
      "step": 3120
    },
    {
      "entropy": 5.575474071502685,
      "epoch": 3.378583017847485,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043016991922472015,
      "loss": 5.2136,
      "mean_token_accuracy": 0.20582512766122818,
      "num_tokens": 7016123.0,
      "step": 3125
    },
    {
      "entropy": 5.647998523712158,
      "epoch": 3.383991346673878,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042985903649525977,
      "loss": 5.1645,
      "mean_token_accuracy": 0.20766933411359786,
      "num_tokens": 7027083.0,
      "step": 3130
    },
    {
      "entropy": 5.58907036781311,
      "epoch": 3.3893996755002704,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042954759100770275,
      "loss": 5.2524,
      "mean_token_accuracy": 0.19670134782791138,
      "num_tokens": 7037892.0,
      "step": 3135
    },
    {
      "entropy": 5.541421937942505,
      "epoch": 3.394808004326663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042923558389384303,
      "loss": 5.238,
      "mean_token_accuracy": 0.19885995984077454,
      "num_tokens": 7048996.0,
      "step": 3140
    },
    {
      "entropy": 5.656047105789185,
      "epoch": 3.4002163331530557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004289230162875149,
      "loss": 5.2447,
      "mean_token_accuracy": 0.19565346091985703,
      "num_tokens": 7060884.0,
      "step": 3145
    },
    {
      "entropy": 5.602205228805542,
      "epoch": 3.4056246619794486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042860988932458987,
      "loss": 5.2461,
      "mean_token_accuracy": 0.20275676846504212,
      "num_tokens": 7072550.0,
      "step": 3150
    },
    {
      "entropy": 5.580437660217285,
      "epoch": 3.411032990805841,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004282962041429721,
      "loss": 5.1573,
      "mean_token_accuracy": 0.21325522810220718,
      "num_tokens": 7085325.0,
      "step": 3155
    },
    {
      "entropy": 5.496772336959839,
      "epoch": 3.4164413196322334,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042798196188259427,
      "loss": 5.0621,
      "mean_token_accuracy": 0.2055182009935379,
      "num_tokens": 7095691.0,
      "step": 3160
    },
    {
      "entropy": 5.43573112487793,
      "epoch": 3.4218496484586263,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004276671636854133,
      "loss": 5.0232,
      "mean_token_accuracy": 0.21906284987926483,
      "num_tokens": 7106552.0,
      "step": 3165
    },
    {
      "entropy": 5.482515716552735,
      "epoch": 3.427257977285019,
      "grad_norm": 0.9375,
      "learning_rate": 0.00042735181069540677,
      "loss": 5.1675,
      "mean_token_accuracy": 0.2021942988038063,
      "num_tokens": 7119708.0,
      "step": 3170
    },
    {
      "entropy": 5.555360269546509,
      "epoch": 3.4326663061114115,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004270359040585681,
      "loss": 5.1658,
      "mean_token_accuracy": 0.19937751442193985,
      "num_tokens": 7130663.0,
      "step": 3175
    },
    {
      "entropy": 5.445105648040771,
      "epoch": 3.4380746349378044,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042671944492290277,
      "loss": 5.0875,
      "mean_token_accuracy": 0.20021066069602966,
      "num_tokens": 7141513.0,
      "step": 3180
    },
    {
      "entropy": 5.526322841644287,
      "epoch": 3.443482963764197,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004264024344384239,
      "loss": 5.1649,
      "mean_token_accuracy": 0.2033783659338951,
      "num_tokens": 7153387.0,
      "step": 3185
    },
    {
      "entropy": 5.565660238265991,
      "epoch": 3.4488912925905897,
      "grad_norm": 0.90625,
      "learning_rate": 0.00042608487375714847,
      "loss": 5.1457,
      "mean_token_accuracy": 0.20955105274915695,
      "num_tokens": 7164905.0,
      "step": 3190
    },
    {
      "entropy": 5.566634798049927,
      "epoch": 3.454299621416982,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042576676403309265,
      "loss": 5.1864,
      "mean_token_accuracy": 0.20546145886182784,
      "num_tokens": 7175636.0,
      "step": 3195
    },
    {
      "entropy": 5.574922800064087,
      "epoch": 3.459707950243375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004254481064222679,
      "loss": 5.1913,
      "mean_token_accuracy": 0.19795469790697098,
      "num_tokens": 7185903.0,
      "step": 3200
    },
    {
      "entropy": 5.474887990951538,
      "epoch": 3.4651162790697674,
      "grad_norm": 0.96875,
      "learning_rate": 0.00042512890208267677,
      "loss": 5.1043,
      "mean_token_accuracy": 0.20120492428541184,
      "num_tokens": 7198257.0,
      "step": 3205
    },
    {
      "entropy": 5.58897385597229,
      "epoch": 3.4705246078961602,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042480915217430853,
      "loss": 5.0898,
      "mean_token_accuracy": 0.2133757948875427,
      "num_tokens": 7208839.0,
      "step": 3210
    },
    {
      "entropy": 5.541980934143067,
      "epoch": 3.4759329367225527,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042448885785913507,
      "loss": 5.208,
      "mean_token_accuracy": 0.19899676293134688,
      "num_tokens": 7221078.0,
      "step": 3215
    },
    {
      "entropy": 5.567216253280639,
      "epoch": 3.4813412655489455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042416802030110677,
      "loss": 5.1178,
      "mean_token_accuracy": 0.20688695162534715,
      "num_tokens": 7230287.0,
      "step": 3220
    },
    {
      "entropy": 5.471492338180542,
      "epoch": 3.486749594375338,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042384664066614787,
      "loss": 5.1987,
      "mean_token_accuracy": 0.2023004412651062,
      "num_tokens": 7242253.0,
      "step": 3225
    },
    {
      "entropy": 5.5609088897705075,
      "epoch": 3.492157923201731,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004235247201221528,
      "loss": 5.1655,
      "mean_token_accuracy": 0.21034539937973024,
      "num_tokens": 7254098.0,
      "step": 3230
    },
    {
      "entropy": 5.456197452545166,
      "epoch": 3.497566252028123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004232022598389816,
      "loss": 5.0707,
      "mean_token_accuracy": 0.20798765271902084,
      "num_tokens": 7264153.0,
      "step": 3235
    },
    {
      "entropy": 5.482938623428344,
      "epoch": 3.502974580854516,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004228792609884557,
      "loss": 5.1784,
      "mean_token_accuracy": 0.20120178908109665,
      "num_tokens": 7276431.0,
      "step": 3240
    },
    {
      "entropy": 5.544559383392334,
      "epoch": 3.5083829096809085,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00042255572474435364,
      "loss": 5.1211,
      "mean_token_accuracy": 0.20562490224838256,
      "num_tokens": 7287882.0,
      "step": 3245
    },
    {
      "entropy": 5.550385999679565,
      "epoch": 3.5137912385073014,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004222316522824069,
      "loss": 5.2576,
      "mean_token_accuracy": 0.19115870147943498,
      "num_tokens": 7298846.0,
      "step": 3250
    },
    {
      "entropy": 5.527417945861816,
      "epoch": 3.5191995673336938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004219070447802955,
      "loss": 5.1274,
      "mean_token_accuracy": 0.20272285342216492,
      "num_tokens": 7310372.0,
      "step": 3255
    },
    {
      "entropy": 5.612939310073853,
      "epoch": 3.5246078961600866,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042158190341764384,
      "loss": 5.2346,
      "mean_token_accuracy": 0.20130084156990052,
      "num_tokens": 7322200.0,
      "step": 3260
    },
    {
      "entropy": 5.557623338699341,
      "epoch": 3.530016224986479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042125622937601645,
      "loss": 5.1011,
      "mean_token_accuracy": 0.21032217890024185,
      "num_tokens": 7331824.0,
      "step": 3265
    },
    {
      "entropy": 5.536936473846436,
      "epoch": 3.535424553812872,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004209300238389135,
      "loss": 5.0972,
      "mean_token_accuracy": 0.20524015426635742,
      "num_tokens": 7341522.0,
      "step": 3270
    },
    {
      "entropy": 5.517943239212036,
      "epoch": 3.5408328826392643,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004206032879917667,
      "loss": 5.1822,
      "mean_token_accuracy": 0.2097307413816452,
      "num_tokens": 7353312.0,
      "step": 3275
    },
    {
      "entropy": 5.55204963684082,
      "epoch": 3.546241211465657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004202760230219348,
      "loss": 5.1901,
      "mean_token_accuracy": 0.2106068253517151,
      "num_tokens": 7363985.0,
      "step": 3280
    },
    {
      "entropy": 5.5757972240448,
      "epoch": 3.5516495402920496,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041994823011869955,
      "loss": 5.1561,
      "mean_token_accuracy": 0.20840810239315033,
      "num_tokens": 7373872.0,
      "step": 3285
    },
    {
      "entropy": 5.562177944183349,
      "epoch": 3.5570578691184425,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00041961991047326116,
      "loss": 5.1795,
      "mean_token_accuracy": 0.21022027730941772,
      "num_tokens": 7386759.0,
      "step": 3290
    },
    {
      "entropy": 5.50989785194397,
      "epoch": 3.5624661979448353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004192910652787338,
      "loss": 5.1279,
      "mean_token_accuracy": 0.21205971986055375,
      "num_tokens": 7398001.0,
      "step": 3295
    },
    {
      "entropy": 5.604728507995605,
      "epoch": 3.5678745267712277,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004189616957301419,
      "loss": 5.1826,
      "mean_token_accuracy": 0.20173380076885222,
      "num_tokens": 7409159.0,
      "step": 3300
    },
    {
      "entropy": 5.458903837203979,
      "epoch": 3.57328285559762,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004186318030244149,
      "loss": 5.1123,
      "mean_token_accuracy": 0.2085228979587555,
      "num_tokens": 7421028.0,
      "step": 3305
    },
    {
      "entropy": 5.482832574844361,
      "epoch": 3.578691184424013,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041830138836038383,
      "loss": 5.219,
      "mean_token_accuracy": 0.20427736788988113,
      "num_tokens": 7432321.0,
      "step": 3310
    },
    {
      "entropy": 5.63370909690857,
      "epoch": 3.584099513250406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004179704529387763,
      "loss": 5.2171,
      "mean_token_accuracy": 0.21171905100345612,
      "num_tokens": 7443607.0,
      "step": 3315
    },
    {
      "entropy": 5.55096435546875,
      "epoch": 3.5895078420767983,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004176389979622125,
      "loss": 5.1023,
      "mean_token_accuracy": 0.20815017223358154,
      "num_tokens": 7455222.0,
      "step": 3320
    },
    {
      "entropy": 5.566087532043457,
      "epoch": 3.5949161709031907,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004173070246352003,
      "loss": 5.2557,
      "mean_token_accuracy": 0.20288895070552826,
      "num_tokens": 7468511.0,
      "step": 3325
    },
    {
      "entropy": 5.5558840274810795,
      "epoch": 3.6003244997295836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041697453416413184,
      "loss": 5.2304,
      "mean_token_accuracy": 0.20506677627563477,
      "num_tokens": 7479810.0,
      "step": 3330
    },
    {
      "entropy": 5.484669351577759,
      "epoch": 3.6057328285559764,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000416641527757278,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20741904079914092,
      "num_tokens": 7491405.0,
      "step": 3335
    },
    {
      "entropy": 5.5714035987854,
      "epoch": 3.611141157382369,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041630800662478514,
      "loss": 5.2042,
      "mean_token_accuracy": 0.19627125561237335,
      "num_tokens": 7501820.0,
      "step": 3340
    },
    {
      "entropy": 5.564830732345581,
      "epoch": 3.6165494862087613,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041597397197866957,
      "loss": 5.1477,
      "mean_token_accuracy": 0.20144274979829788,
      "num_tokens": 7512165.0,
      "step": 3345
    },
    {
      "entropy": 5.4781747341156,
      "epoch": 3.621957815035154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004156394250328141,
      "loss": 5.1396,
      "mean_token_accuracy": 0.2055287703871727,
      "num_tokens": 7523755.0,
      "step": 3350
    },
    {
      "entropy": 5.465416097640992,
      "epoch": 3.627366143861547,
      "grad_norm": 0.984375,
      "learning_rate": 0.000415304367002963,
      "loss": 5.1444,
      "mean_token_accuracy": 0.2058341920375824,
      "num_tokens": 7535674.0,
      "step": 3355
    },
    {
      "entropy": 5.543293809890747,
      "epoch": 3.6327744726879394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041496879910671796,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20769851803779601,
      "num_tokens": 7546885.0,
      "step": 3360
    },
    {
      "entropy": 5.510053586959839,
      "epoch": 3.638182801514332,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004146327225635336,
      "loss": 5.1096,
      "mean_token_accuracy": 0.21097142100334168,
      "num_tokens": 7557718.0,
      "step": 3365
    },
    {
      "entropy": 5.471784353256226,
      "epoch": 3.6435911303407247,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004142961385947127,
      "loss": 5.0855,
      "mean_token_accuracy": 0.20106115192174911,
      "num_tokens": 7569472.0,
      "step": 3370
    },
    {
      "entropy": 5.586308908462525,
      "epoch": 3.6489994591671175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041395904842340215,
      "loss": 5.1288,
      "mean_token_accuracy": 0.20158884376287461,
      "num_tokens": 7579954.0,
      "step": 3375
    },
    {
      "entropy": 5.441673326492309,
      "epoch": 3.65440778799351,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004136214532745887,
      "loss": 5.0713,
      "mean_token_accuracy": 0.21283359080553055,
      "num_tokens": 7590717.0,
      "step": 3380
    },
    {
      "entropy": 5.562170839309692,
      "epoch": 3.6598161168199024,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004132833543750936,
      "loss": 5.3881,
      "mean_token_accuracy": 0.19054288864135743,
      "num_tokens": 7602802.0,
      "step": 3385
    },
    {
      "entropy": 5.70370078086853,
      "epoch": 3.6652244456462952,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00041294475295356917,
      "loss": 5.3212,
      "mean_token_accuracy": 0.19516013115644454,
      "num_tokens": 7614944.0,
      "step": 3390
    },
    {
      "entropy": 5.549120950698852,
      "epoch": 3.670632774472688,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004126056502404937,
      "loss": 5.2095,
      "mean_token_accuracy": 0.19887917637825012,
      "num_tokens": 7626449.0,
      "step": 3395
    },
    {
      "entropy": 5.4932983875274655,
      "epoch": 3.6760411032990805,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004122660474681672,
      "loss": 5.0728,
      "mean_token_accuracy": 0.2106201693415642,
      "num_tokens": 7637878.0,
      "step": 3400
    },
    {
      "entropy": 5.584516954421997,
      "epoch": 3.6814494321254734,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00041192594587070706,
      "loss": 5.3003,
      "mean_token_accuracy": 0.1944027379155159,
      "num_tokens": 7649428.0,
      "step": 3405
    },
    {
      "entropy": 5.644841480255127,
      "epoch": 3.686857760951866,
      "grad_norm": 1.0,
      "learning_rate": 0.00041158534668404325,
      "loss": 5.2193,
      "mean_token_accuracy": 0.19747732132673262,
      "num_tokens": 7660527.0,
      "step": 3410
    },
    {
      "entropy": 5.472689437866211,
      "epoch": 3.6922660897782587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041124425114591386,
      "loss": 5.1513,
      "mean_token_accuracy": 0.2066401407122612,
      "num_tokens": 7671249.0,
      "step": 3415
    },
    {
      "entropy": 5.572272109985351,
      "epoch": 3.697674418604651,
      "grad_norm": 1.125,
      "learning_rate": 0.00041090266049586107,
      "loss": 5.1494,
      "mean_token_accuracy": 0.2017663300037384,
      "num_tokens": 7681824.0,
      "step": 3420
    },
    {
      "entropy": 5.516412258148193,
      "epoch": 3.703082747431044,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004105605759752258,
      "loss": 5.1981,
      "mean_token_accuracy": 0.2008174940943718,
      "num_tokens": 7691916.0,
      "step": 3425
    },
    {
      "entropy": 5.562558031082153,
      "epoch": 3.7084910762574363,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00041021799882714436,
      "loss": 5.233,
      "mean_token_accuracy": 0.19865853488445281,
      "num_tokens": 7703355.0,
      "step": 3430
    },
    {
      "entropy": 5.478209114074707,
      "epoch": 3.713899405083829,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004098749302965426,
      "loss": 5.1716,
      "mean_token_accuracy": 0.20574578046798705,
      "num_tokens": 7714731.0,
      "step": 3435
    },
    {
      "entropy": 5.521606683731079,
      "epoch": 3.7193077339102216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004095313716301324,
      "loss": 5.127,
      "mean_token_accuracy": 0.21381030976772308,
      "num_tokens": 7725696.0,
      "step": 3440
    },
    {
      "entropy": 5.527922010421753,
      "epoch": 3.7247160627366145,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004091873240764069,
      "loss": 5.1366,
      "mean_token_accuracy": 0.20006966292858125,
      "num_tokens": 7737121.0,
      "step": 3445
    },
    {
      "entropy": 5.584472513198852,
      "epoch": 3.730124391563007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040884278888563566,
      "loss": 5.2142,
      "mean_token_accuracy": 0.2017587050795555,
      "num_tokens": 7749088.0,
      "step": 3450
    },
    {
      "entropy": 5.610855054855347,
      "epoch": 3.7355327203893998,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040849776730986014,
      "loss": 5.305,
      "mean_token_accuracy": 0.20229094922542573,
      "num_tokens": 7760676.0,
      "step": 3455
    },
    {
      "entropy": 5.592826652526855,
      "epoch": 3.740941049215792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004081522606028899,
      "loss": 5.2329,
      "mean_token_accuracy": 0.19813970774412154,
      "num_tokens": 7772560.0,
      "step": 3460
    },
    {
      "entropy": 5.548866558074951,
      "epoch": 3.746349378042185,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040780627002029674,
      "loss": 5.1384,
      "mean_token_accuracy": 0.2101344957947731,
      "num_tokens": 7782206.0,
      "step": 3465
    },
    {
      "entropy": 5.514203882217407,
      "epoch": 3.7517577068685775,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004074597968194116,
      "loss": 5.2579,
      "mean_token_accuracy": 0.19164653569459916,
      "num_tokens": 7793862.0,
      "step": 3470
    },
    {
      "entropy": 5.579835033416748,
      "epoch": 3.7571660356949703,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004071128422593186,
      "loss": 5.2788,
      "mean_token_accuracy": 0.19755514860153198,
      "num_tokens": 7805291.0,
      "step": 3475
    },
    {
      "entropy": 5.653624200820923,
      "epoch": 3.7625743645213627,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004067654076008515,
      "loss": 5.2474,
      "mean_token_accuracy": 0.20341532081365585,
      "num_tokens": 7816141.0,
      "step": 3480
    },
    {
      "entropy": 5.508863353729248,
      "epoch": 3.7679826933477556,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040641749410658877,
      "loss": 5.2185,
      "mean_token_accuracy": 0.19252879321575164,
      "num_tokens": 7826209.0,
      "step": 3485
    },
    {
      "entropy": 5.569253540039062,
      "epoch": 3.773391022174148,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004060691030408487,
      "loss": 5.1698,
      "mean_token_accuracy": 0.1992380440235138,
      "num_tokens": 7836734.0,
      "step": 3490
    },
    {
      "entropy": 5.576125907897949,
      "epoch": 3.778799351000541,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040572023566968536,
      "loss": 5.2352,
      "mean_token_accuracy": 0.1927531987428665,
      "num_tokens": 7848283.0,
      "step": 3495
    },
    {
      "entropy": 5.559657907485962,
      "epoch": 3.7842076798269337,
      "grad_norm": 0.90625,
      "learning_rate": 0.00040537089326088345,
      "loss": 5.1341,
      "mean_token_accuracy": 0.20429346114397048,
      "num_tokens": 7861226.0,
      "step": 3500
    },
    {
      "entropy": 5.43853497505188,
      "epoch": 3.789616008653326,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004050210770839541,
      "loss": 5.1125,
      "mean_token_accuracy": 0.21103197038173677,
      "num_tokens": 7872426.0,
      "step": 3505
    },
    {
      "entropy": 5.569460344314575,
      "epoch": 3.7950243374797186,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004046707884101303,
      "loss": 5.2075,
      "mean_token_accuracy": 0.20043655782938002,
      "num_tokens": 7883884.0,
      "step": 3510
    },
    {
      "entropy": 5.478324460983276,
      "epoch": 3.8004326663061114,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040432002851236164,
      "loss": 5.1194,
      "mean_token_accuracy": 0.2073998510837555,
      "num_tokens": 7895415.0,
      "step": 3515
    },
    {
      "entropy": 5.505063533782959,
      "epoch": 3.8058409951325043,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004039687986653106,
      "loss": 5.2168,
      "mean_token_accuracy": 0.20027986615896226,
      "num_tokens": 7907444.0,
      "step": 3520
    },
    {
      "entropy": 5.611638498306275,
      "epoch": 3.8112493239588967,
      "grad_norm": 0.921875,
      "learning_rate": 0.00040361710014534714,
      "loss": 5.2011,
      "mean_token_accuracy": 0.2003824755549431,
      "num_tokens": 7918811.0,
      "step": 3525
    },
    {
      "entropy": 5.5834228038787845,
      "epoch": 3.816657652785289,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004032649342305445,
      "loss": 5.2675,
      "mean_token_accuracy": 0.19360382854938507,
      "num_tokens": 7930657.0,
      "step": 3530
    },
    {
      "entropy": 5.40786919593811,
      "epoch": 3.822065981611682,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004029123022006745,
      "loss": 5.0926,
      "mean_token_accuracy": 0.19876976013183595,
      "num_tokens": 7942469.0,
      "step": 3535
    },
    {
      "entropy": 5.616995429992675,
      "epoch": 3.827474310438075,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004025592053372027,
      "loss": 5.2684,
      "mean_token_accuracy": 0.20128581672906876,
      "num_tokens": 7954034.0,
      "step": 3540
    },
    {
      "entropy": 5.550896263122558,
      "epoch": 3.8328826392644673,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004022056449232839,
      "loss": 5.1775,
      "mean_token_accuracy": 0.20249273926019667,
      "num_tokens": 7966079.0,
      "step": 3545
    },
    {
      "entropy": 5.393668794631958,
      "epoch": 3.8382909680908597,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004018516222437575,
      "loss": 5.133,
      "mean_token_accuracy": 0.21324997246265412,
      "num_tokens": 7977785.0,
      "step": 3550
    },
    {
      "entropy": 5.617745780944825,
      "epoch": 3.8436992969172525,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004014971385851428,
      "loss": 5.293,
      "mean_token_accuracy": 0.20187714248895644,
      "num_tokens": 7988874.0,
      "step": 3555
    },
    {
      "entropy": 5.45885648727417,
      "epoch": 3.8491076257436454,
      "grad_norm": 0.984375,
      "learning_rate": 0.000401142195235634,
      "loss": 5.0776,
      "mean_token_accuracy": 0.21275651901960374,
      "num_tokens": 7999438.0,
      "step": 3560
    },
    {
      "entropy": 5.569678926467896,
      "epoch": 3.854515954570038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004007867934850963,
      "loss": 5.1761,
      "mean_token_accuracy": 0.20520496219396592,
      "num_tokens": 8009658.0,
      "step": 3565
    },
    {
      "entropy": 5.5759193897247314,
      "epoch": 3.8599242833964302,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004004309346250603,
      "loss": 5.1392,
      "mean_token_accuracy": 0.1986655205488205,
      "num_tokens": 8020988.0,
      "step": 3570
    },
    {
      "entropy": 5.4343029975891115,
      "epoch": 3.865332612222823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004000746199487181,
      "loss": 5.0416,
      "mean_token_accuracy": 0.20885435938835145,
      "num_tokens": 8031329.0,
      "step": 3575
    },
    {
      "entropy": 5.450254631042481,
      "epoch": 3.870740941049216,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000399717850750918,
      "loss": 5.1002,
      "mean_token_accuracy": 0.20467742681503295,
      "num_tokens": 8041693.0,
      "step": 3580
    },
    {
      "entropy": 5.545977783203125,
      "epoch": 3.8761492698756084,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00039936062832816,
      "loss": 5.1827,
      "mean_token_accuracy": 0.20977984368801117,
      "num_tokens": 8053123.0,
      "step": 3585
    },
    {
      "entropy": 5.601700687408448,
      "epoch": 3.8815575987020012,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003990029539785913,
      "loss": 5.1613,
      "mean_token_accuracy": 0.2057858407497406,
      "num_tokens": 8064270.0,
      "step": 3590
    },
    {
      "entropy": 5.519832134246826,
      "epoch": 3.8869659275283936,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039864482900200135,
      "loss": 5.1167,
      "mean_token_accuracy": 0.20741852819919587,
      "num_tokens": 8074631.0,
      "step": 3595
    },
    {
      "entropy": 5.465943431854248,
      "epoch": 3.8923742563547865,
      "grad_norm": 1.0,
      "learning_rate": 0.0003982862546998169,
      "loss": 5.1929,
      "mean_token_accuracy": 0.20347485244274138,
      "num_tokens": 8086794.0,
      "step": 3600
    },
    {
      "entropy": 5.532767820358276,
      "epoch": 3.897782585181179,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003979272323750981,
      "loss": 5.1889,
      "mean_token_accuracy": 0.20599882453680038,
      "num_tokens": 8097595.0,
      "step": 3605
    },
    {
      "entropy": 5.527312898635865,
      "epoch": 3.903190914007572,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003975677633325327,
      "loss": 5.1305,
      "mean_token_accuracy": 0.20975493639707565,
      "num_tokens": 8107174.0,
      "step": 3610
    },
    {
      "entropy": 5.525962686538696,
      "epoch": 3.908599242833964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003972078488784321,
      "loss": 5.1754,
      "mean_token_accuracy": 0.20812157541513443,
      "num_tokens": 8118065.0,
      "step": 3615
    },
    {
      "entropy": 5.480575704574585,
      "epoch": 3.914007571660357,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003968474903207263,
      "loss": 5.1429,
      "mean_token_accuracy": 0.19851676672697066,
      "num_tokens": 8130604.0,
      "step": 3620
    },
    {
      "entropy": 5.554414987564087,
      "epoch": 3.9194159004867495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003964866889689592,
      "loss": 5.2174,
      "mean_token_accuracy": 0.2034759357571602,
      "num_tokens": 8141808.0,
      "step": 3625
    },
    {
      "entropy": 5.58361701965332,
      "epoch": 3.9248242293131423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003961254461342834,
      "loss": 5.184,
      "mean_token_accuracy": 0.19987331777811052,
      "num_tokens": 8152101.0,
      "step": 3630
    },
    {
      "entropy": 5.565663814544678,
      "epoch": 3.9302325581395348,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003957637631294566,
      "loss": 5.2034,
      "mean_token_accuracy": 0.20243831127882003,
      "num_tokens": 8163963.0,
      "step": 3635
    },
    {
      "entropy": 5.509068536758423,
      "epoch": 3.9356408869659276,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039540164126883554,
      "loss": 5.1223,
      "mean_token_accuracy": 0.20493561029434204,
      "num_tokens": 8174063.0,
      "step": 3640
    },
    {
      "entropy": 5.428995847702026,
      "epoch": 3.94104921579232,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003950390818683718,
      "loss": 5.1143,
      "mean_token_accuracy": 0.21023259907960892,
      "num_tokens": 8186800.0,
      "step": 3645
    },
    {
      "entropy": 5.533255290985108,
      "epoch": 3.946457544618713,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003946760862456071,
      "loss": 5.1984,
      "mean_token_accuracy": 0.20148286819458008,
      "num_tokens": 8198420.0,
      "step": 3650
    },
    {
      "entropy": 5.6072595596313475,
      "epoch": 3.9518658734451053,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003943126557196685,
      "loss": 5.2355,
      "mean_token_accuracy": 0.19470467716455458,
      "num_tokens": 8209384.0,
      "step": 3655
    },
    {
      "entropy": 5.523781871795654,
      "epoch": 3.957274202271498,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039394879161126333,
      "loss": 5.1691,
      "mean_token_accuracy": 0.20237645506858826,
      "num_tokens": 8220356.0,
      "step": 3660
    },
    {
      "entropy": 5.45121431350708,
      "epoch": 3.9626825310978906,
      "grad_norm": 0.921875,
      "learning_rate": 0.0003935844952426745,
      "loss": 5.1082,
      "mean_token_accuracy": 0.20572350025177003,
      "num_tokens": 8231597.0,
      "step": 3665
    },
    {
      "entropy": 5.464006519317627,
      "epoch": 3.9680908599242835,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00039321976793775586,
      "loss": 5.0818,
      "mean_token_accuracy": 0.20889972895383835,
      "num_tokens": 8242554.0,
      "step": 3670
    },
    {
      "entropy": 5.612120771408081,
      "epoch": 3.973499188750676,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003928546110219275,
      "loss": 5.3351,
      "mean_token_accuracy": 0.19498939365148543,
      "num_tokens": 8253396.0,
      "step": 3675
    },
    {
      "entropy": 5.468735790252685,
      "epoch": 3.9789075175770687,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039248902582217024,
      "loss": 5.0949,
      "mean_token_accuracy": 0.2172741934657097,
      "num_tokens": 8265193.0,
      "step": 3680
    },
    {
      "entropy": 5.454189443588257,
      "epoch": 3.9843158464034616,
      "grad_norm": 1.125,
      "learning_rate": 0.00039212301366702153,
      "loss": 5.1928,
      "mean_token_accuracy": 0.20339322984218597,
      "num_tokens": 8276343.0,
      "step": 3685
    },
    {
      "entropy": 5.57918267250061,
      "epoch": 3.989724175229854,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00039175657588657063,
      "loss": 5.192,
      "mean_token_accuracy": 0.20398834794759751,
      "num_tokens": 8288069.0,
      "step": 3690
    },
    {
      "entropy": 5.478389644622803,
      "epoch": 3.9951325040562464,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003913897138124531,
      "loss": 5.1568,
      "mean_token_accuracy": 0.20175088346004486,
      "num_tokens": 8299382.0,
      "step": 3695
    },
    {
      "entropy": 5.531948990292019,
      "epoch": 4.0,
      "grad_norm": 2.1875,
      "learning_rate": 0.0003910224287778466,
      "loss": 5.0795,
      "mean_token_accuracy": 0.21163836949401432,
      "num_tokens": 8307556.0,
      "step": 3700
    },
    {
      "entropy": 5.520035314559936,
      "epoch": 4.005408328826393,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039065472211746586,
      "loss": 4.9228,
      "mean_token_accuracy": 0.2212802678346634,
      "num_tokens": 8317851.0,
      "step": 3705
    },
    {
      "entropy": 5.408587694168091,
      "epoch": 4.010816657652786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003902865951675576,
      "loss": 4.8294,
      "mean_token_accuracy": 0.2355797693133354,
      "num_tokens": 8328485.0,
      "step": 3710
    },
    {
      "entropy": 5.495800352096557,
      "epoch": 4.016224986479178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038991804926589616,
      "loss": 4.787,
      "mean_token_accuracy": 0.23025935441255568,
      "num_tokens": 8339625.0,
      "step": 3715
    },
    {
      "entropy": 5.468115997314453,
      "epoch": 4.0216333153055706,
      "grad_norm": 1.0625,
      "learning_rate": 0.000389549085751778,
      "loss": 4.8482,
      "mean_token_accuracy": 0.2282474234700203,
      "num_tokens": 8350667.0,
      "step": 3720
    },
    {
      "entropy": 5.4827944278717045,
      "epoch": 4.027041644131963,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00038917970596601765,
      "loss": 4.9119,
      "mean_token_accuracy": 0.22222007513046266,
      "num_tokens": 8362521.0,
      "step": 3725
    },
    {
      "entropy": 5.4329047203063965,
      "epoch": 4.032449972958356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003888099112509419,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21877333968877793,
      "num_tokens": 8372703.0,
      "step": 3730
    },
    {
      "entropy": 5.5117429256439205,
      "epoch": 4.037858301784748,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038843970295038564,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22922151088714598,
      "num_tokens": 8383345.0,
      "step": 3735
    },
    {
      "entropy": 5.449793195724487,
      "epoch": 4.043266630611141,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038806908240968666,
      "loss": 4.8378,
      "mean_token_accuracy": 0.22729576230049134,
      "num_tokens": 8394056.0,
      "step": 3740
    },
    {
      "entropy": 5.394731092453003,
      "epoch": 4.048674959437534,
      "grad_norm": 1.109375,
      "learning_rate": 0.000387698050975681,
      "loss": 4.8664,
      "mean_token_accuracy": 0.21979550272226334,
      "num_tokens": 8403889.0,
      "step": 3745
    },
    {
      "entropy": 5.442848110198975,
      "epoch": 4.054083288263927,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003873266099966976,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22261063605546952,
      "num_tokens": 8415169.0,
      "step": 3750
    },
    {
      "entropy": 5.449613046646118,
      "epoch": 4.059491617090319,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038695476082255395,
      "loss": 4.881,
      "mean_token_accuracy": 0.21994651705026627,
      "num_tokens": 8427056.0,
      "step": 3755
    },
    {
      "entropy": 5.44387264251709,
      "epoch": 4.064899945916712,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003865825048045507,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2262551635503769,
      "num_tokens": 8438900.0,
      "step": 3760
    },
    {
      "entropy": 5.407828283309937,
      "epoch": 4.0703082747431045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003862098432954672,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22506212592124938,
      "num_tokens": 8449037.0,
      "step": 3765
    },
    {
      "entropy": 5.370322561264038,
      "epoch": 4.075716603569497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003858367776495561,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2245958462357521,
      "num_tokens": 8459409.0,
      "step": 3770
    },
    {
      "entropy": 5.483574962615966,
      "epoch": 4.081124932395889,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038546330922253886,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2214529037475586,
      "num_tokens": 8470190.0,
      "step": 3775
    },
    {
      "entropy": 5.470584678649902,
      "epoch": 4.086533261222282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003850894393716007,
      "loss": 4.917,
      "mean_token_accuracy": 0.2233967065811157,
      "num_tokens": 8481559.0,
      "step": 3780
    },
    {
      "entropy": 5.488635921478272,
      "epoch": 4.091941590048675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003847151694553855,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21162661164999008,
      "num_tokens": 8492607.0,
      "step": 3785
    },
    {
      "entropy": 5.490857934951782,
      "epoch": 4.097349918875068,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003843405008339908,
      "loss": 5.0331,
      "mean_token_accuracy": 0.21268152296543122,
      "num_tokens": 8504954.0,
      "step": 3790
    },
    {
      "entropy": 5.566803979873657,
      "epoch": 4.10275824770146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038396543486896354,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2198921263217926,
      "num_tokens": 8516250.0,
      "step": 3795
    },
    {
      "entropy": 5.42306432723999,
      "epoch": 4.108166576527853,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038358997292329406,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22527273148298263,
      "num_tokens": 8527220.0,
      "step": 3800
    },
    {
      "entropy": 5.470781707763672,
      "epoch": 4.113574905354246,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038321411636141214,
      "loss": 4.9746,
      "mean_token_accuracy": 0.22265234142541884,
      "num_tokens": 8538037.0,
      "step": 3805
    },
    {
      "entropy": 5.490799617767334,
      "epoch": 4.1189832341806385,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003828378665491811,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2232275053858757,
      "num_tokens": 8548684.0,
      "step": 3810
    },
    {
      "entropy": 5.374894809722901,
      "epoch": 4.1243915630070305,
      "grad_norm": 1.15625,
      "learning_rate": 0.000382461224853894,
      "loss": 4.8475,
      "mean_token_accuracy": 0.22701484560966492,
      "num_tokens": 8559756.0,
      "step": 3815
    },
    {
      "entropy": 5.443173027038574,
      "epoch": 4.129799891833423,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003820841926442674,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22235102206468582,
      "num_tokens": 8570640.0,
      "step": 3820
    },
    {
      "entropy": 5.488928174972534,
      "epoch": 4.135208220659816,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003817067712904372,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21921021193265916,
      "num_tokens": 8583127.0,
      "step": 3825
    },
    {
      "entropy": 5.407041931152344,
      "epoch": 4.140616549486209,
      "grad_norm": 0.953125,
      "learning_rate": 0.00038132896216395355,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2232694074511528,
      "num_tokens": 8595213.0,
      "step": 3830
    },
    {
      "entropy": 5.487379169464111,
      "epoch": 4.146024878312601,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003809507666377758,
      "loss": 4.8801,
      "mean_token_accuracy": 0.2309890180826187,
      "num_tokens": 8606057.0,
      "step": 3835
    },
    {
      "entropy": 5.457875061035156,
      "epoch": 4.151433207138994,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038057218608626714,
      "loss": 4.9334,
      "mean_token_accuracy": 0.22941599786281586,
      "num_tokens": 8617660.0,
      "step": 3840
    },
    {
      "entropy": 5.3874289989471436,
      "epoch": 4.156841535965387,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003801932218851903,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2234245151281357,
      "num_tokens": 8628595.0,
      "step": 3845
    },
    {
      "entropy": 5.517367935180664,
      "epoch": 4.16224986479178,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000379813875411702,
      "loss": 4.9288,
      "mean_token_accuracy": 0.22356290221214295,
      "num_tokens": 8639831.0,
      "step": 3850
    },
    {
      "entropy": 5.411230659484863,
      "epoch": 4.167658193618172,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037943414804434826,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22538353353738785,
      "num_tokens": 8650470.0,
      "step": 3855
    },
    {
      "entropy": 5.474798679351807,
      "epoch": 4.173066522444564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003790540411630592,
      "loss": 4.9756,
      "mean_token_accuracy": 0.22487561106681825,
      "num_tokens": 8660969.0,
      "step": 3860
    },
    {
      "entropy": 5.402866697311401,
      "epoch": 4.178474851270957,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037867355614914404,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23247915953397752,
      "num_tokens": 8670913.0,
      "step": 3865
    },
    {
      "entropy": 5.449716711044312,
      "epoch": 4.18388318009735,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003782926943852864,
      "loss": 4.9918,
      "mean_token_accuracy": 0.2167653203010559,
      "num_tokens": 8683028.0,
      "step": 3870
    },
    {
      "entropy": 5.56648268699646,
      "epoch": 4.189291508923742,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00037791145725553853,
      "loss": 5.0724,
      "mean_token_accuracy": 0.2143675222992897,
      "num_tokens": 8696454.0,
      "step": 3875
    },
    {
      "entropy": 5.498928880691528,
      "epoch": 4.194699837750135,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003775298461453172,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22822982668876649,
      "num_tokens": 8707112.0,
      "step": 3880
    },
    {
      "entropy": 5.3358461380004885,
      "epoch": 4.200108166576528,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000377147862441398,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23076649606227875,
      "num_tokens": 8717696.0,
      "step": 3885
    },
    {
      "entropy": 5.456122016906738,
      "epoch": 4.205516495402921,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037676550753191075,
      "loss": 4.811,
      "mean_token_accuracy": 0.23404202610254288,
      "num_tokens": 8728571.0,
      "step": 3890
    },
    {
      "entropy": 5.4752216815948485,
      "epoch": 4.210924824229313,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003763827828063339,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22195800691843032,
      "num_tokens": 8739582.0,
      "step": 3895
    },
    {
      "entropy": 5.4541913032531735,
      "epoch": 4.2163331530557056,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037599968965549,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22384155094623565,
      "num_tokens": 8750527.0,
      "step": 3900
    },
    {
      "entropy": 5.364770412445068,
      "epoch": 4.221741481882098,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003756162294715405,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22533732950687407,
      "num_tokens": 8763140.0,
      "step": 3905
    },
    {
      "entropy": 5.395046997070312,
      "epoch": 4.227149810708491,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037523240364798063,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22156916856765746,
      "num_tokens": 8774328.0,
      "step": 3910
    },
    {
      "entropy": 5.4685746192932125,
      "epoch": 4.232558139534884,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037484821357963423,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2204428419470787,
      "num_tokens": 8785119.0,
      "step": 3915
    },
    {
      "entropy": 5.43782148361206,
      "epoch": 4.237966468361276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000374463660662649,
      "loss": 4.8956,
      "mean_token_accuracy": 0.2206850528717041,
      "num_tokens": 8795833.0,
      "step": 3920
    },
    {
      "entropy": 5.426690864562988,
      "epoch": 4.243374797187669,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000374078746294491,
      "loss": 5.0016,
      "mean_token_accuracy": 0.21677581071853638,
      "num_tokens": 8807098.0,
      "step": 3925
    },
    {
      "entropy": 5.458262968063354,
      "epoch": 4.248783126014062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037369347187393996,
      "loss": 4.8209,
      "mean_token_accuracy": 0.2288977846503258,
      "num_tokens": 8818096.0,
      "step": 3930
    },
    {
      "entropy": 5.435509395599365,
      "epoch": 4.254191454840455,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003733078388010842,
      "loss": 4.9842,
      "mean_token_accuracy": 0.21152130514383316,
      "num_tokens": 8829451.0,
      "step": 3935
    },
    {
      "entropy": 5.439328718185425,
      "epoch": 4.259599783666847,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003729218484773149,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22326621413230896,
      "num_tokens": 8840099.0,
      "step": 3940
    },
    {
      "entropy": 5.4143470287322994,
      "epoch": 4.2650081124932395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003725355023053221,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23461422324180603,
      "num_tokens": 8851269.0,
      "step": 3945
    },
    {
      "entropy": 5.427573823928833,
      "epoch": 4.270416441319632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003721488016890884,
      "loss": 4.8576,
      "mean_token_accuracy": 0.23136806935071946,
      "num_tokens": 8862871.0,
      "step": 3950
    },
    {
      "entropy": 5.440256929397583,
      "epoch": 4.275824770146025,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037176174803388496,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22617517709732055,
      "num_tokens": 8874328.0,
      "step": 3955
    },
    {
      "entropy": 5.40334415435791,
      "epoch": 4.281233098972417,
      "grad_norm": 1.125,
      "learning_rate": 0.00037137434274626545,
      "loss": 4.9243,
      "mean_token_accuracy": 0.22224092185497285,
      "num_tokens": 8885133.0,
      "step": 3960
    },
    {
      "entropy": 5.419406366348267,
      "epoch": 4.28664142779881,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003709865872340618,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2234245240688324,
      "num_tokens": 8896479.0,
      "step": 3965
    },
    {
      "entropy": 5.488235378265381,
      "epoch": 4.292049756625203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003705984829063782,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22382310330867766,
      "num_tokens": 8907966.0,
      "step": 3970
    },
    {
      "entropy": 5.310491037368775,
      "epoch": 4.297458085451596,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003702100311735869,
      "loss": 4.8969,
      "mean_token_accuracy": 0.2224091038107872,
      "num_tokens": 8919042.0,
      "step": 3975
    },
    {
      "entropy": 5.443239688873291,
      "epoch": 4.302866414277988,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036982123344732223,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22111062854528427,
      "num_tokens": 8930456.0,
      "step": 3980
    },
    {
      "entropy": 5.407270240783691,
      "epoch": 4.308274743104381,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036943209114047613,
      "loss": 4.9591,
      "mean_token_accuracy": 0.22046105563640594,
      "num_tokens": 8941473.0,
      "step": 3985
    },
    {
      "entropy": 5.404595565795899,
      "epoch": 4.3136830719307735,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003690426056671926,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2196716845035553,
      "num_tokens": 8952927.0,
      "step": 3990
    },
    {
      "entropy": 5.434232759475708,
      "epoch": 4.319091400757166,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003686527784428627,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2178665205836296,
      "num_tokens": 8963831.0,
      "step": 3995
    },
    {
      "entropy": 5.426350879669189,
      "epoch": 4.324499729583558,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036826261088411955,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22362880408763885,
      "num_tokens": 8977147.0,
      "step": 4000
    },
    {
      "epoch": 4.324499729583558,
      "eval_entropy": 5.332500669089231,
      "eval_loss": 5.847503662109375,
      "eval_mean_token_accuracy": 0.18003634424372153,
      "eval_num_tokens": 8977147.0,
      "eval_runtime": 2.1319,
      "eval_samples_per_second": 1648.754,
      "eval_steps_per_second": 206.387,
      "step": 4000
    },
    {
      "entropy": 5.4549534797668455,
      "epoch": 4.329908058409951,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003678721044088329,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2287401780486107,
      "num_tokens": 8990031.0,
      "step": 4005
    },
    {
      "entropy": 5.368061113357544,
      "epoch": 4.335316387236344,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003674812604361043,
      "loss": 4.8282,
      "mean_token_accuracy": 0.22808760404586792,
      "num_tokens": 9000555.0,
      "step": 4010
    },
    {
      "entropy": 5.334468936920166,
      "epoch": 4.340724716062737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003670900803862615,
      "loss": 4.832,
      "mean_token_accuracy": 0.22556958049535752,
      "num_tokens": 9011631.0,
      "step": 4015
    },
    {
      "entropy": 5.375359678268433,
      "epoch": 4.346133044889129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003666985656808537,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22626398801803588,
      "num_tokens": 9023073.0,
      "step": 4020
    },
    {
      "entropy": 5.418436336517334,
      "epoch": 4.351541373715522,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036630671774264635,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21608079671859742,
      "num_tokens": 9033360.0,
      "step": 4025
    },
    {
      "entropy": 5.347191572189331,
      "epoch": 4.356949702541915,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003659145379956158,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22000972479581832,
      "num_tokens": 9044404.0,
      "step": 4030
    },
    {
      "entropy": 5.445996713638306,
      "epoch": 4.3623580313683075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036552202786494393,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22494267374277116,
      "num_tokens": 9054826.0,
      "step": 4035
    },
    {
      "entropy": 5.447552919387817,
      "epoch": 4.367766360194699,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036512918877701367,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22132477462291716,
      "num_tokens": 9065363.0,
      "step": 4040
    },
    {
      "entropy": 5.396865653991699,
      "epoch": 4.373174689021092,
      "grad_norm": 1.125,
      "learning_rate": 0.00036473602215940316,
      "loss": 4.896,
      "mean_token_accuracy": 0.22537353932857512,
      "num_tokens": 9075692.0,
      "step": 4045
    },
    {
      "entropy": 5.445511531829834,
      "epoch": 4.378583017847485,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003643425294408807,
      "loss": 4.9771,
      "mean_token_accuracy": 0.22071280777454377,
      "num_tokens": 9087618.0,
      "step": 4050
    },
    {
      "entropy": 5.427194309234619,
      "epoch": 4.383991346673878,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036394871205139984,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21783801317214965,
      "num_tokens": 9098508.0,
      "step": 4055
    },
    {
      "entropy": 5.386686182022094,
      "epoch": 4.38939967550027,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003635545714220938,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21990583688020707,
      "num_tokens": 9109003.0,
      "step": 4060
    },
    {
      "entropy": 5.386107015609741,
      "epoch": 4.394808004326663,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003631601089852706,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21359293907880783,
      "num_tokens": 9120615.0,
      "step": 4065
    },
    {
      "entropy": 5.513945150375366,
      "epoch": 4.400216333153056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003627653261744075,
      "loss": 5.0472,
      "mean_token_accuracy": 0.21691259145736694,
      "num_tokens": 9131814.0,
      "step": 4070
    },
    {
      "entropy": 5.434678173065185,
      "epoch": 4.405624661979449,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036237022442414636,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22214868664741516,
      "num_tokens": 9142443.0,
      "step": 4075
    },
    {
      "entropy": 5.45235767364502,
      "epoch": 4.411032990805841,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003619748051702878,
      "loss": 4.944,
      "mean_token_accuracy": 0.2173607423901558,
      "num_tokens": 9154433.0,
      "step": 4080
    },
    {
      "entropy": 5.336677598953247,
      "epoch": 4.416441319632233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003615790698497862,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21591013967990874,
      "num_tokens": 9165153.0,
      "step": 4085
    },
    {
      "entropy": 5.499279403686524,
      "epoch": 4.421849648458626,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036118301990074463,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21266622841358185,
      "num_tokens": 9176568.0,
      "step": 4090
    },
    {
      "entropy": 5.406652355194092,
      "epoch": 4.427257977285019,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003607866567624097,
      "loss": 4.9554,
      "mean_token_accuracy": 0.2182571455836296,
      "num_tokens": 9187205.0,
      "step": 4095
    },
    {
      "entropy": 5.482303810119629,
      "epoch": 4.432666306111411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036038998187516583,
      "loss": 5.0184,
      "mean_token_accuracy": 0.21168431639671326,
      "num_tokens": 9198486.0,
      "step": 4100
    },
    {
      "entropy": 5.365210342407226,
      "epoch": 4.438074634937804,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035999299668053043,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22566664814949036,
      "num_tokens": 9209714.0,
      "step": 4105
    },
    {
      "entropy": 5.38793830871582,
      "epoch": 4.443482963764197,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003595957026211487,
      "loss": 4.869,
      "mean_token_accuracy": 0.22279901504516603,
      "num_tokens": 9221968.0,
      "step": 4110
    },
    {
      "entropy": 5.394179391860962,
      "epoch": 4.44889129259059,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035919810114078815,
      "loss": 4.9459,
      "mean_token_accuracy": 0.21513068825006484,
      "num_tokens": 9234437.0,
      "step": 4115
    },
    {
      "entropy": 5.387730836868286,
      "epoch": 4.4542996214169825,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035880019368433354,
      "loss": 4.8425,
      "mean_token_accuracy": 0.22663739919662476,
      "num_tokens": 9246181.0,
      "step": 4120
    },
    {
      "entropy": 5.4676842212677,
      "epoch": 4.4597079502433745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003584019816977813,
      "loss": 4.9465,
      "mean_token_accuracy": 0.22468167841434478,
      "num_tokens": 9256951.0,
      "step": 4125
    },
    {
      "entropy": 5.407054996490478,
      "epoch": 4.465116279069767,
      "grad_norm": 1.125,
      "learning_rate": 0.00035800346662823493,
      "loss": 4.897,
      "mean_token_accuracy": 0.22603341788053513,
      "num_tokens": 9268783.0,
      "step": 4130
    },
    {
      "entropy": 5.42610502243042,
      "epoch": 4.47052460789616,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000357604649923899,
      "loss": 4.947,
      "mean_token_accuracy": 0.21931235641241073,
      "num_tokens": 9279541.0,
      "step": 4135
    },
    {
      "entropy": 5.3961913108825685,
      "epoch": 4.475932936722553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003572055330340744,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21654741019010543,
      "num_tokens": 9291041.0,
      "step": 4140
    },
    {
      "entropy": 5.351862049102783,
      "epoch": 4.481341265548945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035680611740915264,
      "loss": 4.9013,
      "mean_token_accuracy": 0.21511438935995103,
      "num_tokens": 9303451.0,
      "step": 4145
    },
    {
      "entropy": 5.448016023635864,
      "epoch": 4.486749594375338,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003564064045006113,
      "loss": 4.937,
      "mean_token_accuracy": 0.2192645251750946,
      "num_tokens": 9315275.0,
      "step": 4150
    },
    {
      "entropy": 5.446535682678222,
      "epoch": 4.492157923201731,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035600639576100786,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21564442664384842,
      "num_tokens": 9326987.0,
      "step": 4155
    },
    {
      "entropy": 5.377150583267212,
      "epoch": 4.497566252028124,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035560609264397497,
      "loss": 4.9452,
      "mean_token_accuracy": 0.22020510882139205,
      "num_tokens": 9337885.0,
      "step": 4160
    },
    {
      "entropy": 5.366753101348877,
      "epoch": 4.502974580854516,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035520549660421515,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21818556785583496,
      "num_tokens": 9348881.0,
      "step": 4165
    },
    {
      "entropy": 5.428653573989868,
      "epoch": 4.5083829096809085,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003548046090974954,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2252671405673027,
      "num_tokens": 9361066.0,
      "step": 4170
    },
    {
      "entropy": 5.468895149230957,
      "epoch": 4.513791238507301,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00035440343158064176,
      "loss": 4.8659,
      "mean_token_accuracy": 0.23179158419370652,
      "num_tokens": 9371985.0,
      "step": 4175
    },
    {
      "entropy": 5.459458923339843,
      "epoch": 4.519199567333694,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003540019655115341,
      "loss": 4.9885,
      "mean_token_accuracy": 0.21487326472997664,
      "num_tokens": 9383756.0,
      "step": 4180
    },
    {
      "entropy": 5.390002870559693,
      "epoch": 4.524607896160086,
      "grad_norm": 1.125,
      "learning_rate": 0.00035360021234910133,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2235135555267334,
      "num_tokens": 9394892.0,
      "step": 4185
    },
    {
      "entropy": 5.430325174331665,
      "epoch": 4.530016224986479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035319817355331537,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21850583851337432,
      "num_tokens": 9406646.0,
      "step": 4190
    },
    {
      "entropy": 5.450710391998291,
      "epoch": 4.535424553812872,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003527958505851858,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22576112151145936,
      "num_tokens": 9417548.0,
      "step": 4195
    },
    {
      "entropy": 5.40500545501709,
      "epoch": 4.540832882639265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035239324490675546,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22475125640630722,
      "num_tokens": 9428481.0,
      "step": 4200
    },
    {
      "entropy": 5.405156946182251,
      "epoch": 4.546241211465657,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003519903579810943,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2265077441930771,
      "num_tokens": 9439451.0,
      "step": 4205
    },
    {
      "entropy": 5.411196088790893,
      "epoch": 4.55164954029205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003515871912722944,
      "loss": 4.9078,
      "mean_token_accuracy": 0.2201046019792557,
      "num_tokens": 9450141.0,
      "step": 4210
    },
    {
      "entropy": 5.448217344284058,
      "epoch": 4.5570578691184425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003511837462454643,
      "loss": 4.9788,
      "mean_token_accuracy": 0.21594100594520568,
      "num_tokens": 9461309.0,
      "step": 4215
    },
    {
      "entropy": 5.418334341049194,
      "epoch": 4.562466197944835,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035078002436672443,
      "loss": 4.9088,
      "mean_token_accuracy": 0.22491281926631929,
      "num_tokens": 9473147.0,
      "step": 4220
    },
    {
      "entropy": 5.267909193038941,
      "epoch": 4.567874526771227,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003503760271032007,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2195964053273201,
      "num_tokens": 9483106.0,
      "step": 4225
    },
    {
      "entropy": 5.461530494689941,
      "epoch": 4.57328285559762,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003499717559230205,
      "loss": 4.8806,
      "mean_token_accuracy": 0.217904195189476,
      "num_tokens": 9493735.0,
      "step": 4230
    },
    {
      "entropy": 5.366152906417847,
      "epoch": 4.578691184424013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034956721229530586,
      "loss": 4.9503,
      "mean_token_accuracy": 0.22121719568967818,
      "num_tokens": 9505385.0,
      "step": 4235
    },
    {
      "entropy": 5.515100717544556,
      "epoch": 4.584099513250406,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003491623976901695,
      "loss": 5.0456,
      "mean_token_accuracy": 0.21188241392374038,
      "num_tokens": 9518170.0,
      "step": 4240
    },
    {
      "entropy": 5.460223579406739,
      "epoch": 4.589507842076799,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003487573135787085,
      "loss": 4.9882,
      "mean_token_accuracy": 0.2190849557518959,
      "num_tokens": 9530416.0,
      "step": 4245
    },
    {
      "entropy": 5.340732622146606,
      "epoch": 4.594916170903191,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034835196143299956,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2305539444088936,
      "num_tokens": 9542365.0,
      "step": 4250
    },
    {
      "entropy": 5.441924428939819,
      "epoch": 4.600324499729584,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034794634272609313,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21289959400892258,
      "num_tokens": 9552903.0,
      "step": 4255
    },
    {
      "entropy": 5.44968695640564,
      "epoch": 4.605732828555976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003475404589320089,
      "loss": 4.9614,
      "mean_token_accuracy": 0.22229813039302826,
      "num_tokens": 9563311.0,
      "step": 4260
    },
    {
      "entropy": 5.495689392089844,
      "epoch": 4.611141157382368,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003471343115257291,
      "loss": 5.0405,
      "mean_token_accuracy": 0.21441640853881835,
      "num_tokens": 9574673.0,
      "step": 4265
    },
    {
      "entropy": 5.429718828201294,
      "epoch": 4.616549486208761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034672790198319453,
      "loss": 4.9525,
      "mean_token_accuracy": 0.2051247164607048,
      "num_tokens": 9584572.0,
      "step": 4270
    },
    {
      "entropy": 5.360509967803955,
      "epoch": 4.621957815035154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003463212317812985,
      "loss": 4.8095,
      "mean_token_accuracy": 0.23005660027265548,
      "num_tokens": 9594828.0,
      "step": 4275
    },
    {
      "entropy": 5.388811016082764,
      "epoch": 4.627366143861547,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003459143023978813,
      "loss": 4.9576,
      "mean_token_accuracy": 0.21566459834575652,
      "num_tokens": 9606456.0,
      "step": 4280
    },
    {
      "entropy": 5.432236576080323,
      "epoch": 4.63277447268794,
      "grad_norm": 1.0,
      "learning_rate": 0.0003455071153117252,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22271626591682434,
      "num_tokens": 9618092.0,
      "step": 4285
    },
    {
      "entropy": 5.396778631210327,
      "epoch": 4.638182801514332,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003450996720025493,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22625730335712432,
      "num_tokens": 9629344.0,
      "step": 4290
    },
    {
      "entropy": 5.462570238113403,
      "epoch": 4.643591130340725,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00034469197395100323,
      "loss": 4.9606,
      "mean_token_accuracy": 0.20850926786661148,
      "num_tokens": 9639956.0,
      "step": 4295
    },
    {
      "entropy": 5.440243864059449,
      "epoch": 4.6489994591671175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034428402263866285,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22388518303632737,
      "num_tokens": 9649528.0,
      "step": 4300
    },
    {
      "entropy": 5.464076137542724,
      "epoch": 4.6544077879935095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034387581954802406,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21853445768356322,
      "num_tokens": 9661374.0,
      "step": 4305
    },
    {
      "entropy": 5.4249622344970705,
      "epoch": 4.659816116819902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034346736616249786,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21464212983846664,
      "num_tokens": 9672359.0,
      "step": 4310
    },
    {
      "entropy": 5.484592771530151,
      "epoch": 4.665224445646295,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034305866396640473,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21209099739789963,
      "num_tokens": 9683748.0,
      "step": 4315
    },
    {
      "entropy": 5.515557718276978,
      "epoch": 4.670632774472688,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034264971444496954,
      "loss": 4.9458,
      "mean_token_accuracy": 0.22219865769147873,
      "num_tokens": 9696049.0,
      "step": 4320
    },
    {
      "entropy": 5.38921012878418,
      "epoch": 4.676041103299081,
      "grad_norm": 1.25,
      "learning_rate": 0.00034224051908431574,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22372995615005492,
      "num_tokens": 9707358.0,
      "step": 4325
    },
    {
      "entropy": 5.394842195510864,
      "epoch": 4.681449432125473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003418310793714602,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21947879940271378,
      "num_tokens": 9718890.0,
      "step": 4330
    },
    {
      "entropy": 5.3791015625,
      "epoch": 4.686857760951866,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003414213967943077,
      "loss": 4.8017,
      "mean_token_accuracy": 0.23549789190292358,
      "num_tokens": 9730395.0,
      "step": 4335
    },
    {
      "entropy": 5.466121959686279,
      "epoch": 4.692266089778259,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034101147284164583,
      "loss": 4.9148,
      "mean_token_accuracy": 0.2171756759285927,
      "num_tokens": 9741846.0,
      "step": 4340
    },
    {
      "entropy": 5.39025092124939,
      "epoch": 4.6976744186046515,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034060130900313905,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22450713515281678,
      "num_tokens": 9752936.0,
      "step": 4345
    },
    {
      "entropy": 5.378725051879883,
      "epoch": 4.7030827474310435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034019090676932383,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2272382527589798,
      "num_tokens": 9763913.0,
      "step": 4350
    },
    {
      "entropy": 5.436556625366211,
      "epoch": 4.708491076257436,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003397802676316027,
      "loss": 4.984,
      "mean_token_accuracy": 0.2199976548552513,
      "num_tokens": 9775682.0,
      "step": 4355
    },
    {
      "entropy": 5.3924651622772215,
      "epoch": 4.713899405083829,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033936939308223935,
      "loss": 5.0044,
      "mean_token_accuracy": 0.21096239387989044,
      "num_tokens": 9787403.0,
      "step": 4360
    },
    {
      "entropy": 5.526662111282349,
      "epoch": 4.719307733910222,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033895828461435275,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22209493368864058,
      "num_tokens": 9798166.0,
      "step": 4365
    },
    {
      "entropy": 5.46227650642395,
      "epoch": 4.724716062736614,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003385469437219121,
      "loss": 4.9075,
      "mean_token_accuracy": 0.2240492507815361,
      "num_tokens": 9809405.0,
      "step": 4370
    },
    {
      "entropy": 5.448237895965576,
      "epoch": 4.730124391563007,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033813537189973105,
      "loss": 4.9305,
      "mean_token_accuracy": 0.22060854882001876,
      "num_tokens": 9820953.0,
      "step": 4375
    },
    {
      "entropy": 5.467635297775269,
      "epoch": 4.7355327203894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033772357064346266,
      "loss": 4.9842,
      "mean_token_accuracy": 0.22067289799451828,
      "num_tokens": 9831916.0,
      "step": 4380
    },
    {
      "entropy": 5.395278549194336,
      "epoch": 4.740941049215793,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003373115414495935,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22243587225675582,
      "num_tokens": 9841800.0,
      "step": 4385
    },
    {
      "entropy": 5.459128522872925,
      "epoch": 4.746349378042185,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003368992858154389,
      "loss": 5.0388,
      "mean_token_accuracy": 0.21994709074497223,
      "num_tokens": 9854713.0,
      "step": 4390
    },
    {
      "entropy": 5.478680372238159,
      "epoch": 4.7517577068685775,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033648680523913643,
      "loss": 4.9421,
      "mean_token_accuracy": 0.22057987302541732,
      "num_tokens": 9865730.0,
      "step": 4395
    },
    {
      "entropy": 5.514898300170898,
      "epoch": 4.75716603569497,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003360741012196417,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2231214612722397,
      "num_tokens": 9875771.0,
      "step": 4400
    },
    {
      "entropy": 5.425652360916137,
      "epoch": 4.762574364521363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000335661175256722,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2126207932829857,
      "num_tokens": 9888839.0,
      "step": 4405
    },
    {
      "entropy": 5.4566521644592285,
      "epoch": 4.767982693347756,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003352480288509513,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22401460856199265,
      "num_tokens": 9899150.0,
      "step": 4410
    },
    {
      "entropy": 5.47235426902771,
      "epoch": 4.773391022174148,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033483466350370445,
      "loss": 5.0867,
      "mean_token_accuracy": 0.20992355346679686,
      "num_tokens": 9912083.0,
      "step": 4415
    },
    {
      "entropy": 5.445180797576905,
      "epoch": 4.778799351000541,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00033442108071715215,
      "loss": 5.0106,
      "mean_token_accuracy": 0.2141200765967369,
      "num_tokens": 9924217.0,
      "step": 4420
    },
    {
      "entropy": 5.571099424362183,
      "epoch": 4.784207679826934,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033400728199425525,
      "loss": 5.0595,
      "mean_token_accuracy": 0.21216620802879332,
      "num_tokens": 9936298.0,
      "step": 4425
    },
    {
      "entropy": 5.372473907470703,
      "epoch": 4.789616008653326,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033359326883875917,
      "loss": 4.8443,
      "mean_token_accuracy": 0.2292049452662468,
      "num_tokens": 9946752.0,
      "step": 4430
    },
    {
      "entropy": 5.410205125808716,
      "epoch": 4.795024337479719,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033317904275518873,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22421593070030213,
      "num_tokens": 9958445.0,
      "step": 4435
    },
    {
      "entropy": 5.44655933380127,
      "epoch": 4.800432666306111,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003327646052488425,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21902037411928177,
      "num_tokens": 9970021.0,
      "step": 4440
    },
    {
      "entropy": 5.349941968917847,
      "epoch": 4.805840995132504,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003323499578257873,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2325447231531143,
      "num_tokens": 9979703.0,
      "step": 4445
    },
    {
      "entropy": 5.373944044113159,
      "epoch": 4.811249323958897,
      "grad_norm": 1.125,
      "learning_rate": 0.00033193510199285286,
      "loss": 4.9239,
      "mean_token_accuracy": 0.22446109503507614,
      "num_tokens": 9991574.0,
      "step": 4450
    },
    {
      "entropy": 5.347008085250854,
      "epoch": 4.816657652785289,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033152003925762615,
      "loss": 4.872,
      "mean_token_accuracy": 0.22051797211170196,
      "num_tokens": 10002410.0,
      "step": 4455
    },
    {
      "entropy": 5.427417230606079,
      "epoch": 4.822065981611682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003311047711284462,
      "loss": 4.8447,
      "mean_token_accuracy": 0.22524640560150147,
      "num_tokens": 10012576.0,
      "step": 4460
    },
    {
      "entropy": 5.454698085784912,
      "epoch": 4.827474310438075,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003306892991143983,
      "loss": 4.948,
      "mean_token_accuracy": 0.21940114647150039,
      "num_tokens": 10024298.0,
      "step": 4465
    },
    {
      "entropy": 5.33115930557251,
      "epoch": 4.832882639264467,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003302736247253087,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22948517501354218,
      "num_tokens": 10035835.0,
      "step": 4470
    },
    {
      "entropy": 5.473714208602905,
      "epoch": 4.83829096809086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003298577494717391,
      "loss": 4.9685,
      "mean_token_accuracy": 0.22131202816963197,
      "num_tokens": 10046880.0,
      "step": 4475
    },
    {
      "entropy": 5.457976627349853,
      "epoch": 4.8436992969172525,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003294416748649812,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2206039696931839,
      "num_tokens": 10057761.0,
      "step": 4480
    },
    {
      "entropy": 5.282527160644531,
      "epoch": 4.849107625743645,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032902540241705106,
      "loss": 4.799,
      "mean_token_accuracy": 0.23099472671747207,
      "num_tokens": 10069714.0,
      "step": 4485
    },
    {
      "entropy": 5.3357329845428465,
      "epoch": 4.854515954570038,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003286089336406837,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23387452363967895,
      "num_tokens": 10080292.0,
      "step": 4490
    },
    {
      "entropy": 5.361022138595581,
      "epoch": 4.85992428339643,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032819227004932745,
      "loss": 4.9843,
      "mean_token_accuracy": 0.21238050013780593,
      "num_tokens": 10091707.0,
      "step": 4495
    },
    {
      "entropy": 5.413997077941895,
      "epoch": 4.865332612222823,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000327775413157139,
      "loss": 4.8936,
      "mean_token_accuracy": 0.2183866173028946,
      "num_tokens": 10103268.0,
      "step": 4500
    },
    {
      "entropy": 5.377324533462525,
      "epoch": 4.870740941049216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003273583644789772,
      "loss": 4.9202,
      "mean_token_accuracy": 0.22091448903083802,
      "num_tokens": 10114690.0,
      "step": 4505
    },
    {
      "entropy": 5.391185569763183,
      "epoch": 4.876149269875609,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032694112553039795,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22888859510421752,
      "num_tokens": 10125126.0,
      "step": 4510
    },
    {
      "entropy": 5.357927751541138,
      "epoch": 4.881557598702001,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003265236978276484,
      "loss": 4.879,
      "mean_token_accuracy": 0.2212603896856308,
      "num_tokens": 10137093.0,
      "step": 4515
    },
    {
      "entropy": 5.361552143096924,
      "epoch": 4.886965927528394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032610608288766185,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2315201535820961,
      "num_tokens": 10147487.0,
      "step": 4520
    },
    {
      "entropy": 5.500840759277343,
      "epoch": 4.8923742563547865,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032568828222805216,
      "loss": 5.0577,
      "mean_token_accuracy": 0.22175925374031066,
      "num_tokens": 10160351.0,
      "step": 4525
    },
    {
      "entropy": 5.409867429733277,
      "epoch": 4.897782585181179,
      "grad_norm": 1.125,
      "learning_rate": 0.0003252702973671075,
      "loss": 4.9824,
      "mean_token_accuracy": 0.2146667420864105,
      "num_tokens": 10172059.0,
      "step": 4530
    },
    {
      "entropy": 5.4408222198486325,
      "epoch": 4.903190914007571,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032485212982378615,
      "loss": 5.02,
      "mean_token_accuracy": 0.20919327288866044,
      "num_tokens": 10182758.0,
      "step": 4535
    },
    {
      "entropy": 5.472556495666504,
      "epoch": 4.908599242833964,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003244337811177098,
      "loss": 4.9572,
      "mean_token_accuracy": 0.22194432020187377,
      "num_tokens": 10193391.0,
      "step": 4540
    },
    {
      "entropy": 5.397055149078369,
      "epoch": 4.914007571660357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003240152527691586,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2294257566332817,
      "num_tokens": 10204298.0,
      "step": 4545
    },
    {
      "entropy": 5.413761043548584,
      "epoch": 4.91941590048675,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032359654629906555,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21654497534036637,
      "num_tokens": 10215115.0,
      "step": 4550
    },
    {
      "entropy": 5.455242395401001,
      "epoch": 4.924824229313142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032317766322901106,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21748327016830443,
      "num_tokens": 10227649.0,
      "step": 4555
    },
    {
      "entropy": 5.358527135848999,
      "epoch": 4.930232558139535,
      "grad_norm": 1.0625,
      "learning_rate": 0.000322758605081217,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2238612025976181,
      "num_tokens": 10238830.0,
      "step": 4560
    },
    {
      "entropy": 5.345400953292847,
      "epoch": 4.935640886965928,
      "grad_norm": 1.125,
      "learning_rate": 0.00032233937337854167,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21712866723537444,
      "num_tokens": 10250431.0,
      "step": 4565
    },
    {
      "entropy": 5.498257064819336,
      "epoch": 4.9410492157923205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003219199696444741,
      "loss": 5.0571,
      "mean_token_accuracy": 0.21266702860593795,
      "num_tokens": 10263070.0,
      "step": 4570
    },
    {
      "entropy": 5.399999523162842,
      "epoch": 4.9464575446187125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032150039540312843,
      "loss": 4.9816,
      "mean_token_accuracy": 0.21757111698389053,
      "num_tokens": 10275647.0,
      "step": 4575
    },
    {
      "entropy": 5.379177236557007,
      "epoch": 4.951865873445105,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032108065217923836,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21551714539527894,
      "num_tokens": 10286651.0,
      "step": 4580
    },
    {
      "entropy": 5.359277009963989,
      "epoch": 4.957274202271498,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032066074149815184,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2147955670952797,
      "num_tokens": 10296692.0,
      "step": 4585
    },
    {
      "entropy": 5.366557979583741,
      "epoch": 4.962682531097891,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00032024066488582517,
      "loss": 4.955,
      "mean_token_accuracy": 0.21572295129299163,
      "num_tokens": 10307936.0,
      "step": 4590
    },
    {
      "entropy": 5.412725496292114,
      "epoch": 4.968090859924283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003198204238688178,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22834654450416564,
      "num_tokens": 10318637.0,
      "step": 4595
    },
    {
      "entropy": 5.3761858463287355,
      "epoch": 4.973499188750676,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031940001997428645,
      "loss": 4.7814,
      "mean_token_accuracy": 0.22822368890047073,
      "num_tokens": 10328432.0,
      "step": 4600
    },
    {
      "entropy": 5.271241283416748,
      "epoch": 4.978907517577069,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031897945472997996,
      "loss": 4.8322,
      "mean_token_accuracy": 0.22686116844415666,
      "num_tokens": 10338971.0,
      "step": 4605
    },
    {
      "entropy": 5.332295608520508,
      "epoch": 4.984315846403462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003185587296642333,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21835770606994628,
      "num_tokens": 10350586.0,
      "step": 4610
    },
    {
      "entropy": 5.440207290649414,
      "epoch": 4.9897241752298545,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003181378463059625,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22896879017353058,
      "num_tokens": 10363150.0,
      "step": 4615
    },
    {
      "entropy": 5.408982610702514,
      "epoch": 4.995132504056246,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003177168061846584,
      "loss": 4.8826,
      "mean_token_accuracy": 0.21777461022138594,
      "num_tokens": 10374431.0,
      "step": 4620
    },
    {
      "entropy": 5.3562942610846624,
      "epoch": 5.0,
      "grad_norm": 2.640625,
      "learning_rate": 0.0003172956108303819,
      "loss": 4.9467,
      "mean_token_accuracy": 0.2327342066499922,
      "num_tokens": 10384445.0,
      "step": 4625
    },
    {
      "entropy": 5.459956550598145,
      "epoch": 5.005408328826393,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003168742617737579,
      "loss": 4.7684,
      "mean_token_accuracy": 0.23015496581792833,
      "num_tokens": 10395995.0,
      "step": 4630
    },
    {
      "entropy": 5.436410331726075,
      "epoch": 5.010816657652786,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031645276054597006,
      "loss": 4.682,
      "mean_token_accuracy": 0.23784464448690415,
      "num_tokens": 10408385.0,
      "step": 4635
    },
    {
      "entropy": 5.272504138946533,
      "epoch": 5.016224986479178,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031603110867875435,
      "loss": 4.6338,
      "mean_token_accuracy": 0.2485280379652977,
      "num_tokens": 10419629.0,
      "step": 4640
    },
    {
      "entropy": 5.4224607944488525,
      "epoch": 5.0216333153055706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000315609307704395,
      "loss": 4.7715,
      "mean_token_accuracy": 0.2302168309688568,
      "num_tokens": 10431040.0,
      "step": 4645
    },
    {
      "entropy": 5.395181369781494,
      "epoch": 5.027041644131963,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003151873591557176,
      "loss": 4.677,
      "mean_token_accuracy": 0.24433448612689973,
      "num_tokens": 10442909.0,
      "step": 4650
    },
    {
      "entropy": 5.305405378341675,
      "epoch": 5.032449972958356,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031476526456608415,
      "loss": 4.622,
      "mean_token_accuracy": 0.24823571592569352,
      "num_tokens": 10454008.0,
      "step": 4655
    },
    {
      "entropy": 5.385721635818482,
      "epoch": 5.037858301784748,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003143430254693873,
      "loss": 4.6733,
      "mean_token_accuracy": 0.236166612803936,
      "num_tokens": 10465037.0,
      "step": 4660
    },
    {
      "entropy": 5.415099954605102,
      "epoch": 5.043266630611141,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003139206434000451,
      "loss": 4.6904,
      "mean_token_accuracy": 0.2370677560567856,
      "num_tokens": 10475654.0,
      "step": 4665
    },
    {
      "entropy": 5.359461736679077,
      "epoch": 5.048674959437534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003134981198929948,
      "loss": 4.7458,
      "mean_token_accuracy": 0.23621846437454225,
      "num_tokens": 10486700.0,
      "step": 4670
    },
    {
      "entropy": 5.382424783706665,
      "epoch": 5.054083288263927,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031307545648368777,
      "loss": 4.7364,
      "mean_token_accuracy": 0.23580776304006576,
      "num_tokens": 10498055.0,
      "step": 4675
    },
    {
      "entropy": 5.377408361434936,
      "epoch": 5.059491617090319,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031265265470808393,
      "loss": 4.7743,
      "mean_token_accuracy": 0.23825181871652604,
      "num_tokens": 10511418.0,
      "step": 4680
    },
    {
      "entropy": 5.286991453170776,
      "epoch": 5.064899945916712,
      "grad_norm": 1.125,
      "learning_rate": 0.00031222971610264595,
      "loss": 4.5461,
      "mean_token_accuracy": 0.25928134471178055,
      "num_tokens": 10522498.0,
      "step": 4685
    },
    {
      "entropy": 5.393325090408325,
      "epoch": 5.0703082747431045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031180664220433347,
      "loss": 4.7109,
      "mean_token_accuracy": 0.23890808671712876,
      "num_tokens": 10534366.0,
      "step": 4690
    },
    {
      "entropy": 5.494242143630982,
      "epoch": 5.075716603569497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031138343455059823,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23688239008188247,
      "num_tokens": 10545256.0,
      "step": 4695
    },
    {
      "entropy": 5.349709749221802,
      "epoch": 5.081124932395889,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003109600946793778,
      "loss": 4.6432,
      "mean_token_accuracy": 0.2508495718240738,
      "num_tokens": 10555927.0,
      "step": 4700
    },
    {
      "entropy": 5.31465630531311,
      "epoch": 5.086533261222282,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031053662412909024,
      "loss": 4.6787,
      "mean_token_accuracy": 0.24274386167526246,
      "num_tokens": 10567383.0,
      "step": 4705
    },
    {
      "entropy": 5.433224296569824,
      "epoch": 5.091941590048675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003101130244386283,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2346149206161499,
      "num_tokens": 10580257.0,
      "step": 4710
    },
    {
      "entropy": 5.345867156982422,
      "epoch": 5.097349918875068,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030968929714735456,
      "loss": 4.7362,
      "mean_token_accuracy": 0.24116515070199968,
      "num_tokens": 10590838.0,
      "step": 4715
    },
    {
      "entropy": 5.344528102874756,
      "epoch": 5.10275824770146,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003092654437950947,
      "loss": 4.6637,
      "mean_token_accuracy": 0.23820921778678894,
      "num_tokens": 10602254.0,
      "step": 4720
    },
    {
      "entropy": 5.428279399871826,
      "epoch": 5.108166576527853,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003088414659221331,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23291309773921967,
      "num_tokens": 10613630.0,
      "step": 4725
    },
    {
      "entropy": 5.332703256607056,
      "epoch": 5.113574905354246,
      "grad_norm": 1.203125,
      "learning_rate": 0.000308417365069206,
      "loss": 4.5904,
      "mean_token_accuracy": 0.25217919051647186,
      "num_tokens": 10623754.0,
      "step": 4730
    },
    {
      "entropy": 5.369397497177124,
      "epoch": 5.1189832341806385,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030799314277749697,
      "loss": 4.7064,
      "mean_token_accuracy": 0.24109421968460082,
      "num_tokens": 10633071.0,
      "step": 4735
    },
    {
      "entropy": 5.342383003234863,
      "epoch": 5.1243915630070305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030756880058863085,
      "loss": 4.6632,
      "mean_token_accuracy": 0.24114688932895662,
      "num_tokens": 10643963.0,
      "step": 4740
    },
    {
      "entropy": 5.317634534835816,
      "epoch": 5.129799891833423,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030714434004466823,
      "loss": 4.6938,
      "mean_token_accuracy": 0.23998381495475768,
      "num_tokens": 10655716.0,
      "step": 4745
    },
    {
      "entropy": 5.340021371841431,
      "epoch": 5.135208220659816,
      "grad_norm": 1.25,
      "learning_rate": 0.00030671976268809947,
      "loss": 4.746,
      "mean_token_accuracy": 0.23224331438541412,
      "num_tokens": 10666231.0,
      "step": 4750
    },
    {
      "entropy": 5.39808669090271,
      "epoch": 5.140616549486209,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030629507006183995,
      "loss": 4.7023,
      "mean_token_accuracy": 0.2386842116713524,
      "num_tokens": 10677672.0,
      "step": 4755
    },
    {
      "entropy": 5.3467323780059814,
      "epoch": 5.146024878312601,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003058702637092234,
      "loss": 4.7446,
      "mean_token_accuracy": 0.2237921416759491,
      "num_tokens": 10689968.0,
      "step": 4760
    },
    {
      "entropy": 5.383655166625976,
      "epoch": 5.151433207138994,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030544534517399724,
      "loss": 4.7384,
      "mean_token_accuracy": 0.240167972445488,
      "num_tokens": 10699705.0,
      "step": 4765
    },
    {
      "entropy": 5.379343748092651,
      "epoch": 5.156841535965387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003050203160003163,
      "loss": 4.6859,
      "mean_token_accuracy": 0.2364353984594345,
      "num_tokens": 10710464.0,
      "step": 4770
    },
    {
      "entropy": 5.373980474472046,
      "epoch": 5.16224986479178,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003045951777327377,
      "loss": 4.6906,
      "mean_token_accuracy": 0.2424035921692848,
      "num_tokens": 10720592.0,
      "step": 4775
    },
    {
      "entropy": 5.383245468139648,
      "epoch": 5.167658193618172,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003041699319162148,
      "loss": 4.7482,
      "mean_token_accuracy": 0.23516542017459868,
      "num_tokens": 10731670.0,
      "step": 4780
    },
    {
      "entropy": 5.371738243103027,
      "epoch": 5.173066522444564,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030374458009609193,
      "loss": 4.6898,
      "mean_token_accuracy": 0.24281259030103683,
      "num_tokens": 10743123.0,
      "step": 4785
    },
    {
      "entropy": 5.307730102539063,
      "epoch": 5.178474851270957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003033191238180985,
      "loss": 4.7049,
      "mean_token_accuracy": 0.2371360182762146,
      "num_tokens": 10753855.0,
      "step": 4790
    },
    {
      "entropy": 5.292190027236939,
      "epoch": 5.18388318009735,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003028935646283438,
      "loss": 4.6999,
      "mean_token_accuracy": 0.235623337328434,
      "num_tokens": 10765275.0,
      "step": 4795
    },
    {
      "entropy": 5.3043006420135494,
      "epoch": 5.189291508923742,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030246790407331063,
      "loss": 4.6974,
      "mean_token_accuracy": 0.23308577984571457,
      "num_tokens": 10776563.0,
      "step": 4800
    },
    {
      "entropy": 5.379180574417115,
      "epoch": 5.194699837750135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030204214369985066,
      "loss": 4.6498,
      "mean_token_accuracy": 0.2424623504281044,
      "num_tokens": 10786989.0,
      "step": 4805
    },
    {
      "entropy": 5.3146830081939695,
      "epoch": 5.200108166576528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030161628505517797,
      "loss": 4.669,
      "mean_token_accuracy": 0.2349323719739914,
      "num_tokens": 10798460.0,
      "step": 4810
    },
    {
      "entropy": 5.326756763458252,
      "epoch": 5.205516495402921,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030119032968686397,
      "loss": 4.7365,
      "mean_token_accuracy": 0.23800140619277954,
      "num_tokens": 10810100.0,
      "step": 4815
    },
    {
      "entropy": 5.368277502059937,
      "epoch": 5.210924824229313,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030076427914283127,
      "loss": 4.7114,
      "mean_token_accuracy": 0.23949310332536697,
      "num_tokens": 10821019.0,
      "step": 4820
    },
    {
      "entropy": 5.3036675453186035,
      "epoch": 5.2163331530557056,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030033813497134864,
      "loss": 4.6521,
      "mean_token_accuracy": 0.23777441084384918,
      "num_tokens": 10831583.0,
      "step": 4825
    },
    {
      "entropy": 5.235832071304321,
      "epoch": 5.221741481882098,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000299911898721025,
      "loss": 4.6685,
      "mean_token_accuracy": 0.23695015013217927,
      "num_tokens": 10844117.0,
      "step": 4830
    },
    {
      "entropy": 5.304520893096924,
      "epoch": 5.227149810708491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029948557194080385,
      "loss": 4.6908,
      "mean_token_accuracy": 0.24506817311048507,
      "num_tokens": 10855577.0,
      "step": 4835
    },
    {
      "entropy": 5.328809785842895,
      "epoch": 5.232558139534884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002990591561799576,
      "loss": 4.7243,
      "mean_token_accuracy": 0.2375485822558403,
      "num_tokens": 10867153.0,
      "step": 4840
    },
    {
      "entropy": 5.45048041343689,
      "epoch": 5.237966468361276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002986326529880823,
      "loss": 4.814,
      "mean_token_accuracy": 0.2318206861615181,
      "num_tokens": 10878666.0,
      "step": 4845
    },
    {
      "entropy": 5.378794384002686,
      "epoch": 5.243374797187669,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002982060639150913,
      "loss": 4.7382,
      "mean_token_accuracy": 0.2338978186249733,
      "num_tokens": 10890197.0,
      "step": 4850
    },
    {
      "entropy": 5.3411033153533936,
      "epoch": 5.248783126014062,
      "grad_norm": 1.0,
      "learning_rate": 0.0002977793905112106,
      "loss": 4.6894,
      "mean_token_accuracy": 0.23268816322088243,
      "num_tokens": 10901423.0,
      "step": 4855
    },
    {
      "entropy": 5.370946884155273,
      "epoch": 5.254191454840455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029735263432697183,
      "loss": 4.7397,
      "mean_token_accuracy": 0.2379957318305969,
      "num_tokens": 10913023.0,
      "step": 4860
    },
    {
      "entropy": 5.428412055969238,
      "epoch": 5.259599783666847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002969257969132085,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23058093190193177,
      "num_tokens": 10924404.0,
      "step": 4865
    },
    {
      "entropy": 5.408817958831787,
      "epoch": 5.2650081124932395,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002964988798210485,
      "loss": 4.7966,
      "mean_token_accuracy": 0.22653153389692307,
      "num_tokens": 10935001.0,
      "step": 4870
    },
    {
      "entropy": 5.353402948379516,
      "epoch": 5.270416441319632,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002960718846019095,
      "loss": 4.7905,
      "mean_token_accuracy": 0.22905509173870087,
      "num_tokens": 10947035.0,
      "step": 4875
    },
    {
      "entropy": 5.425058221817016,
      "epoch": 5.275824770146025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002956448128074931,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2319728136062622,
      "num_tokens": 10960387.0,
      "step": 4880
    },
    {
      "entropy": 5.280859279632568,
      "epoch": 5.281233098972417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002952176659897794,
      "loss": 4.5967,
      "mean_token_accuracy": 0.2561764642596245,
      "num_tokens": 10972949.0,
      "step": 4885
    },
    {
      "entropy": 5.245252418518066,
      "epoch": 5.28664142779881,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002947904457010207,
      "loss": 4.6778,
      "mean_token_accuracy": 0.2346836656332016,
      "num_tokens": 10983488.0,
      "step": 4890
    },
    {
      "entropy": 5.326966571807861,
      "epoch": 5.292049756625203,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002943631534937366,
      "loss": 4.7068,
      "mean_token_accuracy": 0.23458608239889145,
      "num_tokens": 10995453.0,
      "step": 4895
    },
    {
      "entropy": 5.478303909301758,
      "epoch": 5.297458085451596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002939357909207079,
      "loss": 4.88,
      "mean_token_accuracy": 0.23047704994678497,
      "num_tokens": 11007945.0,
      "step": 4900
    },
    {
      "entropy": 5.400589513778686,
      "epoch": 5.302866414277988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002935083595349713,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23118148744106293,
      "num_tokens": 11020574.0,
      "step": 4905
    },
    {
      "entropy": 5.341559982299804,
      "epoch": 5.308274743104381,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029308086088981335,
      "loss": 4.6776,
      "mean_token_accuracy": 0.2404978409409523,
      "num_tokens": 11030893.0,
      "step": 4910
    },
    {
      "entropy": 5.307660675048828,
      "epoch": 5.3136830719307735,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029265329653876515,
      "loss": 4.7468,
      "mean_token_accuracy": 0.23058863282203673,
      "num_tokens": 11042104.0,
      "step": 4915
    },
    {
      "entropy": 5.349027538299561,
      "epoch": 5.319091400757166,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002922256680355965,
      "loss": 4.7741,
      "mean_token_accuracy": 0.22687351256608962,
      "num_tokens": 11051961.0,
      "step": 4920
    },
    {
      "entropy": 5.441255187988281,
      "epoch": 5.324499729583558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029179797693431046,
      "loss": 4.8304,
      "mean_token_accuracy": 0.22856033593416214,
      "num_tokens": 11063877.0,
      "step": 4925
    },
    {
      "entropy": 5.358956670761108,
      "epoch": 5.329908058409951,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002913702247891374,
      "loss": 4.6585,
      "mean_token_accuracy": 0.25383347272872925,
      "num_tokens": 11074085.0,
      "step": 4930
    },
    {
      "entropy": 5.321387529373169,
      "epoch": 5.335316387236344,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002909424131545296,
      "loss": 4.7267,
      "mean_token_accuracy": 0.23633771538734435,
      "num_tokens": 11085689.0,
      "step": 4935
    },
    {
      "entropy": 5.363430118560791,
      "epoch": 5.340724716062737,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029051454358515553,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23704927414655685,
      "num_tokens": 11096635.0,
      "step": 4940
    },
    {
      "entropy": 5.2950841903686525,
      "epoch": 5.346133044889129,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029008661763589436,
      "loss": 4.7284,
      "mean_token_accuracy": 0.23479934483766557,
      "num_tokens": 11107646.0,
      "step": 4945
    },
    {
      "entropy": 5.3303381443023685,
      "epoch": 5.351541373715522,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002896586368618297,
      "loss": 4.7172,
      "mean_token_accuracy": 0.23344653397798537,
      "num_tokens": 11117785.0,
      "step": 4950
    },
    {
      "entropy": 5.3246660232543945,
      "epoch": 5.356949702541915,
      "grad_norm": 1.25,
      "learning_rate": 0.00028923060281824483,
      "loss": 4.6906,
      "mean_token_accuracy": 0.24123958945274354,
      "num_tokens": 11129668.0,
      "step": 4955
    },
    {
      "entropy": 5.354137754440307,
      "epoch": 5.3623580313683075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002888025170606165,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2327262580394745,
      "num_tokens": 11141549.0,
      "step": 4960
    },
    {
      "entropy": 5.332718086242676,
      "epoch": 5.367766360194699,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028837438114460924,
      "loss": 4.7419,
      "mean_token_accuracy": 0.23853450119495392,
      "num_tokens": 11153599.0,
      "step": 4965
    },
    {
      "entropy": 5.377040863037109,
      "epoch": 5.373174689021092,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002879461966260699,
      "loss": 4.8,
      "mean_token_accuracy": 0.22976029813289642,
      "num_tokens": 11164392.0,
      "step": 4970
    },
    {
      "entropy": 5.3175136089324955,
      "epoch": 5.378583017847485,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028751796506102224,
      "loss": 4.6844,
      "mean_token_accuracy": 0.23883168399333954,
      "num_tokens": 11175707.0,
      "step": 4975
    },
    {
      "entropy": 5.381155920028687,
      "epoch": 5.383991346673878,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028708968800566056,
      "loss": 4.7908,
      "mean_token_accuracy": 0.2357799932360649,
      "num_tokens": 11186921.0,
      "step": 4980
    },
    {
      "entropy": 5.3925779342651365,
      "epoch": 5.38939967550027,
      "grad_norm": 1.125,
      "learning_rate": 0.00028666136701634474,
      "loss": 4.7247,
      "mean_token_accuracy": 0.23328425139188766,
      "num_tokens": 11197917.0,
      "step": 4985
    },
    {
      "entropy": 5.390820598602295,
      "epoch": 5.394808004326663,
      "grad_norm": 1.125,
      "learning_rate": 0.0002862330036495942,
      "loss": 4.7491,
      "mean_token_accuracy": 0.24770502597093583,
      "num_tokens": 11209001.0,
      "step": 4990
    },
    {
      "entropy": 5.305437278747559,
      "epoch": 5.400216333153056,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002858045994620825,
      "loss": 4.7515,
      "mean_token_accuracy": 0.23874713629484176,
      "num_tokens": 11220661.0,
      "step": 4995
    },
    {
      "entropy": 5.356938648223877,
      "epoch": 5.405624661979449,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028537615601063133,
      "loss": 4.7452,
      "mean_token_accuracy": 0.23583322763442993,
      "num_tokens": 11232418.0,
      "step": 5000
    },
    {
      "epoch": 5.405624661979449,
      "eval_entropy": 5.27588460445404,
      "eval_loss": 5.827576160430908,
      "eval_mean_token_accuracy": 0.18168690673668275,
      "eval_num_tokens": 11232418.0,
      "eval_runtime": 1.9444,
      "eval_samples_per_second": 1807.711,
      "eval_steps_per_second": 226.285,
      "step": 5000
    },
    {
      "entropy": 5.369883632659912,
      "epoch": 5.411032990805841,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002849476748522052,
      "loss": 4.7065,
      "mean_token_accuracy": 0.2376004308462143,
      "num_tokens": 11244399.0,
      "step": 5005
    },
    {
      "entropy": 5.351139307022095,
      "epoch": 5.416441319632233,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002845191575439057,
      "loss": 4.8049,
      "mean_token_accuracy": 0.237331560254097,
      "num_tokens": 11254958.0,
      "step": 5010
    },
    {
      "entropy": 5.321223545074463,
      "epoch": 5.421849648458626,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002840906056429658,
      "loss": 4.7466,
      "mean_token_accuracy": 0.23710355013608933,
      "num_tokens": 11264806.0,
      "step": 5015
    },
    {
      "entropy": 5.409817218780518,
      "epoch": 5.427257977285019,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002836620207067439,
      "loss": 4.7621,
      "mean_token_accuracy": 0.23184122443199157,
      "num_tokens": 11274636.0,
      "step": 5020
    },
    {
      "entropy": 5.339342641830444,
      "epoch": 5.432666306111411,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028323340429271875,
      "loss": 4.6769,
      "mean_token_accuracy": 0.24240877777338027,
      "num_tokens": 11285401.0,
      "step": 5025
    },
    {
      "entropy": 5.294991779327392,
      "epoch": 5.438074634937804,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002828047579584833,
      "loss": 4.6369,
      "mean_token_accuracy": 0.2497243747115135,
      "num_tokens": 11295526.0,
      "step": 5030
    },
    {
      "entropy": 5.268231391906738,
      "epoch": 5.443482963764197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028237608326173944,
      "loss": 4.7404,
      "mean_token_accuracy": 0.23345786929130555,
      "num_tokens": 11306588.0,
      "step": 5035
    },
    {
      "entropy": 5.375014209747315,
      "epoch": 5.44889129259059,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028194738176029206,
      "loss": 4.7536,
      "mean_token_accuracy": 0.2354530066251755,
      "num_tokens": 11317290.0,
      "step": 5040
    },
    {
      "entropy": 5.4057434558868405,
      "epoch": 5.4542996214169825,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028151865501204316,
      "loss": 4.762,
      "mean_token_accuracy": 0.23712704330682755,
      "num_tokens": 11327289.0,
      "step": 5045
    },
    {
      "entropy": 5.337244510650635,
      "epoch": 5.4597079502433745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028108990457498696,
      "loss": 4.7234,
      "mean_token_accuracy": 0.24120232462882996,
      "num_tokens": 11339031.0,
      "step": 5050
    },
    {
      "entropy": 5.382588529586792,
      "epoch": 5.465116279069767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028066113200720343,
      "loss": 4.7345,
      "mean_token_accuracy": 0.24146860539913179,
      "num_tokens": 11348964.0,
      "step": 5055
    },
    {
      "entropy": 5.349998998641968,
      "epoch": 5.47052460789616,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028023233886685314,
      "loss": 4.7456,
      "mean_token_accuracy": 0.225068399310112,
      "num_tokens": 11360695.0,
      "step": 5060
    },
    {
      "entropy": 5.315986299514771,
      "epoch": 5.475932936722553,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002798035267121713,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23827139139175416,
      "num_tokens": 11371578.0,
      "step": 5065
    },
    {
      "entropy": 5.2930628776550295,
      "epoch": 5.481341265548945,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002793746971014623,
      "loss": 4.6138,
      "mean_token_accuracy": 0.2450634628534317,
      "num_tokens": 11382228.0,
      "step": 5070
    },
    {
      "entropy": 5.408115291595459,
      "epoch": 5.486749594375338,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002789458515930939,
      "loss": 4.79,
      "mean_token_accuracy": 0.23413893580436707,
      "num_tokens": 11393340.0,
      "step": 5075
    },
    {
      "entropy": 5.340760707855225,
      "epoch": 5.492157923201731,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027851699174549164,
      "loss": 4.769,
      "mean_token_accuracy": 0.2381237715482712,
      "num_tokens": 11404301.0,
      "step": 5080
    },
    {
      "entropy": 5.3433544635772705,
      "epoch": 5.497566252028124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002780881191171333,
      "loss": 4.6818,
      "mean_token_accuracy": 0.2449544534087181,
      "num_tokens": 11413859.0,
      "step": 5085
    },
    {
      "entropy": 5.405519723892212,
      "epoch": 5.502974580854516,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027765923526654304,
      "loss": 4.7357,
      "mean_token_accuracy": 0.24649530798196792,
      "num_tokens": 11425067.0,
      "step": 5090
    },
    {
      "entropy": 5.338283395767212,
      "epoch": 5.5083829096809085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027723034175228554,
      "loss": 4.735,
      "mean_token_accuracy": 0.23247045427560806,
      "num_tokens": 11434814.0,
      "step": 5095
    },
    {
      "entropy": 5.307128143310547,
      "epoch": 5.513791238507301,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000276801440132961,
      "loss": 4.7293,
      "mean_token_accuracy": 0.23710498362779617,
      "num_tokens": 11445534.0,
      "step": 5100
    },
    {
      "entropy": 5.323813438415527,
      "epoch": 5.519199567333694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002763725319671989,
      "loss": 4.7534,
      "mean_token_accuracy": 0.23198260962963105,
      "num_tokens": 11456474.0,
      "step": 5105
    },
    {
      "entropy": 5.389151287078858,
      "epoch": 5.524607896160086,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002759436188136525,
      "loss": 4.7411,
      "mean_token_accuracy": 0.23811507523059844,
      "num_tokens": 11467612.0,
      "step": 5110
    },
    {
      "entropy": 5.368999004364014,
      "epoch": 5.530016224986479,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027551470223099306,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23258546143770217,
      "num_tokens": 11479624.0,
      "step": 5115
    },
    {
      "entropy": 5.300816011428833,
      "epoch": 5.535424553812872,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002750857837779045,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2442478433251381,
      "num_tokens": 11489884.0,
      "step": 5120
    },
    {
      "entropy": 5.397455930709839,
      "epoch": 5.540832882639265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002746568650130775,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23023964166641236,
      "num_tokens": 11502168.0,
      "step": 5125
    },
    {
      "entropy": 5.29607400894165,
      "epoch": 5.546241211465657,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002742279474952039,
      "loss": 4.6496,
      "mean_token_accuracy": 0.23967482149600983,
      "num_tokens": 11512809.0,
      "step": 5130
    },
    {
      "entropy": 5.336775207519532,
      "epoch": 5.55164954029205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027379903278297086,
      "loss": 4.7232,
      "mean_token_accuracy": 0.234770068526268,
      "num_tokens": 11523421.0,
      "step": 5135
    },
    {
      "entropy": 5.394512176513672,
      "epoch": 5.5570578691184425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027337012243505536,
      "loss": 4.7935,
      "mean_token_accuracy": 0.23414106369018556,
      "num_tokens": 11535246.0,
      "step": 5140
    },
    {
      "entropy": 5.319738531112671,
      "epoch": 5.562466197944835,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027294121801011887,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23365508466959,
      "num_tokens": 11546043.0,
      "step": 5145
    },
    {
      "entropy": 5.432832431793213,
      "epoch": 5.567874526771227,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027251232106680084,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23289427161216736,
      "num_tokens": 11557330.0,
      "step": 5150
    },
    {
      "entropy": 5.34397873878479,
      "epoch": 5.57328285559762,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027208343316371383,
      "loss": 4.6943,
      "mean_token_accuracy": 0.24126824587583542,
      "num_tokens": 11568519.0,
      "step": 5155
    },
    {
      "entropy": 5.305225896835327,
      "epoch": 5.578691184424013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002716545558594373,
      "loss": 4.6972,
      "mean_token_accuracy": 0.24296617805957793,
      "num_tokens": 11578676.0,
      "step": 5160
    },
    {
      "entropy": 5.307149887084961,
      "epoch": 5.584099513250406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002712256907125126,
      "loss": 4.6372,
      "mean_token_accuracy": 0.2461378961801529,
      "num_tokens": 11590441.0,
      "step": 5165
    },
    {
      "entropy": 5.384063673019409,
      "epoch": 5.589507842076799,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002707968392814366,
      "loss": 4.7748,
      "mean_token_accuracy": 0.23553876131772994,
      "num_tokens": 11601832.0,
      "step": 5170
    },
    {
      "entropy": 5.380595922470093,
      "epoch": 5.594916170903191,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027036800312465635,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23337744772434235,
      "num_tokens": 11612234.0,
      "step": 5175
    },
    {
      "entropy": 5.308310556411743,
      "epoch": 5.600324499729584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002699391838005634,
      "loss": 4.6624,
      "mean_token_accuracy": 0.2455565020442009,
      "num_tokens": 11622282.0,
      "step": 5180
    },
    {
      "entropy": 5.392118120193482,
      "epoch": 5.605732828555976,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026951038286748824,
      "loss": 4.7409,
      "mean_token_accuracy": 0.23500763475894929,
      "num_tokens": 11632778.0,
      "step": 5185
    },
    {
      "entropy": 5.351109027862549,
      "epoch": 5.611141157382368,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002690816018836945,
      "loss": 4.759,
      "mean_token_accuracy": 0.23361084014177322,
      "num_tokens": 11645142.0,
      "step": 5190
    },
    {
      "entropy": 5.2242638111114506,
      "epoch": 5.616549486208761,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002686528424073731,
      "loss": 4.614,
      "mean_token_accuracy": 0.24482941776514053,
      "num_tokens": 11655405.0,
      "step": 5195
    },
    {
      "entropy": 5.251346158981323,
      "epoch": 5.621957815035154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026822410599663704,
      "loss": 4.6813,
      "mean_token_accuracy": 0.23999152481555938,
      "num_tokens": 11665885.0,
      "step": 5200
    },
    {
      "entropy": 5.384830284118652,
      "epoch": 5.627366143861547,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026779539420951554,
      "loss": 4.7976,
      "mean_token_accuracy": 0.2353444814682007,
      "num_tokens": 11677122.0,
      "step": 5205
    },
    {
      "entropy": 5.329274988174438,
      "epoch": 5.63277447268794,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002673667086039481,
      "loss": 4.7153,
      "mean_token_accuracy": 0.23818994760513307,
      "num_tokens": 11687656.0,
      "step": 5210
    },
    {
      "entropy": 5.288289308547974,
      "epoch": 5.638182801514332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002669380507377792,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2326359823346138,
      "num_tokens": 11698341.0,
      "step": 5215
    },
    {
      "entropy": 5.393973970413208,
      "epoch": 5.643591130340725,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002665094221687525,
      "loss": 4.7661,
      "mean_token_accuracy": 0.23271936029195786,
      "num_tokens": 11709730.0,
      "step": 5220
    },
    {
      "entropy": 5.27545690536499,
      "epoch": 5.6489994591671175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002660808244545054,
      "loss": 4.6164,
      "mean_token_accuracy": 0.244342540204525,
      "num_tokens": 11721337.0,
      "step": 5225
    },
    {
      "entropy": 5.275486803054809,
      "epoch": 5.6544077879935095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026565225915256276,
      "loss": 4.74,
      "mean_token_accuracy": 0.2420819193124771,
      "num_tokens": 11733460.0,
      "step": 5230
    },
    {
      "entropy": 5.31084794998169,
      "epoch": 5.659816116819902,
      "grad_norm": 1.09375,
      "learning_rate": 0.000265223727820332,
      "loss": 4.7431,
      "mean_token_accuracy": 0.23815554082393647,
      "num_tokens": 11746077.0,
      "step": 5235
    },
    {
      "entropy": 5.292631340026856,
      "epoch": 5.665224445646295,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002647952320150969,
      "loss": 4.6907,
      "mean_token_accuracy": 0.2377640873193741,
      "num_tokens": 11757477.0,
      "step": 5240
    },
    {
      "entropy": 5.275875759124756,
      "epoch": 5.670632774472688,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002643667732940122,
      "loss": 4.6786,
      "mean_token_accuracy": 0.24668896347284316,
      "num_tokens": 11768388.0,
      "step": 5245
    },
    {
      "entropy": 5.290028381347656,
      "epoch": 5.676041103299081,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026393835321409796,
      "loss": 4.7034,
      "mean_token_accuracy": 0.23661887049674987,
      "num_tokens": 11779254.0,
      "step": 5250
    },
    {
      "entropy": 5.347932195663452,
      "epoch": 5.681449432125473,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026350997333223375,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2427957087755203,
      "num_tokens": 11789871.0,
      "step": 5255
    },
    {
      "entropy": 5.381182289123535,
      "epoch": 5.686857760951866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026308163520515293,
      "loss": 4.8133,
      "mean_token_accuracy": 0.22983981519937516,
      "num_tokens": 11801973.0,
      "step": 5260
    },
    {
      "entropy": 5.347960424423218,
      "epoch": 5.692266089778259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026265334038943735,
      "loss": 4.7992,
      "mean_token_accuracy": 0.23652246594429016,
      "num_tokens": 11813409.0,
      "step": 5265
    },
    {
      "entropy": 5.338179445266723,
      "epoch": 5.6976744186046515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002622250904415113,
      "loss": 4.7053,
      "mean_token_accuracy": 0.24053310304880143,
      "num_tokens": 11825428.0,
      "step": 5270
    },
    {
      "entropy": 5.276244592666626,
      "epoch": 5.7030827474310435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002617968869176362,
      "loss": 4.6852,
      "mean_token_accuracy": 0.2427355855703354,
      "num_tokens": 11838156.0,
      "step": 5275
    },
    {
      "entropy": 5.3246129035949705,
      "epoch": 5.708491076257436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002613687313739044,
      "loss": 4.8002,
      "mean_token_accuracy": 0.22865838259458543,
      "num_tokens": 11850579.0,
      "step": 5280
    },
    {
      "entropy": 5.367587900161743,
      "epoch": 5.713899405083829,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002609406253662344,
      "loss": 4.788,
      "mean_token_accuracy": 0.2355159193277359,
      "num_tokens": 11860862.0,
      "step": 5285
    },
    {
      "entropy": 5.317709493637085,
      "epoch": 5.719307733910222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002605125704503643,
      "loss": 4.7152,
      "mean_token_accuracy": 0.24033275842666627,
      "num_tokens": 11871470.0,
      "step": 5290
    },
    {
      "entropy": 5.341941165924072,
      "epoch": 5.724716062736614,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002600845681818467,
      "loss": 4.8132,
      "mean_token_accuracy": 0.2238249659538269,
      "num_tokens": 11882830.0,
      "step": 5295
    },
    {
      "entropy": 5.392775583267212,
      "epoch": 5.730124391563007,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002596566201160427,
      "loss": 4.8371,
      "mean_token_accuracy": 0.22723516672849656,
      "num_tokens": 11895076.0,
      "step": 5300
    },
    {
      "entropy": 5.412158536911011,
      "epoch": 5.7355327203894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025922872780811664,
      "loss": 4.7645,
      "mean_token_accuracy": 0.23642301112413405,
      "num_tokens": 11907493.0,
      "step": 5305
    },
    {
      "entropy": 5.25253734588623,
      "epoch": 5.740941049215793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025880089281303027,
      "loss": 4.6791,
      "mean_token_accuracy": 0.2304311439394951,
      "num_tokens": 11917624.0,
      "step": 5310
    },
    {
      "entropy": 5.380647802352906,
      "epoch": 5.746349378042185,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025837311668553683,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23607597947120668,
      "num_tokens": 11928791.0,
      "step": 5315
    },
    {
      "entropy": 5.332422494888306,
      "epoch": 5.7517577068685775,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002579454009801757,
      "loss": 4.6612,
      "mean_token_accuracy": 0.24519112706184387,
      "num_tokens": 11939636.0,
      "step": 5320
    },
    {
      "entropy": 5.34787654876709,
      "epoch": 5.75716603569497,
      "grad_norm": 1.25,
      "learning_rate": 0.0002575177472512669,
      "loss": 4.7505,
      "mean_token_accuracy": 0.2352599620819092,
      "num_tokens": 11950273.0,
      "step": 5325
    },
    {
      "entropy": 5.329688024520874,
      "epoch": 5.762574364521363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000257090157052905,
      "loss": 4.7819,
      "mean_token_accuracy": 0.2303329661488533,
      "num_tokens": 11961760.0,
      "step": 5330
    },
    {
      "entropy": 5.415322113037109,
      "epoch": 5.767982693347756,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025666263193895375,
      "loss": 4.7684,
      "mean_token_accuracy": 0.22530437558889388,
      "num_tokens": 11972832.0,
      "step": 5335
    },
    {
      "entropy": 5.3036223411560055,
      "epoch": 5.773391022174148,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025623517346304027,
      "loss": 4.7506,
      "mean_token_accuracy": 0.2413951575756073,
      "num_tokens": 11983858.0,
      "step": 5340
    },
    {
      "entropy": 5.3472034454345705,
      "epoch": 5.778799351000541,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025580778317855,
      "loss": 4.7493,
      "mean_token_accuracy": 0.24133626818656922,
      "num_tokens": 11996368.0,
      "step": 5345
    },
    {
      "entropy": 5.3773518085479735,
      "epoch": 5.784207679826934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002553804626386199,
      "loss": 4.7574,
      "mean_token_accuracy": 0.23588511794805528,
      "num_tokens": 12007870.0,
      "step": 5350
    },
    {
      "entropy": 5.3499915599823,
      "epoch": 5.789616008653326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002549532133961339,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2259849488735199,
      "num_tokens": 12019764.0,
      "step": 5355
    },
    {
      "entropy": 5.303190612792969,
      "epoch": 5.795024337479719,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025452603700371675,
      "loss": 4.6966,
      "mean_token_accuracy": 0.2453092023730278,
      "num_tokens": 12031581.0,
      "step": 5360
    },
    {
      "entropy": 5.348434686660767,
      "epoch": 5.800432666306111,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002540989350137285,
      "loss": 4.8414,
      "mean_token_accuracy": 0.24196318686008453,
      "num_tokens": 12044156.0,
      "step": 5365
    },
    {
      "entropy": 5.334214401245117,
      "epoch": 5.805840995132504,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025367190897825873,
      "loss": 4.6643,
      "mean_token_accuracy": 0.24767277687788009,
      "num_tokens": 12056117.0,
      "step": 5370
    },
    {
      "entropy": 5.447876405715943,
      "epoch": 5.811249323958897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002532449604491209,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2241116926074028,
      "num_tokens": 12068078.0,
      "step": 5375
    },
    {
      "entropy": 5.383710241317749,
      "epoch": 5.816657652785289,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002528180909778471,
      "loss": 4.7376,
      "mean_token_accuracy": 0.24343141168355942,
      "num_tokens": 12078707.0,
      "step": 5380
    },
    {
      "entropy": 5.320490074157715,
      "epoch": 5.822065981611682,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025239130211568206,
      "loss": 4.7359,
      "mean_token_accuracy": 0.2354983314871788,
      "num_tokens": 12089716.0,
      "step": 5385
    },
    {
      "entropy": 5.344113683700561,
      "epoch": 5.827474310438075,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025196459541357725,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23607688248157502,
      "num_tokens": 12099085.0,
      "step": 5390
    },
    {
      "entropy": 5.342487955093384,
      "epoch": 5.832882639264467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025153797242218604,
      "loss": 4.7461,
      "mean_token_accuracy": 0.2342844933271408,
      "num_tokens": 12110689.0,
      "step": 5395
    },
    {
      "entropy": 5.37053165435791,
      "epoch": 5.83829096809086,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025111143469185715,
      "loss": 4.7622,
      "mean_token_accuracy": 0.23243485987186432,
      "num_tokens": 12121179.0,
      "step": 5400
    },
    {
      "entropy": 5.36965045928955,
      "epoch": 5.8436992969172525,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002506849837726299,
      "loss": 4.6884,
      "mean_token_accuracy": 0.24135384410619737,
      "num_tokens": 12133026.0,
      "step": 5405
    },
    {
      "entropy": 5.315388584136963,
      "epoch": 5.849107625743645,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002502586212142278,
      "loss": 4.7697,
      "mean_token_accuracy": 0.233494308590889,
      "num_tokens": 12143858.0,
      "step": 5410
    },
    {
      "entropy": 5.290375661849976,
      "epoch": 5.854515954570038,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024983234856605337,
      "loss": 4.7022,
      "mean_token_accuracy": 0.23650214225053787,
      "num_tokens": 12154698.0,
      "step": 5415
    },
    {
      "entropy": 5.2817223072052,
      "epoch": 5.85992428339643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024940616737718235,
      "loss": 4.6989,
      "mean_token_accuracy": 0.23575526773929595,
      "num_tokens": 12165657.0,
      "step": 5420
    },
    {
      "entropy": 5.283653688430786,
      "epoch": 5.865332612222823,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002489800791963583,
      "loss": 4.7645,
      "mean_token_accuracy": 0.22857791632413865,
      "num_tokens": 12176737.0,
      "step": 5425
    },
    {
      "entropy": 5.389004039764404,
      "epoch": 5.870740941049216,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024855408557198657,
      "loss": 4.8045,
      "mean_token_accuracy": 0.22992659509181976,
      "num_tokens": 12188216.0,
      "step": 5430
    },
    {
      "entropy": 5.3752086639404295,
      "epoch": 5.876149269875609,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000248128188052129,
      "loss": 4.7481,
      "mean_token_accuracy": 0.23514376878738402,
      "num_tokens": 12198666.0,
      "step": 5435
    },
    {
      "entropy": 5.253864479064942,
      "epoch": 5.881557598702001,
      "grad_norm": 1.140625,
      "learning_rate": 0.00024770238818449804,
      "loss": 4.6962,
      "mean_token_accuracy": 0.2347796842455864,
      "num_tokens": 12210640.0,
      "step": 5440
    },
    {
      "entropy": 5.2314534187316895,
      "epoch": 5.886965927528394,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024727668751645164,
      "loss": 4.6885,
      "mean_token_accuracy": 0.23575307428836823,
      "num_tokens": 12221878.0,
      "step": 5445
    },
    {
      "entropy": 5.340726566314697,
      "epoch": 5.8923742563547865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002468510875949868,
      "loss": 4.6571,
      "mean_token_accuracy": 0.24513103812932968,
      "num_tokens": 12232803.0,
      "step": 5450
    },
    {
      "entropy": 5.367942380905151,
      "epoch": 5.897782585181179,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024642558996673474,
      "loss": 4.7034,
      "mean_token_accuracy": 0.23816127181053162,
      "num_tokens": 12244423.0,
      "step": 5455
    },
    {
      "entropy": 5.38005485534668,
      "epoch": 5.903190914007571,
      "grad_norm": 1.125,
      "learning_rate": 0.00024600019617795465,
      "loss": 4.7358,
      "mean_token_accuracy": 0.2393931642174721,
      "num_tokens": 12256724.0,
      "step": 5460
    },
    {
      "entropy": 5.369742012023925,
      "epoch": 5.908599242833964,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002455749077745288,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23098594099283218,
      "num_tokens": 12267229.0,
      "step": 5465
    },
    {
      "entropy": 5.285031032562256,
      "epoch": 5.914007571660357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024514972630195603,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2427264392375946,
      "num_tokens": 12278370.0,
      "step": 5470
    },
    {
      "entropy": 5.419775724411011,
      "epoch": 5.91941590048675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002447246533053469,
      "loss": 4.7008,
      "mean_token_accuracy": 0.23430182337760924,
      "num_tokens": 12289828.0,
      "step": 5475
    },
    {
      "entropy": 5.32359471321106,
      "epoch": 5.924824229313142,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024429969032941756,
      "loss": 4.7459,
      "mean_token_accuracy": 0.2318441465497017,
      "num_tokens": 12300361.0,
      "step": 5480
    },
    {
      "entropy": 5.3055100440979,
      "epoch": 5.930232558139535,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024387483891848456,
      "loss": 4.6695,
      "mean_token_accuracy": 0.23837276101112365,
      "num_tokens": 12309892.0,
      "step": 5485
    },
    {
      "entropy": 5.245346117019653,
      "epoch": 5.935640886965928,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024345010061645889,
      "loss": 4.6333,
      "mean_token_accuracy": 0.24718346446752548,
      "num_tokens": 12321625.0,
      "step": 5490
    },
    {
      "entropy": 5.340652227401733,
      "epoch": 5.9410492157923205,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024302547696684047,
      "loss": 4.7066,
      "mean_token_accuracy": 0.24184146225452424,
      "num_tokens": 12333492.0,
      "step": 5495
    },
    {
      "entropy": 5.310504198074341,
      "epoch": 5.9464575446187125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00024260096951271253,
      "loss": 4.6995,
      "mean_token_accuracy": 0.24560317248106003,
      "num_tokens": 12344803.0,
      "step": 5500
    },
    {
      "entropy": 5.317621660232544,
      "epoch": 5.951865873445105,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024217657979673634,
      "loss": 4.789,
      "mean_token_accuracy": 0.22861287742853165,
      "num_tokens": 12355956.0,
      "step": 5505
    },
    {
      "entropy": 5.369316482543946,
      "epoch": 5.957274202271498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024175230936114507,
      "loss": 4.8267,
      "mean_token_accuracy": 0.23064886182546615,
      "num_tokens": 12368706.0,
      "step": 5510
    },
    {
      "entropy": 5.296198606491089,
      "epoch": 5.962682531097891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002413281597477384,
      "loss": 4.645,
      "mean_token_accuracy": 0.24565879702568055,
      "num_tokens": 12380737.0,
      "step": 5515
    },
    {
      "entropy": 5.253948307037353,
      "epoch": 5.968090859924283,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024090413249787696,
      "loss": 4.6571,
      "mean_token_accuracy": 0.24345935732126237,
      "num_tokens": 12391147.0,
      "step": 5520
    },
    {
      "entropy": 5.391175651550293,
      "epoch": 5.973499188750676,
      "grad_norm": 1.125,
      "learning_rate": 0.0002404802291524768,
      "loss": 4.7909,
      "mean_token_accuracy": 0.24234597384929657,
      "num_tokens": 12402922.0,
      "step": 5525
    },
    {
      "entropy": 5.29814133644104,
      "epoch": 5.978907517577069,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024005645125200375,
      "loss": 4.7282,
      "mean_token_accuracy": 0.2417369470000267,
      "num_tokens": 12413958.0,
      "step": 5530
    },
    {
      "entropy": 5.387341737747192,
      "epoch": 5.984315846403462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002396328003364675,
      "loss": 4.7492,
      "mean_token_accuracy": 0.23815809041261674,
      "num_tokens": 12424747.0,
      "step": 5535
    },
    {
      "entropy": 5.402175951004028,
      "epoch": 5.9897241752298545,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023920927794541665,
      "loss": 4.6662,
      "mean_token_accuracy": 0.25058837831020353,
      "num_tokens": 12437018.0,
      "step": 5540
    },
    {
      "entropy": 5.367151546478271,
      "epoch": 5.995132504056246,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002387858856179324,
      "loss": 4.795,
      "mean_token_accuracy": 0.2240246281027794,
      "num_tokens": 12449620.0,
      "step": 5545
    },
    {
      "entropy": 5.348214997185601,
      "epoch": 6.0,
      "grad_norm": 2.15625,
      "learning_rate": 0.00023836262489262361,
      "loss": 4.7611,
      "mean_token_accuracy": 0.23656969765822092,
      "num_tokens": 12461334.0,
      "step": 5550
    },
    {
      "entropy": 5.37571349143982,
      "epoch": 6.005408328826393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023793949730762049,
      "loss": 4.5984,
      "mean_token_accuracy": 0.24920047372579573,
      "num_tokens": 12471955.0,
      "step": 5555
    },
    {
      "entropy": 5.362363719940186,
      "epoch": 6.010816657652786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023751650440056982,
      "loss": 4.6019,
      "mean_token_accuracy": 0.2562414988875389,
      "num_tokens": 12482854.0,
      "step": 5560
    },
    {
      "entropy": 5.278171014785767,
      "epoch": 6.016224986479178,
      "grad_norm": 1.25,
      "learning_rate": 0.00023709364770862885,
      "loss": 4.495,
      "mean_token_accuracy": 0.25104211270809174,
      "num_tokens": 12493336.0,
      "step": 5565
    },
    {
      "entropy": 5.302797651290893,
      "epoch": 6.0216333153055706,
      "grad_norm": 1.125,
      "learning_rate": 0.00023667092876845976,
      "loss": 4.5982,
      "mean_token_accuracy": 0.24976315796375276,
      "num_tokens": 12505384.0,
      "step": 5570
    },
    {
      "entropy": 5.337259721755982,
      "epoch": 6.027041644131963,
      "grad_norm": 1.25,
      "learning_rate": 0.0002362483491162239,
      "loss": 4.5264,
      "mean_token_accuracy": 0.25760655850172043,
      "num_tokens": 12515838.0,
      "step": 5575
    },
    {
      "entropy": 5.418482542037964,
      "epoch": 6.032449972958356,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023582591028757706,
      "loss": 4.6364,
      "mean_token_accuracy": 0.24380246102809905,
      "num_tokens": 12527748.0,
      "step": 5580
    },
    {
      "entropy": 5.30024995803833,
      "epoch": 6.037858301784748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023540361381766274,
      "loss": 4.5048,
      "mean_token_accuracy": 0.2591186538338661,
      "num_tokens": 12540401.0,
      "step": 5585
    },
    {
      "entropy": 5.280279302597046,
      "epoch": 6.043266630611141,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002349814612411073,
      "loss": 4.516,
      "mean_token_accuracy": 0.2543742373585701,
      "num_tokens": 12551081.0,
      "step": 5590
    },
    {
      "entropy": 5.289242696762085,
      "epoch": 6.048674959437534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023455945409201411,
      "loss": 4.6095,
      "mean_token_accuracy": 0.2494116574525833,
      "num_tokens": 12563749.0,
      "step": 5595
    },
    {
      "entropy": 5.391827154159546,
      "epoch": 6.054083288263927,
      "grad_norm": 1.140625,
      "learning_rate": 0.00023413759390395828,
      "loss": 4.6328,
      "mean_token_accuracy": 0.2509587720036507,
      "num_tokens": 12575435.0,
      "step": 5600
    },
    {
      "entropy": 5.349138832092285,
      "epoch": 6.059491617090319,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002337158822099807,
      "loss": 4.5183,
      "mean_token_accuracy": 0.2597142681479454,
      "num_tokens": 12586713.0,
      "step": 5605
    },
    {
      "entropy": 5.230661058425904,
      "epoch": 6.064899945916712,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002332943205425825,
      "loss": 4.5279,
      "mean_token_accuracy": 0.25923209488391874,
      "num_tokens": 12598598.0,
      "step": 5610
    },
    {
      "entropy": 5.364458608627319,
      "epoch": 6.0703082747431045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023287291043371978,
      "loss": 4.6682,
      "mean_token_accuracy": 0.24494118988513947,
      "num_tokens": 12609638.0,
      "step": 5615
    },
    {
      "entropy": 5.40669813156128,
      "epoch": 6.075716603569497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023245165341479796,
      "loss": 4.5948,
      "mean_token_accuracy": 0.25283135026693343,
      "num_tokens": 12621316.0,
      "step": 5620
    },
    {
      "entropy": 5.360855770111084,
      "epoch": 6.081124932395889,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023203055101666593,
      "loss": 4.5971,
      "mean_token_accuracy": 0.24927141964435579,
      "num_tokens": 12631374.0,
      "step": 5625
    },
    {
      "entropy": 5.209184837341309,
      "epoch": 6.086533261222282,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002316096047696108,
      "loss": 4.4509,
      "mean_token_accuracy": 0.26711108088493346,
      "num_tokens": 12642520.0,
      "step": 5630
    },
    {
      "entropy": 5.269736576080322,
      "epoch": 6.091941590048675,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002311888162033522,
      "loss": 4.5586,
      "mean_token_accuracy": 0.254153610765934,
      "num_tokens": 12653410.0,
      "step": 5635
    },
    {
      "entropy": 5.290821886062622,
      "epoch": 6.097349918875068,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023076818684703682,
      "loss": 4.5543,
      "mean_token_accuracy": 0.2537793397903442,
      "num_tokens": 12663979.0,
      "step": 5640
    },
    {
      "entropy": 5.349629259109497,
      "epoch": 6.10275824770146,
      "grad_norm": 1.15625,
      "learning_rate": 0.00023034771822923268,
      "loss": 4.5895,
      "mean_token_accuracy": 0.25114935636520386,
      "num_tokens": 12673970.0,
      "step": 5645
    },
    {
      "entropy": 5.311895561218262,
      "epoch": 6.108166576527853,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002299274118779236,
      "loss": 4.5595,
      "mean_token_accuracy": 0.2523087546229362,
      "num_tokens": 12685817.0,
      "step": 5650
    },
    {
      "entropy": 5.315186452865601,
      "epoch": 6.113574905354246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022950726932050386,
      "loss": 4.6641,
      "mean_token_accuracy": 0.24742500185966493,
      "num_tokens": 12697796.0,
      "step": 5655
    },
    {
      "entropy": 5.288755750656128,
      "epoch": 6.1189832341806385,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022908729208377267,
      "loss": 4.5461,
      "mean_token_accuracy": 0.25241935104131696,
      "num_tokens": 12709071.0,
      "step": 5660
    },
    {
      "entropy": 5.27881441116333,
      "epoch": 6.1243915630070305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00022866748169392815,
      "loss": 4.5446,
      "mean_token_accuracy": 0.24807569831609727,
      "num_tokens": 12720763.0,
      "step": 5665
    },
    {
      "entropy": 5.255998992919922,
      "epoch": 6.129799891833423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022824783967656208,
      "loss": 4.5136,
      "mean_token_accuracy": 0.2714682534337044,
      "num_tokens": 12732095.0,
      "step": 5670
    },
    {
      "entropy": 5.3379992008209225,
      "epoch": 6.135208220659816,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022782836755665475,
      "loss": 4.6546,
      "mean_token_accuracy": 0.24739901721477509,
      "num_tokens": 12742056.0,
      "step": 5675
    },
    {
      "entropy": 5.284276342391967,
      "epoch": 6.140616549486209,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00022740906685856876,
      "loss": 4.5333,
      "mean_token_accuracy": 0.25044632256031035,
      "num_tokens": 12753653.0,
      "step": 5680
    },
    {
      "entropy": 5.3539103984832765,
      "epoch": 6.146024878312601,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022698993910604383,
      "loss": 4.587,
      "mean_token_accuracy": 0.25243034958839417,
      "num_tokens": 12764733.0,
      "step": 5685
    },
    {
      "entropy": 5.23184609413147,
      "epoch": 6.151433207138994,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000226570985822191,
      "loss": 4.5161,
      "mean_token_accuracy": 0.26145158559083936,
      "num_tokens": 12776747.0,
      "step": 5690
    },
    {
      "entropy": 5.257935285568237,
      "epoch": 6.156841535965387,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022615220852948765,
      "loss": 4.4723,
      "mean_token_accuracy": 0.25020534098148345,
      "num_tokens": 12786802.0,
      "step": 5695
    },
    {
      "entropy": 5.299817228317261,
      "epoch": 6.16224986479178,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022573360874977145,
      "loss": 4.5585,
      "mean_token_accuracy": 0.2515297636389732,
      "num_tokens": 12796866.0,
      "step": 5700
    },
    {
      "entropy": 5.255823850631714,
      "epoch": 6.167658193618172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022531518800423484,
      "loss": 4.526,
      "mean_token_accuracy": 0.25243674516677855,
      "num_tokens": 12807803.0,
      "step": 5705
    },
    {
      "entropy": 5.287596797943115,
      "epoch": 6.173066522444564,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002248969478134199,
      "loss": 4.6211,
      "mean_token_accuracy": 0.2515898197889328,
      "num_tokens": 12818357.0,
      "step": 5710
    },
    {
      "entropy": 5.305178022384643,
      "epoch": 6.178474851270957,
      "grad_norm": 1.046875,
      "learning_rate": 0.00022447888969721232,
      "loss": 4.6333,
      "mean_token_accuracy": 0.25456716269254687,
      "num_tokens": 12830798.0,
      "step": 5715
    },
    {
      "entropy": 5.294315195083618,
      "epoch": 6.18388318009735,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022406101517483636,
      "loss": 4.5522,
      "mean_token_accuracy": 0.25335372537374495,
      "num_tokens": 12842424.0,
      "step": 5720
    },
    {
      "entropy": 5.343192720413208,
      "epoch": 6.189291508923742,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022364332576484897,
      "loss": 4.5722,
      "mean_token_accuracy": 0.25098187774419783,
      "num_tokens": 12854468.0,
      "step": 5725
    },
    {
      "entropy": 5.3400555610656735,
      "epoch": 6.194699837750135,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002232258229851345,
      "loss": 4.5508,
      "mean_token_accuracy": 0.2592270910739899,
      "num_tokens": 12866054.0,
      "step": 5730
    },
    {
      "entropy": 5.292687034606933,
      "epoch": 6.200108166576528,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022280850835289885,
      "loss": 4.6011,
      "mean_token_accuracy": 0.250145810842514,
      "num_tokens": 12876612.0,
      "step": 5735
    },
    {
      "entropy": 5.265779209136963,
      "epoch": 6.205516495402921,
      "grad_norm": 1.1875,
      "learning_rate": 0.00022239138338466452,
      "loss": 4.5347,
      "mean_token_accuracy": 0.2543587848544121,
      "num_tokens": 12887762.0,
      "step": 5740
    },
    {
      "entropy": 5.292228126525879,
      "epoch": 6.210924824229313,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022197444959626451,
      "loss": 4.5591,
      "mean_token_accuracy": 0.25506618022918703,
      "num_tokens": 12899110.0,
      "step": 5745
    },
    {
      "entropy": 5.261582756042481,
      "epoch": 6.2163331530557056,
      "grad_norm": 1.015625,
      "learning_rate": 0.00022155770850283723,
      "loss": 4.5371,
      "mean_token_accuracy": 0.25177046060562136,
      "num_tokens": 12912136.0,
      "step": 5750
    },
    {
      "entropy": 5.279011392593384,
      "epoch": 6.221741481882098,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022114116161882065,
      "loss": 4.5673,
      "mean_token_accuracy": 0.251338429749012,
      "num_tokens": 12922902.0,
      "step": 5755
    },
    {
      "entropy": 5.268643856048584,
      "epoch": 6.227149810708491,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022072481045794733,
      "loss": 4.4968,
      "mean_token_accuracy": 0.25733707249164584,
      "num_tokens": 12933831.0,
      "step": 5760
    },
    {
      "entropy": 5.331746530532837,
      "epoch": 6.232558139534884,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022030865653323823,
      "loss": 4.5435,
      "mean_token_accuracy": 0.2539557427167892,
      "num_tokens": 12945299.0,
      "step": 5765
    },
    {
      "entropy": 5.250093650817871,
      "epoch": 6.237966468361276,
      "grad_norm": 1.25,
      "learning_rate": 0.0002198927013569977,
      "loss": 4.5336,
      "mean_token_accuracy": 0.26108190715312957,
      "num_tokens": 12955932.0,
      "step": 5770
    },
    {
      "entropy": 5.241962051391601,
      "epoch": 6.243374797187669,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002194769464408076,
      "loss": 4.5509,
      "mean_token_accuracy": 0.24996897876262664,
      "num_tokens": 12965346.0,
      "step": 5775
    },
    {
      "entropy": 5.2777721881866455,
      "epoch": 6.248783126014062,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002190613932955226,
      "loss": 4.573,
      "mean_token_accuracy": 0.2455021157860756,
      "num_tokens": 12977164.0,
      "step": 5780
    },
    {
      "entropy": 5.354558277130127,
      "epoch": 6.254191454840455,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021864604343126365,
      "loss": 4.5994,
      "mean_token_accuracy": 0.24705723226070403,
      "num_tokens": 12987992.0,
      "step": 5785
    },
    {
      "entropy": 5.289169263839722,
      "epoch": 6.259599783666847,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021823089835741312,
      "loss": 4.6184,
      "mean_token_accuracy": 0.2465651050209999,
      "num_tokens": 12999321.0,
      "step": 5790
    },
    {
      "entropy": 5.286135244369507,
      "epoch": 6.2650081124932395,
      "grad_norm": 1.203125,
      "learning_rate": 0.000217815959582609,
      "loss": 4.5377,
      "mean_token_accuracy": 0.2561722308397293,
      "num_tokens": 13010324.0,
      "step": 5795
    },
    {
      "entropy": 5.210115528106689,
      "epoch": 6.270416441319632,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021740122861473992,
      "loss": 4.5651,
      "mean_token_accuracy": 0.2520148456096649,
      "num_tokens": 13021277.0,
      "step": 5800
    },
    {
      "entropy": 5.334791564941407,
      "epoch": 6.275824770146025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021698670696093932,
      "loss": 4.5385,
      "mean_token_accuracy": 0.2583196133375168,
      "num_tokens": 13032995.0,
      "step": 5805
    },
    {
      "entropy": 5.372949838638306,
      "epoch": 6.281233098972417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021657239612757963,
      "loss": 4.5688,
      "mean_token_accuracy": 0.2507487341761589,
      "num_tokens": 13043696.0,
      "step": 5810
    },
    {
      "entropy": 5.2793067455291744,
      "epoch": 6.28664142779881,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021615829762026734,
      "loss": 4.6105,
      "mean_token_accuracy": 0.2511099621653557,
      "num_tokens": 13056526.0,
      "step": 5815
    },
    {
      "entropy": 5.294821691513062,
      "epoch": 6.292049756625203,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021574441294383752,
      "loss": 4.6084,
      "mean_token_accuracy": 0.2471725881099701,
      "num_tokens": 13067820.0,
      "step": 5820
    },
    {
      "entropy": 5.234441184997559,
      "epoch": 6.297458085451596,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021533074360234806,
      "loss": 4.5425,
      "mean_token_accuracy": 0.25148710906505584,
      "num_tokens": 13078798.0,
      "step": 5825
    },
    {
      "entropy": 5.3169139385223385,
      "epoch": 6.302866414277988,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021491729109907414,
      "loss": 4.5384,
      "mean_token_accuracy": 0.2610315665602684,
      "num_tokens": 13090104.0,
      "step": 5830
    },
    {
      "entropy": 5.360798788070679,
      "epoch": 6.308274743104381,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021450405693650333,
      "loss": 4.6188,
      "mean_token_accuracy": 0.24809301346540452,
      "num_tokens": 13101905.0,
      "step": 5835
    },
    {
      "entropy": 5.312351894378662,
      "epoch": 6.3136830719307735,
      "grad_norm": 1.25,
      "learning_rate": 0.00021409104261632932,
      "loss": 4.5652,
      "mean_token_accuracy": 0.24606993645429612,
      "num_tokens": 13113275.0,
      "step": 5840
    },
    {
      "entropy": 5.267687368392944,
      "epoch": 6.319091400757166,
      "grad_norm": 1.1875,
      "learning_rate": 0.00021367824963944727,
      "loss": 4.6543,
      "mean_token_accuracy": 0.24162089973688125,
      "num_tokens": 13124531.0,
      "step": 5845
    },
    {
      "entropy": 5.366695308685303,
      "epoch": 6.324499729583558,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021326567950594766,
      "loss": 4.6896,
      "mean_token_accuracy": 0.24018312990665436,
      "num_tokens": 13135324.0,
      "step": 5850
    },
    {
      "entropy": 5.347139453887939,
      "epoch": 6.329908058409951,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002128533337151114,
      "loss": 4.5715,
      "mean_token_accuracy": 0.25151847302913666,
      "num_tokens": 13145323.0,
      "step": 5855
    },
    {
      "entropy": 5.257876968383789,
      "epoch": 6.335316387236344,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002124412137654041,
      "loss": 4.5986,
      "mean_token_accuracy": 0.24776281416416168,
      "num_tokens": 13157294.0,
      "step": 5860
    },
    {
      "entropy": 5.173339176177978,
      "epoch": 6.340724716062737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021202932115447056,
      "loss": 4.5002,
      "mean_token_accuracy": 0.2550232321023941,
      "num_tokens": 13167576.0,
      "step": 5865
    },
    {
      "entropy": 5.3480432510375975,
      "epoch": 6.346133044889129,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021161765737912937,
      "loss": 4.639,
      "mean_token_accuracy": 0.2464173212647438,
      "num_tokens": 13180326.0,
      "step": 5870
    },
    {
      "entropy": 5.306275939941406,
      "epoch": 6.351541373715522,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021120622393536775,
      "loss": 4.5931,
      "mean_token_accuracy": 0.24826602935791015,
      "num_tokens": 13190707.0,
      "step": 5875
    },
    {
      "entropy": 5.26748423576355,
      "epoch": 6.356949702541915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021079502231833585,
      "loss": 4.5459,
      "mean_token_accuracy": 0.26566562205553057,
      "num_tokens": 13203902.0,
      "step": 5880
    },
    {
      "entropy": 5.311595678329468,
      "epoch": 6.3623580313683075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021038405402234116,
      "loss": 4.5721,
      "mean_token_accuracy": 0.25951351672410966,
      "num_tokens": 13214192.0,
      "step": 5885
    },
    {
      "entropy": 5.269037389755249,
      "epoch": 6.367766360194699,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002099733205408435,
      "loss": 4.5691,
      "mean_token_accuracy": 0.24761241525411606,
      "num_tokens": 13225525.0,
      "step": 5890
    },
    {
      "entropy": 5.235737752914429,
      "epoch": 6.373174689021092,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002095628233664492,
      "loss": 4.5643,
      "mean_token_accuracy": 0.24890918880701066,
      "num_tokens": 13237406.0,
      "step": 5895
    },
    {
      "entropy": 5.347858858108521,
      "epoch": 6.378583017847485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002091525639909061,
      "loss": 4.5738,
      "mean_token_accuracy": 0.25947401821613314,
      "num_tokens": 13249376.0,
      "step": 5900
    },
    {
      "entropy": 5.254463720321655,
      "epoch": 6.383991346673878,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020874254390509757,
      "loss": 4.579,
      "mean_token_accuracy": 0.2501704037189484,
      "num_tokens": 13259687.0,
      "step": 5905
    },
    {
      "entropy": 5.2755491733551025,
      "epoch": 6.38939967550027,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020833276459903766,
      "loss": 4.555,
      "mean_token_accuracy": 0.24834691733121872,
      "num_tokens": 13271834.0,
      "step": 5910
    },
    {
      "entropy": 5.3026878356933596,
      "epoch": 6.394808004326663,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002079232275618651,
      "loss": 4.6441,
      "mean_token_accuracy": 0.24866369515657424,
      "num_tokens": 13283910.0,
      "step": 5915
    },
    {
      "entropy": 5.345313024520874,
      "epoch": 6.400216333153056,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002075139342818388,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2535760045051575,
      "num_tokens": 13294421.0,
      "step": 5920
    },
    {
      "entropy": 5.371544265747071,
      "epoch": 6.405624661979449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020710488624633118,
      "loss": 4.6352,
      "mean_token_accuracy": 0.2537995457649231,
      "num_tokens": 13304629.0,
      "step": 5925
    },
    {
      "entropy": 5.403523635864258,
      "epoch": 6.411032990805841,
      "grad_norm": 1.09375,
      "learning_rate": 0.00020669608494182384,
      "loss": 4.6767,
      "mean_token_accuracy": 0.24457396119832991,
      "num_tokens": 13315704.0,
      "step": 5930
    },
    {
      "entropy": 5.344007968902588,
      "epoch": 6.416441319632233,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020628753185390164,
      "loss": 4.6486,
      "mean_token_accuracy": 0.24794667810201645,
      "num_tokens": 13326079.0,
      "step": 5935
    },
    {
      "entropy": 5.337772035598755,
      "epoch": 6.421849648458626,
      "grad_norm": 1.125,
      "learning_rate": 0.0002058792284672475,
      "loss": 4.594,
      "mean_token_accuracy": 0.25287126898765566,
      "num_tokens": 13336561.0,
      "step": 5940
    },
    {
      "entropy": 5.30888557434082,
      "epoch": 6.427257977285019,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002054711762656369,
      "loss": 4.6604,
      "mean_token_accuracy": 0.23571989834308624,
      "num_tokens": 13347465.0,
      "step": 5945
    },
    {
      "entropy": 5.340973329544068,
      "epoch": 6.432666306111411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00020506337673193236,
      "loss": 4.6154,
      "mean_token_accuracy": 0.2540865257382393,
      "num_tokens": 13358728.0,
      "step": 5950
    },
    {
      "entropy": 5.338106346130371,
      "epoch": 6.438074634937804,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020465583134807836,
      "loss": 4.624,
      "mean_token_accuracy": 0.2462385207414627,
      "num_tokens": 13370407.0,
      "step": 5955
    },
    {
      "entropy": 5.338262748718262,
      "epoch": 6.443482963764197,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00020424854159509587,
      "loss": 4.5722,
      "mean_token_accuracy": 0.25845286101102827,
      "num_tokens": 13382563.0,
      "step": 5960
    },
    {
      "entropy": 5.304811429977417,
      "epoch": 6.44889129259059,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002038415089530767,
      "loss": 4.6526,
      "mean_token_accuracy": 0.2471668913960457,
      "num_tokens": 13394623.0,
      "step": 5965
    },
    {
      "entropy": 5.2662702083587645,
      "epoch": 6.4542996214169825,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020343473490117846,
      "loss": 4.566,
      "mean_token_accuracy": 0.25581531673669816,
      "num_tokens": 13406565.0,
      "step": 5970
    },
    {
      "entropy": 5.277949333190918,
      "epoch": 6.4597079502433745,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020302822091761884,
      "loss": 4.6234,
      "mean_token_accuracy": 0.24462800920009614,
      "num_tokens": 13418228.0,
      "step": 5975
    },
    {
      "entropy": 5.276081228256226,
      "epoch": 6.465116279069767,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020262196847967091,
      "loss": 4.5299,
      "mean_token_accuracy": 0.2561643585562706,
      "num_tokens": 13428654.0,
      "step": 5980
    },
    {
      "entropy": 5.371704483032227,
      "epoch": 6.47052460789616,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020221597906365667,
      "loss": 4.6779,
      "mean_token_accuracy": 0.2517550319433212,
      "num_tokens": 13439918.0,
      "step": 5985
    },
    {
      "entropy": 5.316441822052002,
      "epoch": 6.475932936722553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00020181025414494285,
      "loss": 4.6616,
      "mean_token_accuracy": 0.24777029305696488,
      "num_tokens": 13452241.0,
      "step": 5990
    },
    {
      "entropy": 5.357630586624145,
      "epoch": 6.481341265548945,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002014047951979344,
      "loss": 4.6251,
      "mean_token_accuracy": 0.2510747790336609,
      "num_tokens": 13463463.0,
      "step": 5995
    },
    {
      "entropy": 5.245645809173584,
      "epoch": 6.486749594375338,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002009996036960704,
      "loss": 4.5428,
      "mean_token_accuracy": 0.25608101636171343,
      "num_tokens": 13474612.0,
      "step": 6000
    },
    {
      "epoch": 6.486749594375338,
      "eval_entropy": 5.215320254455913,
      "eval_loss": 5.834592819213867,
      "eval_mean_token_accuracy": 0.1832794749262658,
      "eval_num_tokens": 13474612.0,
      "eval_runtime": 1.9448,
      "eval_samples_per_second": 1807.341,
      "eval_steps_per_second": 226.239,
      "step": 6000
    },
    {
      "entropy": 5.2015345096588135,
      "epoch": 6.492157923201731,
      "grad_norm": 1.234375,
      "learning_rate": 0.00020059468111181757,
      "loss": 4.5125,
      "mean_token_accuracy": 0.2505683571100235,
      "num_tokens": 13485499.0,
      "step": 6005
    },
    {
      "entropy": 5.246391820907593,
      "epoch": 6.497566252028124,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020019002891666555,
      "loss": 4.6103,
      "mean_token_accuracy": 0.24595464617013932,
      "num_tokens": 13496328.0,
      "step": 6010
    },
    {
      "entropy": 5.323586130142212,
      "epoch": 6.502974580854516,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019978564858112125,
      "loss": 4.6637,
      "mean_token_accuracy": 0.2476791635155678,
      "num_tokens": 13508996.0,
      "step": 6015
    },
    {
      "entropy": 5.342580080032349,
      "epoch": 6.5083829096809085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019938154157470378,
      "loss": 4.6234,
      "mean_token_accuracy": 0.24441204220056534,
      "num_tokens": 13520111.0,
      "step": 6020
    },
    {
      "entropy": 5.376033353805542,
      "epoch": 6.513791238507301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019897770936593912,
      "loss": 4.6653,
      "mean_token_accuracy": 0.23879428952932358,
      "num_tokens": 13532271.0,
      "step": 6025
    },
    {
      "entropy": 5.305279493331909,
      "epoch": 6.519199567333694,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001985741534223543,
      "loss": 4.566,
      "mean_token_accuracy": 0.2562160164117813,
      "num_tokens": 13543757.0,
      "step": 6030
    },
    {
      "entropy": 5.297739410400391,
      "epoch": 6.524607896160086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019817087521047244,
      "loss": 4.6334,
      "mean_token_accuracy": 0.25294891744852066,
      "num_tokens": 13555484.0,
      "step": 6035
    },
    {
      "entropy": 5.314838790893555,
      "epoch": 6.530016224986479,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019776787619580767,
      "loss": 4.6443,
      "mean_token_accuracy": 0.2456320196390152,
      "num_tokens": 13566864.0,
      "step": 6040
    },
    {
      "entropy": 5.355596828460693,
      "epoch": 6.535424553812872,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001973651578428593,
      "loss": 4.5526,
      "mean_token_accuracy": 0.2595892041921616,
      "num_tokens": 13577497.0,
      "step": 6045
    },
    {
      "entropy": 5.247674751281738,
      "epoch": 6.540832882639265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019696272161510663,
      "loss": 4.5649,
      "mean_token_accuracy": 0.2531619071960449,
      "num_tokens": 13589714.0,
      "step": 6050
    },
    {
      "entropy": 5.283963632583618,
      "epoch": 6.546241211465657,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019656056897500408,
      "loss": 4.5691,
      "mean_token_accuracy": 0.24599054157733918,
      "num_tokens": 13600805.0,
      "step": 6055
    },
    {
      "entropy": 5.326214981079102,
      "epoch": 6.55164954029205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019615870138397506,
      "loss": 4.6377,
      "mean_token_accuracy": 0.2401846244931221,
      "num_tokens": 13613160.0,
      "step": 6060
    },
    {
      "entropy": 5.260635471343994,
      "epoch": 6.5570578691184425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001957571203024075,
      "loss": 4.5635,
      "mean_token_accuracy": 0.2505684420466423,
      "num_tokens": 13623023.0,
      "step": 6065
    },
    {
      "entropy": 5.306850481033325,
      "epoch": 6.562466197944835,
      "grad_norm": 1.1875,
      "learning_rate": 0.00019535582718964773,
      "loss": 4.6372,
      "mean_token_accuracy": 0.2545121371746063,
      "num_tokens": 13634416.0,
      "step": 6070
    },
    {
      "entropy": 5.3699287414550785,
      "epoch": 6.567874526771227,
      "grad_norm": 1.25,
      "learning_rate": 0.00019495482350399612,
      "loss": 4.6351,
      "mean_token_accuracy": 0.2515858471393585,
      "num_tokens": 13645123.0,
      "step": 6075
    },
    {
      "entropy": 5.261383581161499,
      "epoch": 6.57328285559762,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001945541107027009,
      "loss": 4.5388,
      "mean_token_accuracy": 0.25285120606422423,
      "num_tokens": 13657279.0,
      "step": 6080
    },
    {
      "entropy": 5.3344128131866455,
      "epoch": 6.578691184424013,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001941536902419533,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2479914665222168,
      "num_tokens": 13668142.0,
      "step": 6085
    },
    {
      "entropy": 5.329736280441284,
      "epoch": 6.584099513250406,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019375356357688206,
      "loss": 4.6413,
      "mean_token_accuracy": 0.2559228852391243,
      "num_tokens": 13679092.0,
      "step": 6090
    },
    {
      "entropy": 5.297796964645386,
      "epoch": 6.589507842076799,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019335373216154862,
      "loss": 4.593,
      "mean_token_accuracy": 0.2499736040830612,
      "num_tokens": 13689470.0,
      "step": 6095
    },
    {
      "entropy": 5.315606164932251,
      "epoch": 6.594916170903191,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019295419744894116,
      "loss": 4.6287,
      "mean_token_accuracy": 0.24721593409776688,
      "num_tokens": 13700067.0,
      "step": 6100
    },
    {
      "entropy": 5.181361675262451,
      "epoch": 6.600324499729584,
      "grad_norm": 1.25,
      "learning_rate": 0.00019255496089096985,
      "loss": 4.4151,
      "mean_token_accuracy": 0.2589213490486145,
      "num_tokens": 13711759.0,
      "step": 6105
    },
    {
      "entropy": 5.289453077316284,
      "epoch": 6.605732828555976,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019215602393846104,
      "loss": 4.6315,
      "mean_token_accuracy": 0.25430271178483965,
      "num_tokens": 13724308.0,
      "step": 6110
    },
    {
      "entropy": 5.348058605194092,
      "epoch": 6.611141157382368,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019175738804115279,
      "loss": 4.631,
      "mean_token_accuracy": 0.24599414318799973,
      "num_tokens": 13735007.0,
      "step": 6115
    },
    {
      "entropy": 5.28420581817627,
      "epoch": 6.616549486208761,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019135905464768882,
      "loss": 4.5246,
      "mean_token_accuracy": 0.25925385653972627,
      "num_tokens": 13746426.0,
      "step": 6120
    },
    {
      "entropy": 5.375229740142823,
      "epoch": 6.621957815035154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019096102520561344,
      "loss": 4.6806,
      "mean_token_accuracy": 0.24620629101991653,
      "num_tokens": 13757766.0,
      "step": 6125
    },
    {
      "entropy": 5.351023244857788,
      "epoch": 6.627366143861547,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019056330116136666,
      "loss": 4.6444,
      "mean_token_accuracy": 0.24531188160181044,
      "num_tokens": 13770013.0,
      "step": 6130
    },
    {
      "entropy": 5.21490159034729,
      "epoch": 6.63277447268794,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019016588396027862,
      "loss": 4.4478,
      "mean_token_accuracy": 0.2641163021326065,
      "num_tokens": 13781541.0,
      "step": 6135
    },
    {
      "entropy": 5.306779098510742,
      "epoch": 6.638182801514332,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00018976877504656438,
      "loss": 4.6066,
      "mean_token_accuracy": 0.25005657523870467,
      "num_tokens": 13792652.0,
      "step": 6140
    },
    {
      "entropy": 5.282522392272949,
      "epoch": 6.643591130340725,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001893719758633185,
      "loss": 4.5775,
      "mean_token_accuracy": 0.2505491301417351,
      "num_tokens": 13803458.0,
      "step": 6145
    },
    {
      "entropy": 5.290247869491577,
      "epoch": 6.6489994591671175,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00018897548785251018,
      "loss": 4.6465,
      "mean_token_accuracy": 0.24882402420043945,
      "num_tokens": 13813943.0,
      "step": 6150
    },
    {
      "entropy": 5.357444715499878,
      "epoch": 6.6544077879935095,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001885793124549779,
      "loss": 4.6556,
      "mean_token_accuracy": 0.2519255310297012,
      "num_tokens": 13825671.0,
      "step": 6155
    },
    {
      "entropy": 5.213391447067261,
      "epoch": 6.659816116819902,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00018818345111042383,
      "loss": 4.4604,
      "mean_token_accuracy": 0.26117071211338044,
      "num_tokens": 13836695.0,
      "step": 6160
    },
    {
      "entropy": 5.296697521209717,
      "epoch": 6.665224445646295,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018778790525740907,
      "loss": 4.6086,
      "mean_token_accuracy": 0.25300288051366804,
      "num_tokens": 13847283.0,
      "step": 6165
    },
    {
      "entropy": 5.243547534942627,
      "epoch": 6.670632774472688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018739267633334806,
      "loss": 4.589,
      "mean_token_accuracy": 0.25304637253284457,
      "num_tokens": 13858396.0,
      "step": 6170
    },
    {
      "entropy": 5.354640769958496,
      "epoch": 6.676041103299081,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018699776577450378,
      "loss": 4.6422,
      "mean_token_accuracy": 0.24621736854314805,
      "num_tokens": 13868532.0,
      "step": 6175
    },
    {
      "entropy": 5.346032810211182,
      "epoch": 6.681449432125473,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018660317501598212,
      "loss": 4.619,
      "mean_token_accuracy": 0.25640145540237425,
      "num_tokens": 13879252.0,
      "step": 6180
    },
    {
      "entropy": 5.244696950912475,
      "epoch": 6.686857760951866,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001862089054917266,
      "loss": 4.6022,
      "mean_token_accuracy": 0.2534037992358208,
      "num_tokens": 13891827.0,
      "step": 6185
    },
    {
      "entropy": 5.285698127746582,
      "epoch": 6.692266089778259,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001858149586345137,
      "loss": 4.6093,
      "mean_token_accuracy": 0.2469403401017189,
      "num_tokens": 13903540.0,
      "step": 6190
    },
    {
      "entropy": 5.267003297805786,
      "epoch": 6.6976744186046515,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018542133587594735,
      "loss": 4.4702,
      "mean_token_accuracy": 0.2619795188307762,
      "num_tokens": 13913700.0,
      "step": 6195
    },
    {
      "entropy": 5.291816711425781,
      "epoch": 6.7030827474310435,
      "grad_norm": 1.375,
      "learning_rate": 0.00018502803864645334,
      "loss": 4.5647,
      "mean_token_accuracy": 0.2547138750553131,
      "num_tokens": 13923389.0,
      "step": 6200
    },
    {
      "entropy": 5.299374389648437,
      "epoch": 6.708491076257436,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018463506837527484,
      "loss": 4.7377,
      "mean_token_accuracy": 0.23576038032770158,
      "num_tokens": 13935576.0,
      "step": 6205
    },
    {
      "entropy": 5.2892955303192135,
      "epoch": 6.713899405083829,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018424242649046663,
      "loss": 4.6727,
      "mean_token_accuracy": 0.24158228039741517,
      "num_tokens": 13947247.0,
      "step": 6210
    },
    {
      "entropy": 5.344982433319092,
      "epoch": 6.719307733910222,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018385011441889037,
      "loss": 4.5876,
      "mean_token_accuracy": 0.25422210842370985,
      "num_tokens": 13958070.0,
      "step": 6215
    },
    {
      "entropy": 5.302200651168823,
      "epoch": 6.724716062736614,
      "grad_norm": 1.234375,
      "learning_rate": 0.00018345813358620894,
      "loss": 4.6006,
      "mean_token_accuracy": 0.24975501149892806,
      "num_tokens": 13969117.0,
      "step": 6220
    },
    {
      "entropy": 5.2954981327056885,
      "epoch": 6.730124391563007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001830664854168816,
      "loss": 4.6327,
      "mean_token_accuracy": 0.24950479567050934,
      "num_tokens": 13980901.0,
      "step": 6225
    },
    {
      "entropy": 5.342351198196411,
      "epoch": 6.7355327203894,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018267517133415862,
      "loss": 4.6543,
      "mean_token_accuracy": 0.24442943036556244,
      "num_tokens": 13992702.0,
      "step": 6230
    },
    {
      "entropy": 5.330401086807251,
      "epoch": 6.740941049215793,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018228419276007646,
      "loss": 4.6237,
      "mean_token_accuracy": 0.24718918651342392,
      "num_tokens": 14004500.0,
      "step": 6235
    },
    {
      "entropy": 5.253543901443481,
      "epoch": 6.746349378042185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018189355111545208,
      "loss": 4.544,
      "mean_token_accuracy": 0.2490151196718216,
      "num_tokens": 14014902.0,
      "step": 6240
    },
    {
      "entropy": 5.247914600372314,
      "epoch": 6.7517577068685775,
      "grad_norm": 1.125,
      "learning_rate": 0.0001815032478198781,
      "loss": 4.554,
      "mean_token_accuracy": 0.255257448554039,
      "num_tokens": 14027496.0,
      "step": 6245
    },
    {
      "entropy": 5.364413070678711,
      "epoch": 6.75716603569497,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018111328429171747,
      "loss": 4.6995,
      "mean_token_accuracy": 0.24186500608921052,
      "num_tokens": 14039178.0,
      "step": 6250
    },
    {
      "entropy": 5.370802736282348,
      "epoch": 6.762574364521363,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018072366194809886,
      "loss": 4.6607,
      "mean_token_accuracy": 0.2507839873433113,
      "num_tokens": 14050819.0,
      "step": 6255
    },
    {
      "entropy": 5.337333106994629,
      "epoch": 6.767982693347756,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018033438220491055,
      "loss": 4.6341,
      "mean_token_accuracy": 0.24007685631513595,
      "num_tokens": 14061216.0,
      "step": 6260
    },
    {
      "entropy": 5.275794506072998,
      "epoch": 6.773391022174148,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000179945446476796,
      "loss": 4.5298,
      "mean_token_accuracy": 0.25582854002714156,
      "num_tokens": 14072141.0,
      "step": 6265
    },
    {
      "entropy": 5.31727442741394,
      "epoch": 6.778799351000541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017955685617714867,
      "loss": 4.6155,
      "mean_token_accuracy": 0.2427059143781662,
      "num_tokens": 14082940.0,
      "step": 6270
    },
    {
      "entropy": 5.354738998413086,
      "epoch": 6.784207679826934,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001791686127181066,
      "loss": 4.6398,
      "mean_token_accuracy": 0.24576787799596786,
      "num_tokens": 14093461.0,
      "step": 6275
    },
    {
      "entropy": 5.312971067428589,
      "epoch": 6.789616008653326,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017878071751054736,
      "loss": 4.5964,
      "mean_token_accuracy": 0.24794669300317765,
      "num_tokens": 14104728.0,
      "step": 6280
    },
    {
      "entropy": 5.330068922042846,
      "epoch": 6.795024337479719,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017839317196408305,
      "loss": 4.6293,
      "mean_token_accuracy": 0.25056772381067277,
      "num_tokens": 14115073.0,
      "step": 6285
    },
    {
      "entropy": 5.239559745788574,
      "epoch": 6.800432666306111,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00017800597748705494,
      "loss": 4.5206,
      "mean_token_accuracy": 0.2523434415459633,
      "num_tokens": 14126764.0,
      "step": 6290
    },
    {
      "entropy": 5.291147232055664,
      "epoch": 6.805840995132504,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001776191354865289,
      "loss": 4.56,
      "mean_token_accuracy": 0.24884624481201173,
      "num_tokens": 14137220.0,
      "step": 6295
    },
    {
      "entropy": 5.279448413848877,
      "epoch": 6.811249323958897,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017723264736828952,
      "loss": 4.554,
      "mean_token_accuracy": 0.2547372728586197,
      "num_tokens": 14148760.0,
      "step": 6300
    },
    {
      "entropy": 5.221875810623169,
      "epoch": 6.816657652785289,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017684651453683554,
      "loss": 4.5857,
      "mean_token_accuracy": 0.25068951994180677,
      "num_tokens": 14158925.0,
      "step": 6305
    },
    {
      "entropy": 5.336827659606934,
      "epoch": 6.822065981611682,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017646073839537458,
      "loss": 4.6335,
      "mean_token_accuracy": 0.25039210766553877,
      "num_tokens": 14169661.0,
      "step": 6310
    },
    {
      "entropy": 5.253059577941895,
      "epoch": 6.827474310438075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00017607532034581797,
      "loss": 4.5294,
      "mean_token_accuracy": 0.2492964968085289,
      "num_tokens": 14180975.0,
      "step": 6315
    },
    {
      "entropy": 5.2756753921508786,
      "epoch": 6.832882639264467,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017569026178877584,
      "loss": 4.5815,
      "mean_token_accuracy": 0.2516352981328964,
      "num_tokens": 14191713.0,
      "step": 6320
    },
    {
      "entropy": 5.372540473937988,
      "epoch": 6.83829096809086,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001753055641235519,
      "loss": 4.7511,
      "mean_token_accuracy": 0.2351868912577629,
      "num_tokens": 14204985.0,
      "step": 6325
    },
    {
      "entropy": 5.354352235794067,
      "epoch": 6.8436992969172525,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017492122874813824,
      "loss": 4.7,
      "mean_token_accuracy": 0.24390213340520858,
      "num_tokens": 14215893.0,
      "step": 6330
    },
    {
      "entropy": 5.368043708801269,
      "epoch": 6.849107625743645,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001745372570592107,
      "loss": 4.6917,
      "mean_token_accuracy": 0.24358184337615968,
      "num_tokens": 14229752.0,
      "step": 6335
    },
    {
      "entropy": 5.2836912155151365,
      "epoch": 6.854515954570038,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017415365045212312,
      "loss": 4.5355,
      "mean_token_accuracy": 0.25412821769714355,
      "num_tokens": 14240755.0,
      "step": 6340
    },
    {
      "entropy": 5.3007872104644775,
      "epoch": 6.85992428339643,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017377041032090278,
      "loss": 4.6086,
      "mean_token_accuracy": 0.2436917245388031,
      "num_tokens": 14251115.0,
      "step": 6345
    },
    {
      "entropy": 5.267897605895996,
      "epoch": 6.865332612222823,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017338753805824536,
      "loss": 4.5688,
      "mean_token_accuracy": 0.24938051253557206,
      "num_tokens": 14263565.0,
      "step": 6350
    },
    {
      "entropy": 5.336873245239258,
      "epoch": 6.870740941049216,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017300503505550939,
      "loss": 4.6356,
      "mean_token_accuracy": 0.24724473506212236,
      "num_tokens": 14275834.0,
      "step": 6355
    },
    {
      "entropy": 5.218852806091308,
      "epoch": 6.876149269875609,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017262290270271175,
      "loss": 4.5321,
      "mean_token_accuracy": 0.25103600025177003,
      "num_tokens": 14286650.0,
      "step": 6360
    },
    {
      "entropy": 5.26631350517273,
      "epoch": 6.881557598702001,
      "grad_norm": 1.375,
      "learning_rate": 0.00017224114238852234,
      "loss": 4.6023,
      "mean_token_accuracy": 0.24578307718038558,
      "num_tokens": 14298075.0,
      "step": 6365
    },
    {
      "entropy": 5.286407089233398,
      "epoch": 6.886965927528394,
      "grad_norm": 1.375,
      "learning_rate": 0.0001718597555002588,
      "loss": 4.5195,
      "mean_token_accuracy": 0.252801188826561,
      "num_tokens": 14308063.0,
      "step": 6370
    },
    {
      "entropy": 5.3068461418151855,
      "epoch": 6.8923742563547865,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017147874342388227,
      "loss": 4.5741,
      "mean_token_accuracy": 0.25174898505210874,
      "num_tokens": 14317677.0,
      "step": 6375
    },
    {
      "entropy": 5.320057201385498,
      "epoch": 6.897782585181179,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001710981075439913,
      "loss": 4.6229,
      "mean_token_accuracy": 0.24292435646057128,
      "num_tokens": 14328047.0,
      "step": 6380
    },
    {
      "entropy": 5.306702661514282,
      "epoch": 6.903190914007571,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017071784924381766,
      "loss": 4.6468,
      "mean_token_accuracy": 0.2496078759431839,
      "num_tokens": 14339284.0,
      "step": 6385
    },
    {
      "entropy": 5.297817659378052,
      "epoch": 6.908599242833964,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00017033796990522081,
      "loss": 4.604,
      "mean_token_accuracy": 0.24904375970363618,
      "num_tokens": 14349966.0,
      "step": 6390
    },
    {
      "entropy": 5.3069933414459225,
      "epoch": 6.914007571660357,
      "grad_norm": 1.21875,
      "learning_rate": 0.00016995847090868322,
      "loss": 4.6666,
      "mean_token_accuracy": 0.24434513300657273,
      "num_tokens": 14362071.0,
      "step": 6395
    },
    {
      "entropy": 5.29845609664917,
      "epoch": 6.91941590048675,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016957935363330506,
      "loss": 4.6178,
      "mean_token_accuracy": 0.2553987205028534,
      "num_tokens": 14372438.0,
      "step": 6400
    },
    {
      "entropy": 5.288075160980225,
      "epoch": 6.924824229313142,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016920061945679951,
      "loss": 4.5495,
      "mean_token_accuracy": 0.2537979394197464,
      "num_tokens": 14382864.0,
      "step": 6405
    },
    {
      "entropy": 5.402813148498535,
      "epoch": 6.930232558139535,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001688222697554872,
      "loss": 4.6928,
      "mean_token_accuracy": 0.25468402802944184,
      "num_tokens": 14395146.0,
      "step": 6410
    },
    {
      "entropy": 5.226742601394653,
      "epoch": 6.935640886965928,
      "grad_norm": 1.1875,
      "learning_rate": 0.000168444305904292,
      "loss": 4.5326,
      "mean_token_accuracy": 0.24845264106988907,
      "num_tokens": 14406765.0,
      "step": 6415
    },
    {
      "entropy": 5.334794378280639,
      "epoch": 6.9410492157923205,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016806672927673552,
      "loss": 4.5903,
      "mean_token_accuracy": 0.2594348445534706,
      "num_tokens": 14417964.0,
      "step": 6420
    },
    {
      "entropy": 5.2798051834106445,
      "epoch": 6.9464575446187125,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016768954124493188,
      "loss": 4.5915,
      "mean_token_accuracy": 0.2533974811434746,
      "num_tokens": 14428584.0,
      "step": 6425
    },
    {
      "entropy": 5.348612070083618,
      "epoch": 6.951865873445105,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016731274317958333,
      "loss": 4.6314,
      "mean_token_accuracy": 0.2467951700091362,
      "num_tokens": 14440191.0,
      "step": 6430
    },
    {
      "entropy": 5.293036317825317,
      "epoch": 6.957274202271498,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001669363364499749,
      "loss": 4.6071,
      "mean_token_accuracy": 0.25187247097492216,
      "num_tokens": 14450174.0,
      "step": 6435
    },
    {
      "entropy": 5.262106275558471,
      "epoch": 6.962682531097891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00016656032242396958,
      "loss": 4.55,
      "mean_token_accuracy": 0.26385889798402784,
      "num_tokens": 14462632.0,
      "step": 6440
    },
    {
      "entropy": 5.331650733947754,
      "epoch": 6.968090859924283,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016618470246800322,
      "loss": 4.6357,
      "mean_token_accuracy": 0.24445628374814987,
      "num_tokens": 14473236.0,
      "step": 6445
    },
    {
      "entropy": 5.26259298324585,
      "epoch": 6.973499188750676,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016580947794707937,
      "loss": 4.522,
      "mean_token_accuracy": 0.2543488755822182,
      "num_tokens": 14485944.0,
      "step": 6450
    },
    {
      "entropy": 5.323462963104248,
      "epoch": 6.978907517577069,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016543465022476512,
      "loss": 4.5894,
      "mean_token_accuracy": 0.2549700617790222,
      "num_tokens": 14496592.0,
      "step": 6455
    },
    {
      "entropy": 5.263682079315186,
      "epoch": 6.984315846403462,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001650602206631852,
      "loss": 4.5774,
      "mean_token_accuracy": 0.2492980554699898,
      "num_tokens": 14507256.0,
      "step": 6460
    },
    {
      "entropy": 5.3388104915618895,
      "epoch": 6.9897241752298545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016468619062301739,
      "loss": 4.6107,
      "mean_token_accuracy": 0.2570935606956482,
      "num_tokens": 14518920.0,
      "step": 6465
    },
    {
      "entropy": 5.341052675247193,
      "epoch": 6.995132504056246,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001643125614634877,
      "loss": 4.6104,
      "mean_token_accuracy": 0.25232918858528136,
      "num_tokens": 14529621.0,
      "step": 6470
    },
    {
      "entropy": 5.159323268466526,
      "epoch": 7.0,
      "grad_norm": 2.609375,
      "learning_rate": 0.0001639393345423656,
      "loss": 4.5044,
      "mean_token_accuracy": 0.2490339544084337,
      "num_tokens": 14538223.0,
      "step": 6475
    },
    {
      "entropy": 5.229571104049683,
      "epoch": 7.005408328826393,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016356651121595832,
      "loss": 4.4541,
      "mean_token_accuracy": 0.26700296700000764,
      "num_tokens": 14548750.0,
      "step": 6480
    },
    {
      "entropy": 5.376835489273072,
      "epoch": 7.010816657652786,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001631940928391068,
      "loss": 4.5554,
      "mean_token_accuracy": 0.25943927466869354,
      "num_tokens": 14560289.0,
      "step": 6485
    },
    {
      "entropy": 5.322033214569092,
      "epoch": 7.016224986479178,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016282208076518032,
      "loss": 4.5338,
      "mean_token_accuracy": 0.26359072178602216,
      "num_tokens": 14571886.0,
      "step": 6490
    },
    {
      "entropy": 5.234636068344116,
      "epoch": 7.0216333153055706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016245047634607155,
      "loss": 4.4747,
      "mean_token_accuracy": 0.26687618643045424,
      "num_tokens": 14583743.0,
      "step": 6495
    },
    {
      "entropy": 5.282580089569092,
      "epoch": 7.027041644131963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001620792809321919,
      "loss": 4.5024,
      "mean_token_accuracy": 0.26070408523082733,
      "num_tokens": 14594295.0,
      "step": 6500
    },
    {
      "entropy": 5.321728086471557,
      "epoch": 7.032449972958356,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016170849587246623,
      "loss": 4.5228,
      "mean_token_accuracy": 0.25736051648855207,
      "num_tokens": 14606079.0,
      "step": 6505
    },
    {
      "entropy": 5.285666990280151,
      "epoch": 7.037858301784748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00016133812251432826,
      "loss": 4.4632,
      "mean_token_accuracy": 0.2676470935344696,
      "num_tokens": 14616883.0,
      "step": 6510
    },
    {
      "entropy": 5.277490854263306,
      "epoch": 7.043266630611141,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001609681622037159,
      "loss": 4.4534,
      "mean_token_accuracy": 0.27279370129108427,
      "num_tokens": 14629135.0,
      "step": 6515
    },
    {
      "entropy": 5.251485538482666,
      "epoch": 7.048674959437534,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016059861628506557,
      "loss": 4.4199,
      "mean_token_accuracy": 0.2676568925380707,
      "num_tokens": 14639861.0,
      "step": 6520
    },
    {
      "entropy": 5.217998027801514,
      "epoch": 7.054083288263927,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016022948610130804,
      "loss": 4.4344,
      "mean_token_accuracy": 0.26319105476140975,
      "num_tokens": 14649821.0,
      "step": 6525
    },
    {
      "entropy": 5.213973760604858,
      "epoch": 7.059491617090319,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001598607729938632,
      "loss": 4.4927,
      "mean_token_accuracy": 0.25378472805023194,
      "num_tokens": 14661970.0,
      "step": 6530
    },
    {
      "entropy": 5.3245594024658205,
      "epoch": 7.064899945916712,
      "grad_norm": 1.078125,
      "learning_rate": 0.00015949247830263538,
      "loss": 4.5492,
      "mean_token_accuracy": 0.25951498448848725,
      "num_tokens": 14675174.0,
      "step": 6535
    },
    {
      "entropy": 5.307151460647583,
      "epoch": 7.0703082747431045,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015912460336600838,
      "loss": 4.6651,
      "mean_token_accuracy": 0.242779740691185,
      "num_tokens": 14687636.0,
      "step": 6540
    },
    {
      "entropy": 5.322556495666504,
      "epoch": 7.075716603569497,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015875714952084054,
      "loss": 4.4214,
      "mean_token_accuracy": 0.26470517069101335,
      "num_tokens": 14699222.0,
      "step": 6545
    },
    {
      "entropy": 5.267972660064697,
      "epoch": 7.081124932395889,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015839011810245984,
      "loss": 4.4547,
      "mean_token_accuracy": 0.26688225120306014,
      "num_tokens": 14710018.0,
      "step": 6550
    },
    {
      "entropy": 5.286082363128662,
      "epoch": 7.086533261222282,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001580235104446596,
      "loss": 4.4505,
      "mean_token_accuracy": 0.2653915137052536,
      "num_tokens": 14721369.0,
      "step": 6555
    },
    {
      "entropy": 5.373891115188599,
      "epoch": 7.091941590048675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00015765732787969267,
      "loss": 4.573,
      "mean_token_accuracy": 0.2633775010704994,
      "num_tokens": 14734805.0,
      "step": 6560
    },
    {
      "entropy": 5.239556455612183,
      "epoch": 7.097349918875068,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015729157173826724,
      "loss": 4.4572,
      "mean_token_accuracy": 0.2645534723997116,
      "num_tokens": 14746370.0,
      "step": 6565
    },
    {
      "entropy": 5.32322998046875,
      "epoch": 7.10275824770146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015692624334954236,
      "loss": 4.5145,
      "mean_token_accuracy": 0.2602105110883713,
      "num_tokens": 14758011.0,
      "step": 6570
    },
    {
      "entropy": 5.360467100143433,
      "epoch": 7.108166576527853,
      "grad_norm": 1.375,
      "learning_rate": 0.0001565613440411219,
      "loss": 4.4995,
      "mean_token_accuracy": 0.26418299078941343,
      "num_tokens": 14769602.0,
      "step": 6575
    },
    {
      "entropy": 5.231982898712158,
      "epoch": 7.113574905354246,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001561968751390509,
      "loss": 4.4893,
      "mean_token_accuracy": 0.261266128718853,
      "num_tokens": 14781766.0,
      "step": 6580
    },
    {
      "entropy": 5.29602017402649,
      "epoch": 7.1189832341806385,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00015583283796781022,
      "loss": 4.4949,
      "mean_token_accuracy": 0.2598620608448982,
      "num_tokens": 14792785.0,
      "step": 6585
    },
    {
      "entropy": 5.24949688911438,
      "epoch": 7.1243915630070305,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015546923385031169,
      "loss": 4.4936,
      "mean_token_accuracy": 0.2654686361551285,
      "num_tokens": 14804343.0,
      "step": 6590
    },
    {
      "entropy": 5.2736735343933105,
      "epoch": 7.129799891833423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015510606410789368,
      "loss": 4.4714,
      "mean_token_accuracy": 0.2620068401098251,
      "num_tokens": 14816316.0,
      "step": 6595
    },
    {
      "entropy": 5.28694634437561,
      "epoch": 7.135208220659816,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001547433300603157,
      "loss": 4.4322,
      "mean_token_accuracy": 0.2681753680109978,
      "num_tokens": 14827461.0,
      "step": 6600
    },
    {
      "entropy": 5.2796392917633055,
      "epoch": 7.140616549486209,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015438103302575413,
      "loss": 4.4605,
      "mean_token_accuracy": 0.2686718851327896,
      "num_tokens": 14838942.0,
      "step": 6605
    },
    {
      "entropy": 5.210146570205689,
      "epoch": 7.146024878312601,
      "grad_norm": 1.125,
      "learning_rate": 0.0001540191743207974,
      "loss": 4.4667,
      "mean_token_accuracy": 0.2599436193704605,
      "num_tokens": 14850346.0,
      "step": 6610
    },
    {
      "entropy": 5.244104099273682,
      "epoch": 7.151433207138994,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001536577552604407,
      "loss": 4.4703,
      "mean_token_accuracy": 0.2606957614421844,
      "num_tokens": 14860467.0,
      "step": 6615
    },
    {
      "entropy": 5.288947010040284,
      "epoch": 7.156841535965387,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015329677715808176,
      "loss": 4.5147,
      "mean_token_accuracy": 0.26436121612787244,
      "num_tokens": 14872444.0,
      "step": 6620
    },
    {
      "entropy": 5.269265031814575,
      "epoch": 7.16224986479178,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015293624132551591,
      "loss": 4.5547,
      "mean_token_accuracy": 0.2619938477873802,
      "num_tokens": 14883389.0,
      "step": 6625
    },
    {
      "entropy": 5.257782363891602,
      "epoch": 7.167658193618172,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001525761490729312,
      "loss": 4.4727,
      "mean_token_accuracy": 0.26103770881891253,
      "num_tokens": 14894059.0,
      "step": 6630
    },
    {
      "entropy": 5.261534738540649,
      "epoch": 7.173066522444564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001522165017089036,
      "loss": 4.5218,
      "mean_token_accuracy": 0.26021108627319334,
      "num_tokens": 14906049.0,
      "step": 6635
    },
    {
      "entropy": 5.215019845962525,
      "epoch": 7.178474851270957,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015185730054039266,
      "loss": 4.4634,
      "mean_token_accuracy": 0.2686006724834442,
      "num_tokens": 14917743.0,
      "step": 6640
    },
    {
      "entropy": 5.269952726364136,
      "epoch": 7.18388318009735,
      "grad_norm": 1.25,
      "learning_rate": 0.00015149854687273594,
      "loss": 4.5406,
      "mean_token_accuracy": 0.24950063824653626,
      "num_tokens": 14929190.0,
      "step": 6645
    },
    {
      "entropy": 5.278566217422485,
      "epoch": 7.189291508923742,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001511402420096455,
      "loss": 4.4923,
      "mean_token_accuracy": 0.26192358136177063,
      "num_tokens": 14939722.0,
      "step": 6650
    },
    {
      "entropy": 5.303974580764771,
      "epoch": 7.194699837750135,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001507823872532017,
      "loss": 4.501,
      "mean_token_accuracy": 0.25300774723291397,
      "num_tokens": 14951154.0,
      "step": 6655
    },
    {
      "entropy": 5.304397344589233,
      "epoch": 7.200108166576528,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015042498390384966,
      "loss": 4.4716,
      "mean_token_accuracy": 0.26376015692949295,
      "num_tokens": 14961882.0,
      "step": 6660
    },
    {
      "entropy": 5.219845724105835,
      "epoch": 7.205516495402921,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015006803326039403,
      "loss": 4.4959,
      "mean_token_accuracy": 0.2523091286420822,
      "num_tokens": 14973106.0,
      "step": 6665
    },
    {
      "entropy": 5.31283655166626,
      "epoch": 7.210924824229313,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014971153661999425,
      "loss": 4.5191,
      "mean_token_accuracy": 0.2620886564254761,
      "num_tokens": 14983222.0,
      "step": 6670
    },
    {
      "entropy": 5.355231380462646,
      "epoch": 7.2163331530557056,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001493554952781599,
      "loss": 4.4862,
      "mean_token_accuracy": 0.2581422865390778,
      "num_tokens": 14993704.0,
      "step": 6675
    },
    {
      "entropy": 5.310689258575439,
      "epoch": 7.221741481882098,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014899991052874618,
      "loss": 4.5535,
      "mean_token_accuracy": 0.25814926624298096,
      "num_tokens": 15005328.0,
      "step": 6680
    },
    {
      "entropy": 5.3164530277252195,
      "epoch": 7.227149810708491,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014864478366394853,
      "loss": 4.556,
      "mean_token_accuracy": 0.25675363689661024,
      "num_tokens": 15016165.0,
      "step": 6685
    },
    {
      "entropy": 5.273098802566528,
      "epoch": 7.232558139534884,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001482901159742992,
      "loss": 4.4472,
      "mean_token_accuracy": 0.2683747708797455,
      "num_tokens": 15027383.0,
      "step": 6690
    },
    {
      "entropy": 5.293208980560303,
      "epoch": 7.237966468361276,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001479359087486613,
      "loss": 4.5053,
      "mean_token_accuracy": 0.2581513822078705,
      "num_tokens": 15038284.0,
      "step": 6695
    },
    {
      "entropy": 5.285342741012573,
      "epoch": 7.243374797187669,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014758216327422457,
      "loss": 4.5005,
      "mean_token_accuracy": 0.2557239904999733,
      "num_tokens": 15048401.0,
      "step": 6700
    },
    {
      "entropy": 5.237166786193848,
      "epoch": 7.248783126014062,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014722888083650094,
      "loss": 4.3723,
      "mean_token_accuracy": 0.2718945950269699,
      "num_tokens": 15059114.0,
      "step": 6705
    },
    {
      "entropy": 5.2753761291503904,
      "epoch": 7.254191454840455,
      "grad_norm": 1.234375,
      "learning_rate": 0.00014687606271931961,
      "loss": 4.5528,
      "mean_token_accuracy": 0.2572538271546364,
      "num_tokens": 15069750.0,
      "step": 6710
    },
    {
      "entropy": 5.218373203277588,
      "epoch": 7.259599783666847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014652371020482246,
      "loss": 4.381,
      "mean_token_accuracy": 0.27517917305231093,
      "num_tokens": 15081300.0,
      "step": 6715
    },
    {
      "entropy": 5.2420666217803955,
      "epoch": 7.2650081124932395,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001461718245734594,
      "loss": 4.4494,
      "mean_token_accuracy": 0.2641877353191376,
      "num_tokens": 15092249.0,
      "step": 6720
    },
    {
      "entropy": 5.274439287185669,
      "epoch": 7.270416441319632,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001458204071039834,
      "loss": 4.4443,
      "mean_token_accuracy": 0.26725870221853254,
      "num_tokens": 15103109.0,
      "step": 6725
    },
    {
      "entropy": 5.2901873111724855,
      "epoch": 7.275824770146025,
      "grad_norm": 1.125,
      "learning_rate": 0.0001454694590734466,
      "loss": 4.5059,
      "mean_token_accuracy": 0.26258442997932435,
      "num_tokens": 15115324.0,
      "step": 6730
    },
    {
      "entropy": 5.305406618118286,
      "epoch": 7.281233098972417,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001451189817571949,
      "loss": 4.5225,
      "mean_token_accuracy": 0.2591644212603569,
      "num_tokens": 15127547.0,
      "step": 6735
    },
    {
      "entropy": 5.232668399810791,
      "epoch": 7.28664142779881,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014476897642886355,
      "loss": 4.4672,
      "mean_token_accuracy": 0.26526825577020646,
      "num_tokens": 15138296.0,
      "step": 6740
    },
    {
      "entropy": 5.238512086868286,
      "epoch": 7.292049756625203,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014441944436037287,
      "loss": 4.4938,
      "mean_token_accuracy": 0.25960848331451414,
      "num_tokens": 15149027.0,
      "step": 6745
    },
    {
      "entropy": 5.254163503646851,
      "epoch": 7.297458085451596,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014407038682192312,
      "loss": 4.4793,
      "mean_token_accuracy": 0.26082787960767745,
      "num_tokens": 15159694.0,
      "step": 6750
    },
    {
      "entropy": 5.2690185546875,
      "epoch": 7.302866414277988,
      "grad_norm": 1.125,
      "learning_rate": 0.0001437218050819902,
      "loss": 4.4061,
      "mean_token_accuracy": 0.2678681746125221,
      "num_tokens": 15171215.0,
      "step": 6755
    },
    {
      "entropy": 5.248695516586304,
      "epoch": 7.308274743104381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000143373700407321,
      "loss": 4.522,
      "mean_token_accuracy": 0.2559132695198059,
      "num_tokens": 15183575.0,
      "step": 6760
    },
    {
      "entropy": 5.244789934158325,
      "epoch": 7.3136830719307735,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014302607406292873,
      "loss": 4.512,
      "mean_token_accuracy": 0.2635293662548065,
      "num_tokens": 15195036.0,
      "step": 6765
    },
    {
      "entropy": 5.206980466842651,
      "epoch": 7.319091400757166,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014267892731208826,
      "loss": 4.422,
      "mean_token_accuracy": 0.2687758654356003,
      "num_tokens": 15206532.0,
      "step": 6770
    },
    {
      "entropy": 5.302237844467163,
      "epoch": 7.324499729583558,
      "grad_norm": 1.125,
      "learning_rate": 0.0001423322614163318,
      "loss": 4.5079,
      "mean_token_accuracy": 0.25992675721645353,
      "num_tokens": 15218426.0,
      "step": 6775
    },
    {
      "entropy": 5.3017583847045895,
      "epoch": 7.329908058409951,
      "grad_norm": 1.265625,
      "learning_rate": 0.00014198607763544392,
      "loss": 4.5236,
      "mean_token_accuracy": 0.2611626639962196,
      "num_tokens": 15229302.0,
      "step": 6780
    },
    {
      "entropy": 5.26635160446167,
      "epoch": 7.335316387236344,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014164037722745715,
      "loss": 4.4758,
      "mean_token_accuracy": 0.26295520216226576,
      "num_tokens": 15240409.0,
      "step": 6785
    },
    {
      "entropy": 5.346647691726685,
      "epoch": 7.340724716062737,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001412951614486479,
      "loss": 4.5149,
      "mean_token_accuracy": 0.2561726912856102,
      "num_tokens": 15250891.0,
      "step": 6790
    },
    {
      "entropy": 5.285936880111694,
      "epoch": 7.346133044889129,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014095043155353085,
      "loss": 4.4518,
      "mean_token_accuracy": 0.26183116286993025,
      "num_tokens": 15261274.0,
      "step": 6795
    },
    {
      "entropy": 5.295490741729736,
      "epoch": 7.351541373715522,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001406061887948553,
      "loss": 4.5091,
      "mean_token_accuracy": 0.2573329910635948,
      "num_tokens": 15271817.0,
      "step": 6800
    },
    {
      "entropy": 5.239786195755005,
      "epoch": 7.356949702541915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014026243442360025,
      "loss": 4.4776,
      "mean_token_accuracy": 0.2662931874394417,
      "num_tokens": 15282973.0,
      "step": 6805
    },
    {
      "entropy": 5.290739440917969,
      "epoch": 7.3623580313683075,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001399191696889699,
      "loss": 4.5169,
      "mean_token_accuracy": 0.26422159373760223,
      "num_tokens": 15294631.0,
      "step": 6810
    },
    {
      "entropy": 5.2489439964294435,
      "epoch": 7.367766360194699,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001395763958383891,
      "loss": 4.5244,
      "mean_token_accuracy": 0.2637367844581604,
      "num_tokens": 15306272.0,
      "step": 6815
    },
    {
      "entropy": 5.2919933795928955,
      "epoch": 7.373174689021092,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001392341141174987,
      "loss": 4.4847,
      "mean_token_accuracy": 0.2632103517651558,
      "num_tokens": 15318103.0,
      "step": 6820
    },
    {
      "entropy": 5.312174558639526,
      "epoch": 7.378583017847485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013889232577015125,
      "loss": 4.5335,
      "mean_token_accuracy": 0.25433182120323183,
      "num_tokens": 15329683.0,
      "step": 6825
    },
    {
      "entropy": 5.253097915649414,
      "epoch": 7.383991346673878,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013855103203840673,
      "loss": 4.4637,
      "mean_token_accuracy": 0.26057019680738447,
      "num_tokens": 15340374.0,
      "step": 6830
    },
    {
      "entropy": 5.292276096343994,
      "epoch": 7.38939967550027,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013821023416252693,
      "loss": 4.4332,
      "mean_token_accuracy": 0.259838205575943,
      "num_tokens": 15351690.0,
      "step": 6835
    },
    {
      "entropy": 5.352556467056274,
      "epoch": 7.394808004326663,
      "grad_norm": 1.375,
      "learning_rate": 0.00013786993338097236,
      "loss": 4.5641,
      "mean_token_accuracy": 0.2602698192000389,
      "num_tokens": 15363290.0,
      "step": 6840
    },
    {
      "entropy": 5.2232481956481935,
      "epoch": 7.400216333153056,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013753013093039673,
      "loss": 4.4397,
      "mean_token_accuracy": 0.26292750537395476,
      "num_tokens": 15373531.0,
      "step": 6845
    },
    {
      "entropy": 5.27802357673645,
      "epoch": 7.405624661979449,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000137190828045643,
      "loss": 4.5215,
      "mean_token_accuracy": 0.2595431089401245,
      "num_tokens": 15385210.0,
      "step": 6850
    },
    {
      "entropy": 5.254598474502563,
      "epoch": 7.411032990805841,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013685202595973861,
      "loss": 4.5009,
      "mean_token_accuracy": 0.2553673520684242,
      "num_tokens": 15396048.0,
      "step": 6855
    },
    {
      "entropy": 5.271158027648926,
      "epoch": 7.416441319632233,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013651372590389122,
      "loss": 4.4867,
      "mean_token_accuracy": 0.25666648894548416,
      "num_tokens": 15407157.0,
      "step": 6860
    },
    {
      "entropy": 5.291407442092895,
      "epoch": 7.421849648458626,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013617592910748378,
      "loss": 4.5008,
      "mean_token_accuracy": 0.26287313997745515,
      "num_tokens": 15418930.0,
      "step": 6865
    },
    {
      "entropy": 5.386734437942505,
      "epoch": 7.427257977285019,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013583863679807092,
      "loss": 4.5351,
      "mean_token_accuracy": 0.261413948237896,
      "num_tokens": 15430668.0,
      "step": 6870
    },
    {
      "entropy": 5.366476202011109,
      "epoch": 7.432666306111411,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013550185020137344,
      "loss": 4.5711,
      "mean_token_accuracy": 0.25355122834444044,
      "num_tokens": 15440693.0,
      "step": 6875
    },
    {
      "entropy": 5.323111915588379,
      "epoch": 7.438074634937804,
      "grad_norm": 1.25,
      "learning_rate": 0.0001351655705412747,
      "loss": 4.5369,
      "mean_token_accuracy": 0.25716243088245394,
      "num_tokens": 15451380.0,
      "step": 6880
    },
    {
      "entropy": 5.376512145996093,
      "epoch": 7.443482963764197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001348297990398158,
      "loss": 4.6215,
      "mean_token_accuracy": 0.25879693031311035,
      "num_tokens": 15462913.0,
      "step": 6885
    },
    {
      "entropy": 5.292003107070923,
      "epoch": 7.44889129259059,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013449453691719108,
      "loss": 4.4606,
      "mean_token_accuracy": 0.2675152450799942,
      "num_tokens": 15473353.0,
      "step": 6890
    },
    {
      "entropy": 5.293996715545655,
      "epoch": 7.4542996214169825,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001341597853917439,
      "loss": 4.497,
      "mean_token_accuracy": 0.2631390243768692,
      "num_tokens": 15483088.0,
      "step": 6895
    },
    {
      "entropy": 5.224536991119384,
      "epoch": 7.4597079502433745,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013382554567996214,
      "loss": 4.5066,
      "mean_token_accuracy": 0.2572442710399628,
      "num_tokens": 15495723.0,
      "step": 6900
    },
    {
      "entropy": 5.256581163406372,
      "epoch": 7.465116279069767,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013349181899647346,
      "loss": 4.4002,
      "mean_token_accuracy": 0.2732828453183174,
      "num_tokens": 15507950.0,
      "step": 6905
    },
    {
      "entropy": 5.320991516113281,
      "epoch": 7.47052460789616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013315860655404167,
      "loss": 4.544,
      "mean_token_accuracy": 0.2576855421066284,
      "num_tokens": 15519419.0,
      "step": 6910
    },
    {
      "entropy": 5.2594499588012695,
      "epoch": 7.475932936722553,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001328259095635615,
      "loss": 4.5678,
      "mean_token_accuracy": 0.25405617505311967,
      "num_tokens": 15530201.0,
      "step": 6915
    },
    {
      "entropy": 5.237321090698242,
      "epoch": 7.481341265548945,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001324937292340545,
      "loss": 4.4526,
      "mean_token_accuracy": 0.25767187774181366,
      "num_tokens": 15540462.0,
      "step": 6920
    },
    {
      "entropy": 5.234324312210083,
      "epoch": 7.486749594375338,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013216206677266475,
      "loss": 4.4657,
      "mean_token_accuracy": 0.2611553281545639,
      "num_tokens": 15552226.0,
      "step": 6925
    },
    {
      "entropy": 5.299695348739624,
      "epoch": 7.492157923201731,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001318309233846544,
      "loss": 4.5078,
      "mean_token_accuracy": 0.2598936125636101,
      "num_tokens": 15562081.0,
      "step": 6930
    },
    {
      "entropy": 5.309095430374145,
      "epoch": 7.497566252028124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013150030027339935,
      "loss": 4.6273,
      "mean_token_accuracy": 0.25217877328395844,
      "num_tokens": 15573082.0,
      "step": 6935
    },
    {
      "entropy": 5.309836673736572,
      "epoch": 7.502974580854516,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013117019864038483,
      "loss": 4.5328,
      "mean_token_accuracy": 0.25549083948135376,
      "num_tokens": 15585977.0,
      "step": 6940
    },
    {
      "entropy": 5.2518167972564695,
      "epoch": 7.5083829096809085,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013084061968520063,
      "loss": 4.4756,
      "mean_token_accuracy": 0.2641377940773964,
      "num_tokens": 15597722.0,
      "step": 6945
    },
    {
      "entropy": 5.282698059082032,
      "epoch": 7.513791238507301,
      "grad_norm": 1.40625,
      "learning_rate": 0.00013051156460553776,
      "loss": 4.5335,
      "mean_token_accuracy": 0.2580183491110802,
      "num_tokens": 15608493.0,
      "step": 6950
    },
    {
      "entropy": 5.272831726074219,
      "epoch": 7.519199567333694,
      "grad_norm": 1.3125,
      "learning_rate": 0.00013018303459718305,
      "loss": 4.4818,
      "mean_token_accuracy": 0.26111680567264556,
      "num_tokens": 15618261.0,
      "step": 6955
    },
    {
      "entropy": 5.308541011810303,
      "epoch": 7.524607896160086,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012985503085401523,
      "loss": 4.5511,
      "mean_token_accuracy": 0.25359287708997724,
      "num_tokens": 15629743.0,
      "step": 6960
    },
    {
      "entropy": 5.257376337051392,
      "epoch": 7.530016224986479,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012952755456800067,
      "loss": 4.4847,
      "mean_token_accuracy": 0.26177666932344434,
      "num_tokens": 15641749.0,
      "step": 6965
    },
    {
      "entropy": 5.227441692352295,
      "epoch": 7.535424553812872,
      "grad_norm": 1.359375,
      "learning_rate": 0.000129200606929189,
      "loss": 4.4686,
      "mean_token_accuracy": 0.265121765434742,
      "num_tokens": 15651890.0,
      "step": 6970
    },
    {
      "entropy": 5.293875789642334,
      "epoch": 7.540832882639265,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001288741891257087,
      "loss": 4.5681,
      "mean_token_accuracy": 0.2551468551158905,
      "num_tokens": 15663892.0,
      "step": 6975
    },
    {
      "entropy": 5.254330348968506,
      "epoch": 7.546241211465657,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012854830234376277,
      "loss": 4.4883,
      "mean_token_accuracy": 0.2573761820793152,
      "num_tokens": 15674025.0,
      "step": 6980
    },
    {
      "entropy": 5.271991539001465,
      "epoch": 7.55164954029205,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001282229477676246,
      "loss": 4.5127,
      "mean_token_accuracy": 0.2571659341454506,
      "num_tokens": 15686352.0,
      "step": 6985
    },
    {
      "entropy": 5.320364284515381,
      "epoch": 7.5570578691184425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001278981265796334,
      "loss": 4.5515,
      "mean_token_accuracy": 0.2638563051819801,
      "num_tokens": 15697082.0,
      "step": 6990
    },
    {
      "entropy": 5.340955066680908,
      "epoch": 7.562466197944835,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012757383996019024,
      "loss": 4.5929,
      "mean_token_accuracy": 0.2504038974642754,
      "num_tokens": 15707898.0,
      "step": 6995
    },
    {
      "entropy": 5.25543441772461,
      "epoch": 7.567874526771227,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001272500890877533,
      "loss": 4.5524,
      "mean_token_accuracy": 0.2552588298916817,
      "num_tokens": 15719665.0,
      "step": 7000
    },
    {
      "epoch": 7.567874526771227,
      "eval_entropy": 5.186373500390486,
      "eval_loss": 5.843907356262207,
      "eval_mean_token_accuracy": 0.18266003998843106,
      "eval_num_tokens": 15719665.0,
      "eval_runtime": 1.9442,
      "eval_samples_per_second": 1807.903,
      "eval_steps_per_second": 226.309,
      "step": 7000
    },
    {
      "entropy": 5.324835300445557,
      "epoch": 7.57328285559762,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012692687513883393,
      "loss": 4.5438,
      "mean_token_accuracy": 0.2625204935669899,
      "num_tokens": 15731245.0,
      "step": 7005
    },
    {
      "entropy": 5.350880670547485,
      "epoch": 7.578691184424013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00012660419928799265,
      "loss": 4.5643,
      "mean_token_accuracy": 0.25950399935245516,
      "num_tokens": 15742582.0,
      "step": 7010
    },
    {
      "entropy": 5.28609709739685,
      "epoch": 7.584099513250406,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012628206270783404,
      "loss": 4.4705,
      "mean_token_accuracy": 0.26656648218631745,
      "num_tokens": 15753491.0,
      "step": 7015
    },
    {
      "entropy": 5.345163440704345,
      "epoch": 7.589507842076799,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012596046656900323,
      "loss": 4.5873,
      "mean_token_accuracy": 0.2559844419360161,
      "num_tokens": 15764325.0,
      "step": 7020
    },
    {
      "entropy": 5.234419918060302,
      "epoch": 7.594916170903191,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001256394120401813,
      "loss": 4.4434,
      "mean_token_accuracy": 0.266838526725769,
      "num_tokens": 15776113.0,
      "step": 7025
    },
    {
      "entropy": 5.25258617401123,
      "epoch": 7.600324499729584,
      "grad_norm": 1.1875,
      "learning_rate": 0.00012531890028808122,
      "loss": 4.4294,
      "mean_token_accuracy": 0.26675664633512497,
      "num_tokens": 15787159.0,
      "step": 7030
    },
    {
      "entropy": 5.259580659866333,
      "epoch": 7.605732828555976,
      "grad_norm": 1.25,
      "learning_rate": 0.0001249989324774434,
      "loss": 4.519,
      "mean_token_accuracy": 0.26144583374261854,
      "num_tokens": 15798329.0,
      "step": 7035
    },
    {
      "entropy": 5.315686798095703,
      "epoch": 7.611141157382368,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012467950977103158,
      "loss": 4.5649,
      "mean_token_accuracy": 0.25382937043905257,
      "num_tokens": 15808782.0,
      "step": 7040
    },
    {
      "entropy": 5.289865112304687,
      "epoch": 7.616549486208761,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00012436063332962852,
      "loss": 4.5424,
      "mean_token_accuracy": 0.2591539531946182,
      "num_tokens": 15821474.0,
      "step": 7045
    },
    {
      "entropy": 5.177752065658569,
      "epoch": 7.621957815035154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012404230431203227,
      "loss": 4.408,
      "mean_token_accuracy": 0.2710533171892166,
      "num_tokens": 15832097.0,
      "step": 7050
    },
    {
      "entropy": 5.31380877494812,
      "epoch": 7.627366143861547,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012372452387505087,
      "loss": 4.6006,
      "mean_token_accuracy": 0.2502247586846352,
      "num_tokens": 15843581.0,
      "step": 7055
    },
    {
      "entropy": 5.362049150466919,
      "epoch": 7.63277447268794,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001234072931734993,
      "loss": 4.5847,
      "mean_token_accuracy": 0.2536924973130226,
      "num_tokens": 15855296.0,
      "step": 7060
    },
    {
      "entropy": 5.2861621379852295,
      "epoch": 7.638182801514332,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012309061336019465,
      "loss": 4.4342,
      "mean_token_accuracy": 0.2704297050833702,
      "num_tokens": 15866548.0,
      "step": 7065
    },
    {
      "entropy": 5.283108520507812,
      "epoch": 7.643591130340725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012277448558595212,
      "loss": 4.4665,
      "mean_token_accuracy": 0.26218216121196747,
      "num_tokens": 15877356.0,
      "step": 7070
    },
    {
      "entropy": 5.238054466247559,
      "epoch": 7.6489994591671175,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012245891099958074,
      "loss": 4.4881,
      "mean_token_accuracy": 0.259418448805809,
      "num_tokens": 15888658.0,
      "step": 7075
    },
    {
      "entropy": 5.293131637573242,
      "epoch": 7.6544077879935095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001221438907478793,
      "loss": 4.5039,
      "mean_token_accuracy": 0.25609496980905533,
      "num_tokens": 15899581.0,
      "step": 7080
    },
    {
      "entropy": 5.263376998901367,
      "epoch": 7.659816116819902,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012182942597563213,
      "loss": 4.5175,
      "mean_token_accuracy": 0.2642672747373581,
      "num_tokens": 15911545.0,
      "step": 7085
    },
    {
      "entropy": 5.294560861587525,
      "epoch": 7.665224445646295,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012151551782560499,
      "loss": 4.5058,
      "mean_token_accuracy": 0.25972277373075486,
      "num_tokens": 15922108.0,
      "step": 7090
    },
    {
      "entropy": 5.281133937835693,
      "epoch": 7.670632774472688,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012120216743854068,
      "loss": 4.5194,
      "mean_token_accuracy": 0.25772617310285567,
      "num_tokens": 15933366.0,
      "step": 7095
    },
    {
      "entropy": 5.271060228347778,
      "epoch": 7.676041103299081,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00012088937595315525,
      "loss": 4.4741,
      "mean_token_accuracy": 0.26725546568632125,
      "num_tokens": 15943645.0,
      "step": 7100
    },
    {
      "entropy": 5.304051494598388,
      "epoch": 7.681449432125473,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001205771445061339,
      "loss": 4.5283,
      "mean_token_accuracy": 0.26238342821598054,
      "num_tokens": 15954147.0,
      "step": 7105
    },
    {
      "entropy": 5.282717561721801,
      "epoch": 7.686857760951866,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001202654742321262,
      "loss": 4.455,
      "mean_token_accuracy": 0.26161874532699586,
      "num_tokens": 15965897.0,
      "step": 7110
    },
    {
      "entropy": 5.308388090133667,
      "epoch": 7.692266089778259,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011995436626374272,
      "loss": 4.4952,
      "mean_token_accuracy": 0.26179309487342833,
      "num_tokens": 15976709.0,
      "step": 7115
    },
    {
      "entropy": 5.277400636672974,
      "epoch": 7.6976744186046515,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011964382173155053,
      "loss": 4.5102,
      "mean_token_accuracy": 0.2595001384615898,
      "num_tokens": 15988520.0,
      "step": 7120
    },
    {
      "entropy": 5.18413896560669,
      "epoch": 7.7030827474310435,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011933384176406915,
      "loss": 4.4556,
      "mean_token_accuracy": 0.25904988795518874,
      "num_tokens": 15999087.0,
      "step": 7125
    },
    {
      "entropy": 5.196502828598023,
      "epoch": 7.708491076257436,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001190244274877666,
      "loss": 4.4754,
      "mean_token_accuracy": 0.25747917741537096,
      "num_tokens": 16009574.0,
      "step": 7130
    },
    {
      "entropy": 5.30810694694519,
      "epoch": 7.713899405083829,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011871558002705501,
      "loss": 4.4895,
      "mean_token_accuracy": 0.25934955179691316,
      "num_tokens": 16019599.0,
      "step": 7135
    },
    {
      "entropy": 5.274073076248169,
      "epoch": 7.719307733910222,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011840730050428666,
      "loss": 4.5031,
      "mean_token_accuracy": 0.2583166003227234,
      "num_tokens": 16029713.0,
      "step": 7140
    },
    {
      "entropy": 5.184285259246826,
      "epoch": 7.724716062736614,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001180995900397503,
      "loss": 4.4112,
      "mean_token_accuracy": 0.2671406403183937,
      "num_tokens": 16040457.0,
      "step": 7145
    },
    {
      "entropy": 5.286316776275635,
      "epoch": 7.730124391563007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011779244975166623,
      "loss": 4.592,
      "mean_token_accuracy": 0.26164825558662413,
      "num_tokens": 16051828.0,
      "step": 7150
    },
    {
      "entropy": 5.290556526184082,
      "epoch": 7.7355327203894,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011748588075618305,
      "loss": 4.4688,
      "mean_token_accuracy": 0.2647244408726692,
      "num_tokens": 16062432.0,
      "step": 7155
    },
    {
      "entropy": 5.209062385559082,
      "epoch": 7.740941049215793,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011717988416737322,
      "loss": 4.3947,
      "mean_token_accuracy": 0.2751539722084999,
      "num_tokens": 16073348.0,
      "step": 7160
    },
    {
      "entropy": 5.350267601013184,
      "epoch": 7.746349378042185,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011687446109722904,
      "loss": 4.5256,
      "mean_token_accuracy": 0.2624422937631607,
      "num_tokens": 16084654.0,
      "step": 7165
    },
    {
      "entropy": 5.235049390792847,
      "epoch": 7.7517577068685775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00011656961265565864,
      "loss": 4.4826,
      "mean_token_accuracy": 0.25973933935165405,
      "num_tokens": 16097180.0,
      "step": 7170
    },
    {
      "entropy": 5.233228445053101,
      "epoch": 7.75716603569497,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011626533995048204,
      "loss": 4.4472,
      "mean_token_accuracy": 0.2616032138466835,
      "num_tokens": 16108335.0,
      "step": 7175
    },
    {
      "entropy": 5.2383324146270756,
      "epoch": 7.762574364521363,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011596164408742675,
      "loss": 4.4156,
      "mean_token_accuracy": 0.26660248041152956,
      "num_tokens": 16119588.0,
      "step": 7180
    },
    {
      "entropy": 5.243097925186158,
      "epoch": 7.767982693347756,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011565852617012452,
      "loss": 4.4813,
      "mean_token_accuracy": 0.26303700953722,
      "num_tokens": 16130478.0,
      "step": 7185
    },
    {
      "entropy": 5.2632612705230715,
      "epoch": 7.773391022174148,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011535598730010633,
      "loss": 4.5299,
      "mean_token_accuracy": 0.2550530344247818,
      "num_tokens": 16141388.0,
      "step": 7190
    },
    {
      "entropy": 5.320485544204712,
      "epoch": 7.778799351000541,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011505402857679921,
      "loss": 4.5845,
      "mean_token_accuracy": 0.25544782280921935,
      "num_tokens": 16152965.0,
      "step": 7195
    },
    {
      "entropy": 5.356726360321045,
      "epoch": 7.784207679826934,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011475265109752187,
      "loss": 4.5835,
      "mean_token_accuracy": 0.2590448632836342,
      "num_tokens": 16164409.0,
      "step": 7200
    },
    {
      "entropy": 5.231152868270874,
      "epoch": 7.789616008653326,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011445185595748074,
      "loss": 4.4648,
      "mean_token_accuracy": 0.2581787496805191,
      "num_tokens": 16175875.0,
      "step": 7205
    },
    {
      "entropy": 5.194439649581909,
      "epoch": 7.795024337479719,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001141516442497661,
      "loss": 4.4934,
      "mean_token_accuracy": 0.2662027314305305,
      "num_tokens": 16187495.0,
      "step": 7210
    },
    {
      "entropy": 5.2879363059997555,
      "epoch": 7.800432666306111,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011385201706534803,
      "loss": 4.4794,
      "mean_token_accuracy": 0.26003775000572205,
      "num_tokens": 16197092.0,
      "step": 7215
    },
    {
      "entropy": 5.191719055175781,
      "epoch": 7.805840995132504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001135529754930722,
      "loss": 4.394,
      "mean_token_accuracy": 0.2709829092025757,
      "num_tokens": 16209321.0,
      "step": 7220
    },
    {
      "entropy": 5.2472635269165036,
      "epoch": 7.811249323958897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011325452061965662,
      "loss": 4.4707,
      "mean_token_accuracy": 0.26589575707912444,
      "num_tokens": 16219961.0,
      "step": 7225
    },
    {
      "entropy": 5.229808616638183,
      "epoch": 7.816657652785289,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011295665352968692,
      "loss": 4.458,
      "mean_token_accuracy": 0.26534799933433534,
      "num_tokens": 16231209.0,
      "step": 7230
    },
    {
      "entropy": 5.258481740951538,
      "epoch": 7.822065981611682,
      "grad_norm": 1.203125,
      "learning_rate": 0.00011265937530561271,
      "loss": 4.4855,
      "mean_token_accuracy": 0.2593889608979225,
      "num_tokens": 16242163.0,
      "step": 7235
    },
    {
      "entropy": 5.267598581314087,
      "epoch": 7.827474310438075,
      "grad_norm": 1.359375,
      "learning_rate": 0.00011236268702774377,
      "loss": 4.5856,
      "mean_token_accuracy": 0.2564650386571884,
      "num_tokens": 16253552.0,
      "step": 7240
    },
    {
      "entropy": 5.286518573760986,
      "epoch": 7.832882639264467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011206658977424605,
      "loss": 4.4979,
      "mean_token_accuracy": 0.25697878152132037,
      "num_tokens": 16264491.0,
      "step": 7245
    },
    {
      "entropy": 5.342361783981323,
      "epoch": 7.83829096809086,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011177108462113761,
      "loss": 4.5254,
      "mean_token_accuracy": 0.26248190402984617,
      "num_tokens": 16276088.0,
      "step": 7250
    },
    {
      "entropy": 5.208338022232056,
      "epoch": 7.8436992969172525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00011147617264228491,
      "loss": 4.3769,
      "mean_token_accuracy": 0.27556352466344836,
      "num_tokens": 16287942.0,
      "step": 7255
    },
    {
      "entropy": 5.180019760131836,
      "epoch": 7.849107625743645,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011118185490939861,
      "loss": 4.3797,
      "mean_token_accuracy": 0.2680892840027809,
      "num_tokens": 16299029.0,
      "step": 7260
    },
    {
      "entropy": 5.2202392578125,
      "epoch": 7.854515954570038,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011088813249203019,
      "loss": 4.4738,
      "mean_token_accuracy": 0.25906888842582704,
      "num_tokens": 16309723.0,
      "step": 7265
    },
    {
      "entropy": 5.250007390975952,
      "epoch": 7.85992428339643,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011059500645756762,
      "loss": 4.499,
      "mean_token_accuracy": 0.25598124861717225,
      "num_tokens": 16322002.0,
      "step": 7270
    },
    {
      "entropy": 5.218455791473389,
      "epoch": 7.865332612222823,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011030247787123147,
      "loss": 4.4988,
      "mean_token_accuracy": 0.24705973267555237,
      "num_tokens": 16334103.0,
      "step": 7275
    },
    {
      "entropy": 5.264587640762329,
      "epoch": 7.870740941049216,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011001054779607121,
      "loss": 4.4635,
      "mean_token_accuracy": 0.2678821459412575,
      "num_tokens": 16345927.0,
      "step": 7280
    },
    {
      "entropy": 5.303491401672363,
      "epoch": 7.876149269875609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00010971921729296172,
      "loss": 4.4977,
      "mean_token_accuracy": 0.260498908162117,
      "num_tokens": 16357033.0,
      "step": 7285
    },
    {
      "entropy": 5.254551076889038,
      "epoch": 7.881557598702001,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010942848742059845,
      "loss": 4.5149,
      "mean_token_accuracy": 0.25765434503555296,
      "num_tokens": 16367097.0,
      "step": 7290
    },
    {
      "entropy": 5.277462148666382,
      "epoch": 7.886965927528394,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010913835923549453,
      "loss": 4.4877,
      "mean_token_accuracy": 0.2657338410615921,
      "num_tokens": 16378474.0,
      "step": 7295
    },
    {
      "entropy": 5.276457929611206,
      "epoch": 7.8923742563547865,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00010884883379197643,
      "loss": 4.4381,
      "mean_token_accuracy": 0.2684517174959183,
      "num_tokens": 16390518.0,
      "step": 7300
    },
    {
      "entropy": 5.280886316299439,
      "epoch": 7.897782585181179,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010855991214218028,
      "loss": 4.5232,
      "mean_token_accuracy": 0.2570959195494652,
      "num_tokens": 16400340.0,
      "step": 7305
    },
    {
      "entropy": 5.282779932022095,
      "epoch": 7.903190914007571,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010827159533604814,
      "loss": 4.5473,
      "mean_token_accuracy": 0.2527189373970032,
      "num_tokens": 16411658.0,
      "step": 7310
    },
    {
      "entropy": 5.347098398208618,
      "epoch": 7.908599242833964,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010798388442132375,
      "loss": 4.606,
      "mean_token_accuracy": 0.24744042605161667,
      "num_tokens": 16421985.0,
      "step": 7315
    },
    {
      "entropy": 5.247751998901367,
      "epoch": 7.914007571660357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010769678044354927,
      "loss": 4.5144,
      "mean_token_accuracy": 0.2527934968471527,
      "num_tokens": 16433589.0,
      "step": 7320
    },
    {
      "entropy": 5.20989294052124,
      "epoch": 7.91941590048675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010741028444606141,
      "loss": 4.4341,
      "mean_token_accuracy": 0.27007447481155394,
      "num_tokens": 16446210.0,
      "step": 7325
    },
    {
      "entropy": 5.218980503082276,
      "epoch": 7.924824229313142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010712439746998711,
      "loss": 4.3994,
      "mean_token_accuracy": 0.2631054908037186,
      "num_tokens": 16457289.0,
      "step": 7330
    },
    {
      "entropy": 5.242149543762207,
      "epoch": 7.930232558139535,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010683912055424036,
      "loss": 4.4667,
      "mean_token_accuracy": 0.26201580464839935,
      "num_tokens": 16468106.0,
      "step": 7335
    },
    {
      "entropy": 5.244215440750122,
      "epoch": 7.935640886965928,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010655445473551809,
      "loss": 4.503,
      "mean_token_accuracy": 0.26487087160348893,
      "num_tokens": 16478484.0,
      "step": 7340
    },
    {
      "entropy": 5.281041288375855,
      "epoch": 7.9410492157923205,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010627040104829658,
      "loss": 4.5659,
      "mean_token_accuracy": 0.2536240413784981,
      "num_tokens": 16490546.0,
      "step": 7345
    },
    {
      "entropy": 5.337679767608643,
      "epoch": 7.9464575446187125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010598696052482759,
      "loss": 4.5879,
      "mean_token_accuracy": 0.253469780087471,
      "num_tokens": 16502503.0,
      "step": 7350
    },
    {
      "entropy": 5.29638409614563,
      "epoch": 7.951865873445105,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001057041341951347,
      "loss": 4.4814,
      "mean_token_accuracy": 0.2630394175648689,
      "num_tokens": 16515048.0,
      "step": 7355
    },
    {
      "entropy": 5.29359540939331,
      "epoch": 7.957274202271498,
      "grad_norm": 1.25,
      "learning_rate": 0.00010542192308700925,
      "loss": 4.4898,
      "mean_token_accuracy": 0.25899193435907364,
      "num_tokens": 16525723.0,
      "step": 7360
    },
    {
      "entropy": 5.283379030227661,
      "epoch": 7.962682531097891,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010514032822600733,
      "loss": 4.4412,
      "mean_token_accuracy": 0.27005851566791533,
      "num_tokens": 16535479.0,
      "step": 7365
    },
    {
      "entropy": 5.304073715209961,
      "epoch": 7.968090859924283,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010485935063544507,
      "loss": 4.5312,
      "mean_token_accuracy": 0.252606189250946,
      "num_tokens": 16546103.0,
      "step": 7370
    },
    {
      "entropy": 5.296210384368896,
      "epoch": 7.973499188750676,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010457899133639579,
      "loss": 4.5267,
      "mean_token_accuracy": 0.25252842754125593,
      "num_tokens": 16557655.0,
      "step": 7375
    },
    {
      "entropy": 5.250873851776123,
      "epoch": 7.978907517577069,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00010429925134768584,
      "loss": 4.5122,
      "mean_token_accuracy": 0.25824074149131776,
      "num_tokens": 16570069.0,
      "step": 7380
    },
    {
      "entropy": 5.31987771987915,
      "epoch": 7.984315846403462,
      "grad_norm": 1.21875,
      "learning_rate": 0.00010402013168589092,
      "loss": 4.5643,
      "mean_token_accuracy": 0.26417982429265974,
      "num_tokens": 16581838.0,
      "step": 7385
    },
    {
      "entropy": 5.290887498855591,
      "epoch": 7.9897241752298545,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010374163336533261,
      "loss": 4.5234,
      "mean_token_accuracy": 0.25693392008543015,
      "num_tokens": 16593093.0,
      "step": 7390
    },
    {
      "entropy": 5.288318252563476,
      "epoch": 7.995132504056246,
      "grad_norm": 1.25,
      "learning_rate": 0.00010346375739807448,
      "loss": 4.503,
      "mean_token_accuracy": 0.2617397397756577,
      "num_tokens": 16605178.0,
      "step": 7395
    },
    {
      "entropy": 5.237538867526585,
      "epoch": 8.0,
      "grad_norm": 2.53125,
      "learning_rate": 0.00010318650479391825,
      "loss": 4.4283,
      "mean_token_accuracy": 0.25736671023898655,
      "num_tokens": 16615112.0,
      "step": 7400
    },
    {
      "entropy": 5.262867879867554,
      "epoch": 8.005408328826393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00010290987656040076,
      "loss": 4.3814,
      "mean_token_accuracy": 0.26766972839832304,
      "num_tokens": 16626266.0,
      "step": 7405
    },
    {
      "entropy": 5.229944181442261,
      "epoch": 8.010816657652786,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010263387370278948,
      "loss": 4.3368,
      "mean_token_accuracy": 0.2781370908021927,
      "num_tokens": 16636086.0,
      "step": 7410
    },
    {
      "entropy": 5.302399015426635,
      "epoch": 8.016224986479179,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001023584972240795,
      "loss": 4.4762,
      "mean_token_accuracy": 0.2599455267190933,
      "num_tokens": 16646640.0,
      "step": 7415
    },
    {
      "entropy": 5.227972030639648,
      "epoch": 8.021633315305571,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010208374812498955,
      "loss": 4.3978,
      "mean_token_accuracy": 0.26698335707187654,
      "num_tokens": 16656509.0,
      "step": 7420
    },
    {
      "entropy": 5.271275806427002,
      "epoch": 8.027041644131963,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010180962740395853,
      "loss": 4.4593,
      "mean_token_accuracy": 0.2672934129834175,
      "num_tokens": 16668061.0,
      "step": 7425
    },
    {
      "entropy": 5.274111080169678,
      "epoch": 8.032449972958355,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010153613605714173,
      "loss": 4.3801,
      "mean_token_accuracy": 0.27344235181808474,
      "num_tokens": 16679588.0,
      "step": 7430
    },
    {
      "entropy": 5.3198981285095215,
      "epoch": 8.037858301784748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010126327507840744,
      "loss": 4.4614,
      "mean_token_accuracy": 0.26232910603284837,
      "num_tokens": 16690621.0,
      "step": 7435
    },
    {
      "entropy": 5.2781342506408695,
      "epoch": 8.043266630611141,
      "grad_norm": 1.15625,
      "learning_rate": 0.00010099104545933288,
      "loss": 4.4605,
      "mean_token_accuracy": 0.2660065948963165,
      "num_tokens": 16702504.0,
      "step": 7440
    },
    {
      "entropy": 5.197895288467407,
      "epoch": 8.048674959437534,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010071944818920127,
      "loss": 4.3718,
      "mean_token_accuracy": 0.26527439802885056,
      "num_tokens": 16712402.0,
      "step": 7445
    },
    {
      "entropy": 5.264601135253907,
      "epoch": 8.054083288263927,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001004484842549978,
      "loss": 4.4081,
      "mean_token_accuracy": 0.26822811365127563,
      "num_tokens": 16722057.0,
      "step": 7450
    },
    {
      "entropy": 5.2956307888031,
      "epoch": 8.05949161709032,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010017815464140583,
      "loss": 4.456,
      "mean_token_accuracy": 0.2696679443120956,
      "num_tokens": 16733966.0,
      "step": 7455
    },
    {
      "entropy": 5.251373147964477,
      "epoch": 8.064899945916713,
      "grad_norm": 1.3046875,
      "learning_rate": 9.990846033080389e-05,
      "loss": 4.4527,
      "mean_token_accuracy": 0.26384402364492415,
      "num_tokens": 16744975.0,
      "step": 7460
    },
    {
      "entropy": 5.265444612503051,
      "epoch": 8.070308274743104,
      "grad_norm": 1.2734375,
      "learning_rate": 9.963940230326176e-05,
      "loss": 4.3916,
      "mean_token_accuracy": 0.2660031422972679,
      "num_tokens": 16755539.0,
      "step": 7465
    },
    {
      "entropy": 5.300448179244995,
      "epoch": 8.075716603569496,
      "grad_norm": 1.3046875,
      "learning_rate": 9.937098153653687e-05,
      "loss": 4.5077,
      "mean_token_accuracy": 0.2601543188095093,
      "num_tokens": 16766524.0,
      "step": 7470
    },
    {
      "entropy": 5.308611154556274,
      "epoch": 8.08112493239589,
      "grad_norm": 1.109375,
      "learning_rate": 9.910319900607104e-05,
      "loss": 4.4933,
      "mean_token_accuracy": 0.2692590057849884,
      "num_tokens": 16779860.0,
      "step": 7475
    },
    {
      "entropy": 5.257492113113403,
      "epoch": 8.086533261222282,
      "grad_norm": 1.0625,
      "learning_rate": 9.883605568498635e-05,
      "loss": 4.434,
      "mean_token_accuracy": 0.26507392078638076,
      "num_tokens": 16792439.0,
      "step": 7480
    },
    {
      "entropy": 5.26414647102356,
      "epoch": 8.091941590048675,
      "grad_norm": 1.5078125,
      "learning_rate": 9.856955254408248e-05,
      "loss": 4.4055,
      "mean_token_accuracy": 0.2695435851812363,
      "num_tokens": 16802284.0,
      "step": 7485
    },
    {
      "entropy": 5.249683570861817,
      "epoch": 8.097349918875068,
      "grad_norm": 1.3984375,
      "learning_rate": 9.830369055183244e-05,
      "loss": 4.383,
      "mean_token_accuracy": 0.2745277762413025,
      "num_tokens": 16814141.0,
      "step": 7490
    },
    {
      "entropy": 5.377056980133057,
      "epoch": 8.10275824770146,
      "grad_norm": 1.1640625,
      "learning_rate": 9.80384706743792e-05,
      "loss": 4.5182,
      "mean_token_accuracy": 0.26022704392671586,
      "num_tokens": 16825222.0,
      "step": 7495
    },
    {
      "entropy": 5.2033610343933105,
      "epoch": 8.108166576527854,
      "grad_norm": 1.2890625,
      "learning_rate": 9.777389387553243e-05,
      "loss": 4.2799,
      "mean_token_accuracy": 0.27551053166389466,
      "num_tokens": 16836989.0,
      "step": 7500
    },
    {
      "entropy": 5.278771495819091,
      "epoch": 8.113574905354245,
      "grad_norm": 1.4921875,
      "learning_rate": 9.750996111676497e-05,
      "loss": 4.4539,
      "mean_token_accuracy": 0.27028381526470185,
      "num_tokens": 16848490.0,
      "step": 7505
    },
    {
      "entropy": 5.222098445892334,
      "epoch": 8.118983234180638,
      "grad_norm": 1.3359375,
      "learning_rate": 9.724667335720894e-05,
      "loss": 4.3971,
      "mean_token_accuracy": 0.27681301832199096,
      "num_tokens": 16859216.0,
      "step": 7510
    },
    {
      "entropy": 5.224892663955688,
      "epoch": 8.12439156300703,
      "grad_norm": 1.2734375,
      "learning_rate": 9.698403155365268e-05,
      "loss": 4.4985,
      "mean_token_accuracy": 0.2623336657881737,
      "num_tokens": 16870748.0,
      "step": 7515
    },
    {
      "entropy": 5.232490205764771,
      "epoch": 8.129799891833423,
      "grad_norm": 1.2109375,
      "learning_rate": 9.672203666053714e-05,
      "loss": 4.4366,
      "mean_token_accuracy": 0.2604313090443611,
      "num_tokens": 16881248.0,
      "step": 7520
    },
    {
      "entropy": 5.273748111724854,
      "epoch": 8.135208220659816,
      "grad_norm": 1.203125,
      "learning_rate": 9.646068962995235e-05,
      "loss": 4.391,
      "mean_token_accuracy": 0.2770028322935104,
      "num_tokens": 16893363.0,
      "step": 7525
    },
    {
      "entropy": 5.258162641525269,
      "epoch": 8.140616549486209,
      "grad_norm": 1.2421875,
      "learning_rate": 9.619999141163415e-05,
      "loss": 4.4316,
      "mean_token_accuracy": 0.26557767689228057,
      "num_tokens": 16904125.0,
      "step": 7530
    },
    {
      "entropy": 5.2424256801605225,
      "epoch": 8.146024878312602,
      "grad_norm": 1.2890625,
      "learning_rate": 9.593994295296024e-05,
      "loss": 4.4115,
      "mean_token_accuracy": 0.2757879078388214,
      "num_tokens": 16915587.0,
      "step": 7535
    },
    {
      "entropy": 5.262409591674805,
      "epoch": 8.151433207138995,
      "grad_norm": 1.3671875,
      "learning_rate": 9.56805451989474e-05,
      "loss": 4.4783,
      "mean_token_accuracy": 0.2641302540898323,
      "num_tokens": 16926729.0,
      "step": 7540
    },
    {
      "entropy": 5.2850995540618895,
      "epoch": 8.156841535965386,
      "grad_norm": 1.3125,
      "learning_rate": 9.54217990922478e-05,
      "loss": 4.4461,
      "mean_token_accuracy": 0.26386301070451734,
      "num_tokens": 16939032.0,
      "step": 7545
    },
    {
      "entropy": 5.224449968338012,
      "epoch": 8.162249864791779,
      "grad_norm": 1.2734375,
      "learning_rate": 9.516370557314525e-05,
      "loss": 4.3996,
      "mean_token_accuracy": 0.272972272336483,
      "num_tokens": 16950274.0,
      "step": 7550
    },
    {
      "entropy": 5.308033990859985,
      "epoch": 8.167658193618172,
      "grad_norm": 1.359375,
      "learning_rate": 9.490626557955221e-05,
      "loss": 4.4747,
      "mean_token_accuracy": 0.2653449773788452,
      "num_tokens": 16960636.0,
      "step": 7555
    },
    {
      "entropy": 5.2518449306488035,
      "epoch": 8.173066522444564,
      "grad_norm": 1.21875,
      "learning_rate": 9.464948004700626e-05,
      "loss": 4.3934,
      "mean_token_accuracy": 0.2658199265599251,
      "num_tokens": 16972360.0,
      "step": 7560
    },
    {
      "entropy": 5.245331478118897,
      "epoch": 8.178474851270957,
      "grad_norm": 1.28125,
      "learning_rate": 9.439334990866658e-05,
      "loss": 4.3919,
      "mean_token_accuracy": 0.2638754054903984,
      "num_tokens": 16983241.0,
      "step": 7565
    },
    {
      "entropy": 5.28767991065979,
      "epoch": 8.18388318009735,
      "grad_norm": 1.1796875,
      "learning_rate": 9.413787609531072e-05,
      "loss": 4.5047,
      "mean_token_accuracy": 0.2624309316277504,
      "num_tokens": 16994875.0,
      "step": 7570
    },
    {
      "entropy": 5.347606945037842,
      "epoch": 8.189291508923743,
      "grad_norm": 1.2109375,
      "learning_rate": 9.38830595353311e-05,
      "loss": 4.4919,
      "mean_token_accuracy": 0.2663153186440468,
      "num_tokens": 17006683.0,
      "step": 7575
    },
    {
      "entropy": 5.217359304428101,
      "epoch": 8.194699837750136,
      "grad_norm": 1.21875,
      "learning_rate": 9.362890115473161e-05,
      "loss": 4.4181,
      "mean_token_accuracy": 0.2673170804977417,
      "num_tokens": 17018018.0,
      "step": 7580
    },
    {
      "entropy": 5.2789397716522215,
      "epoch": 8.200108166576527,
      "grad_norm": 1.1953125,
      "learning_rate": 9.337540187712453e-05,
      "loss": 4.4613,
      "mean_token_accuracy": 0.2664405331015587,
      "num_tokens": 17029115.0,
      "step": 7585
    },
    {
      "entropy": 5.301168012619018,
      "epoch": 8.20551649540292,
      "grad_norm": 1.328125,
      "learning_rate": 9.312256262372674e-05,
      "loss": 4.4891,
      "mean_token_accuracy": 0.2602037563920021,
      "num_tokens": 17041325.0,
      "step": 7590
    },
    {
      "entropy": 5.336178112030029,
      "epoch": 8.210924824229313,
      "grad_norm": 1.2890625,
      "learning_rate": 9.28703843133566e-05,
      "loss": 4.5643,
      "mean_token_accuracy": 0.2603288248181343,
      "num_tokens": 17052392.0,
      "step": 7595
    },
    {
      "entropy": 5.281763887405395,
      "epoch": 8.216333153055706,
      "grad_norm": 1.3046875,
      "learning_rate": 9.261886786243087e-05,
      "loss": 4.4468,
      "mean_token_accuracy": 0.26056930273771284,
      "num_tokens": 17063424.0,
      "step": 7600
    },
    {
      "entropy": 5.246744871139526,
      "epoch": 8.221741481882098,
      "grad_norm": 1.3984375,
      "learning_rate": 9.236801418496071e-05,
      "loss": 4.3838,
      "mean_token_accuracy": 0.26994180381298066,
      "num_tokens": 17073706.0,
      "step": 7605
    },
    {
      "entropy": 5.304289960861206,
      "epoch": 8.227149810708491,
      "grad_norm": 1.4453125,
      "learning_rate": 9.21178241925491e-05,
      "loss": 4.5124,
      "mean_token_accuracy": 0.2515603303909302,
      "num_tokens": 17084448.0,
      "step": 7610
    },
    {
      "entropy": 5.251445865631103,
      "epoch": 8.232558139534884,
      "grad_norm": 1.375,
      "learning_rate": 9.186829879438696e-05,
      "loss": 4.3606,
      "mean_token_accuracy": 0.2791702464222908,
      "num_tokens": 17094288.0,
      "step": 7615
    },
    {
      "entropy": 5.2446675300598145,
      "epoch": 8.237966468361277,
      "grad_norm": 1.25,
      "learning_rate": 9.161943889725021e-05,
      "loss": 4.4611,
      "mean_token_accuracy": 0.26674124598503113,
      "num_tokens": 17105418.0,
      "step": 7620
    },
    {
      "entropy": 5.278581857681274,
      "epoch": 8.24337479718767,
      "grad_norm": 1.2578125,
      "learning_rate": 9.137124540549638e-05,
      "loss": 4.4853,
      "mean_token_accuracy": 0.25673758387565615,
      "num_tokens": 17116482.0,
      "step": 7625
    },
    {
      "entropy": 5.230424308776856,
      "epoch": 8.248783126014061,
      "grad_norm": 1.359375,
      "learning_rate": 9.1123719221061e-05,
      "loss": 4.3953,
      "mean_token_accuracy": 0.2835983008146286,
      "num_tokens": 17127501.0,
      "step": 7630
    },
    {
      "entropy": 5.3159096240997314,
      "epoch": 8.254191454840454,
      "grad_norm": 1.5234375,
      "learning_rate": 9.087686124345486e-05,
      "loss": 4.4551,
      "mean_token_accuracy": 0.26668847948312757,
      "num_tokens": 17138108.0,
      "step": 7635
    },
    {
      "entropy": 5.227493190765381,
      "epoch": 8.259599783666847,
      "grad_norm": 1.09375,
      "learning_rate": 9.063067236976055e-05,
      "loss": 4.4603,
      "mean_token_accuracy": 0.2671823799610138,
      "num_tokens": 17150302.0,
      "step": 7640
    },
    {
      "entropy": 5.263261413574218,
      "epoch": 8.26500811249324,
      "grad_norm": 1.3671875,
      "learning_rate": 9.038515349462887e-05,
      "loss": 4.4151,
      "mean_token_accuracy": 0.27597633600234983,
      "num_tokens": 17160530.0,
      "step": 7645
    },
    {
      "entropy": 5.295974445343018,
      "epoch": 8.270416441319632,
      "grad_norm": 1.2421875,
      "learning_rate": 9.014030551027601e-05,
      "loss": 4.4677,
      "mean_token_accuracy": 0.26183866113424303,
      "num_tokens": 17171484.0,
      "step": 7650
    },
    {
      "entropy": 5.229229211807251,
      "epoch": 8.275824770146025,
      "grad_norm": 1.40625,
      "learning_rate": 8.989612930648014e-05,
      "loss": 4.4013,
      "mean_token_accuracy": 0.2611523687839508,
      "num_tokens": 17182225.0,
      "step": 7655
    },
    {
      "entropy": 5.200949764251709,
      "epoch": 8.281233098972418,
      "grad_norm": 1.3203125,
      "learning_rate": 8.965262577057811e-05,
      "loss": 4.3666,
      "mean_token_accuracy": 0.27381090372800826,
      "num_tokens": 17193366.0,
      "step": 7660
    },
    {
      "entropy": 5.179641199111939,
      "epoch": 8.286641427798811,
      "grad_norm": 1.4140625,
      "learning_rate": 8.940979578746236e-05,
      "loss": 4.3555,
      "mean_token_accuracy": 0.27325813174247743,
      "num_tokens": 17203368.0,
      "step": 7665
    },
    {
      "entropy": 5.246249055862426,
      "epoch": 8.292049756625202,
      "grad_norm": 1.3046875,
      "learning_rate": 8.916764023957766e-05,
      "loss": 4.4051,
      "mean_token_accuracy": 0.27124026864767076,
      "num_tokens": 17214280.0,
      "step": 7670
    },
    {
      "entropy": 5.195784664154052,
      "epoch": 8.297458085451595,
      "grad_norm": 1.4140625,
      "learning_rate": 8.892616000691766e-05,
      "loss": 4.3973,
      "mean_token_accuracy": 0.2736124873161316,
      "num_tokens": 17225996.0,
      "step": 7675
    },
    {
      "entropy": 5.261968088150025,
      "epoch": 8.302866414277988,
      "grad_norm": 1.234375,
      "learning_rate": 8.868535596702232e-05,
      "loss": 4.4106,
      "mean_token_accuracy": 0.26900136321783064,
      "num_tokens": 17237519.0,
      "step": 7680
    },
    {
      "entropy": 5.295371770858765,
      "epoch": 8.30827474310438,
      "grad_norm": 1.34375,
      "learning_rate": 8.844522899497387e-05,
      "loss": 4.4577,
      "mean_token_accuracy": 0.26408430337905886,
      "num_tokens": 17249241.0,
      "step": 7685
    },
    {
      "entropy": 5.226178073883057,
      "epoch": 8.313683071930773,
      "grad_norm": 1.5078125,
      "learning_rate": 8.82057799633944e-05,
      "loss": 4.4389,
      "mean_token_accuracy": 0.2679689586162567,
      "num_tokens": 17259772.0,
      "step": 7690
    },
    {
      "entropy": 5.2919622421264645,
      "epoch": 8.319091400757166,
      "grad_norm": 1.2890625,
      "learning_rate": 8.796700974244218e-05,
      "loss": 4.4403,
      "mean_token_accuracy": 0.26347077637910843,
      "num_tokens": 17270403.0,
      "step": 7695
    },
    {
      "entropy": 5.300587272644043,
      "epoch": 8.32449972958356,
      "grad_norm": 1.21875,
      "learning_rate": 8.772891919980881e-05,
      "loss": 4.4504,
      "mean_token_accuracy": 0.26710911840200424,
      "num_tokens": 17281086.0,
      "step": 7700
    },
    {
      "entropy": 5.209040641784668,
      "epoch": 8.329908058409952,
      "grad_norm": 1.1640625,
      "learning_rate": 8.74915092007158e-05,
      "loss": 4.4075,
      "mean_token_accuracy": 0.2666740193963051,
      "num_tokens": 17291955.0,
      "step": 7705
    },
    {
      "entropy": 5.266893291473389,
      "epoch": 8.335316387236343,
      "grad_norm": 1.46875,
      "learning_rate": 8.725478060791175e-05,
      "loss": 4.4143,
      "mean_token_accuracy": 0.2725271493196487,
      "num_tokens": 17302513.0,
      "step": 7710
    },
    {
      "entropy": 5.249132966995239,
      "epoch": 8.340724716062736,
      "grad_norm": 1.40625,
      "learning_rate": 8.701873428166874e-05,
      "loss": 4.4725,
      "mean_token_accuracy": 0.269053652882576,
      "num_tokens": 17314312.0,
      "step": 7715
    },
    {
      "entropy": 5.191773414611816,
      "epoch": 8.346133044889129,
      "grad_norm": 1.328125,
      "learning_rate": 8.678337107977976e-05,
      "loss": 4.4158,
      "mean_token_accuracy": 0.27164565324783324,
      "num_tokens": 17325585.0,
      "step": 7720
    },
    {
      "entropy": 5.275494956970215,
      "epoch": 8.351541373715522,
      "grad_norm": 1.4140625,
      "learning_rate": 8.654869185755532e-05,
      "loss": 4.4509,
      "mean_token_accuracy": 0.26960608810186387,
      "num_tokens": 17336288.0,
      "step": 7725
    },
    {
      "entropy": 5.30947642326355,
      "epoch": 8.356949702541915,
      "grad_norm": 1.390625,
      "learning_rate": 8.631469746782011e-05,
      "loss": 4.5198,
      "mean_token_accuracy": 0.24734018743038177,
      "num_tokens": 17347378.0,
      "step": 7730
    },
    {
      "entropy": 5.142044639587402,
      "epoch": 8.362358031368307,
      "grad_norm": 1.390625,
      "learning_rate": 8.608138876091036e-05,
      "loss": 4.2802,
      "mean_token_accuracy": 0.28453837186098097,
      "num_tokens": 17357990.0,
      "step": 7735
    },
    {
      "entropy": 5.273561906814575,
      "epoch": 8.3677663601947,
      "grad_norm": 1.296875,
      "learning_rate": 8.584876658467041e-05,
      "loss": 4.3592,
      "mean_token_accuracy": 0.2719022586941719,
      "num_tokens": 17368891.0,
      "step": 7740
    },
    {
      "entropy": 5.300571012496948,
      "epoch": 8.373174689021093,
      "grad_norm": 1.2578125,
      "learning_rate": 8.561683178444979e-05,
      "loss": 4.4655,
      "mean_token_accuracy": 0.2622487723827362,
      "num_tokens": 17380291.0,
      "step": 7745
    },
    {
      "entropy": 5.318579006195068,
      "epoch": 8.378583017847484,
      "grad_norm": 1.2890625,
      "learning_rate": 8.53855852031002e-05,
      "loss": 4.5172,
      "mean_token_accuracy": 0.26102431267499926,
      "num_tokens": 17390293.0,
      "step": 7750
    },
    {
      "entropy": 5.2953681468963625,
      "epoch": 8.383991346673877,
      "grad_norm": 1.40625,
      "learning_rate": 8.515502768097204e-05,
      "loss": 4.502,
      "mean_token_accuracy": 0.2643964484333992,
      "num_tokens": 17401707.0,
      "step": 7755
    },
    {
      "entropy": 5.27127103805542,
      "epoch": 8.38939967550027,
      "grad_norm": 1.296875,
      "learning_rate": 8.492516005591204e-05,
      "loss": 4.4274,
      "mean_token_accuracy": 0.26136756837368014,
      "num_tokens": 17412369.0,
      "step": 7760
    },
    {
      "entropy": 5.28918833732605,
      "epoch": 8.394808004326663,
      "grad_norm": 1.375,
      "learning_rate": 8.469598316325971e-05,
      "loss": 4.4728,
      "mean_token_accuracy": 0.25418974459171295,
      "num_tokens": 17423562.0,
      "step": 7765
    },
    {
      "entropy": 5.284426307678222,
      "epoch": 8.400216333153056,
      "grad_norm": 1.2421875,
      "learning_rate": 8.446749783584432e-05,
      "loss": 4.4641,
      "mean_token_accuracy": 0.2607898622751236,
      "num_tokens": 17435587.0,
      "step": 7770
    },
    {
      "entropy": 5.274924945831299,
      "epoch": 8.405624661979449,
      "grad_norm": 1.328125,
      "learning_rate": 8.423970490398206e-05,
      "loss": 4.4694,
      "mean_token_accuracy": 0.2613364353775978,
      "num_tokens": 17446718.0,
      "step": 7775
    },
    {
      "entropy": 5.287240600585937,
      "epoch": 8.411032990805841,
      "grad_norm": 1.4921875,
      "learning_rate": 8.401260519547305e-05,
      "loss": 4.4526,
      "mean_token_accuracy": 0.27272864282131193,
      "num_tokens": 17456674.0,
      "step": 7780
    },
    {
      "entropy": 5.236607027053833,
      "epoch": 8.416441319632234,
      "grad_norm": 1.296875,
      "learning_rate": 8.378619953559807e-05,
      "loss": 4.4294,
      "mean_token_accuracy": 0.26108056455850603,
      "num_tokens": 17467289.0,
      "step": 7785
    },
    {
      "entropy": 5.233558320999146,
      "epoch": 8.421849648458625,
      "grad_norm": 1.28125,
      "learning_rate": 8.356048874711589e-05,
      "loss": 4.4736,
      "mean_token_accuracy": 0.26244513839483263,
      "num_tokens": 17478533.0,
      "step": 7790
    },
    {
      "entropy": 5.303229475021363,
      "epoch": 8.427257977285018,
      "grad_norm": 1.3515625,
      "learning_rate": 8.333547365026e-05,
      "loss": 4.5562,
      "mean_token_accuracy": 0.2564991101622581,
      "num_tokens": 17489967.0,
      "step": 7795
    },
    {
      "entropy": 5.22686357498169,
      "epoch": 8.432666306111411,
      "grad_norm": 1.265625,
      "learning_rate": 8.311115506273579e-05,
      "loss": 4.4414,
      "mean_token_accuracy": 0.26282148212194445,
      "num_tokens": 17501521.0,
      "step": 7800
    },
    {
      "entropy": 5.335459566116333,
      "epoch": 8.438074634937804,
      "grad_norm": 1.34375,
      "learning_rate": 8.288753379971756e-05,
      "loss": 4.5108,
      "mean_token_accuracy": 0.260001565515995,
      "num_tokens": 17513235.0,
      "step": 7805
    },
    {
      "entropy": 5.288670921325684,
      "epoch": 8.443482963764197,
      "grad_norm": 1.3203125,
      "learning_rate": 8.26646106738454e-05,
      "loss": 4.4164,
      "mean_token_accuracy": 0.2734836250543594,
      "num_tokens": 17524777.0,
      "step": 7810
    },
    {
      "entropy": 5.226439762115478,
      "epoch": 8.44889129259059,
      "grad_norm": 1.421875,
      "learning_rate": 8.244238649522249e-05,
      "loss": 4.3916,
      "mean_token_accuracy": 0.274645034968853,
      "num_tokens": 17535956.0,
      "step": 7815
    },
    {
      "entropy": 5.281897068023682,
      "epoch": 8.454299621416983,
      "grad_norm": 1.1484375,
      "learning_rate": 8.222086207141214e-05,
      "loss": 4.4564,
      "mean_token_accuracy": 0.2662057265639305,
      "num_tokens": 17549574.0,
      "step": 7820
    },
    {
      "entropy": 5.25428352355957,
      "epoch": 8.459707950243375,
      "grad_norm": 1.3046875,
      "learning_rate": 8.200003820743449e-05,
      "loss": 4.429,
      "mean_token_accuracy": 0.26675207167863846,
      "num_tokens": 17559999.0,
      "step": 7825
    },
    {
      "entropy": 5.312931489944458,
      "epoch": 8.465116279069768,
      "grad_norm": 1.265625,
      "learning_rate": 8.177991570576396e-05,
      "loss": 4.5008,
      "mean_token_accuracy": 0.2570738151669502,
      "num_tokens": 17573351.0,
      "step": 7830
    },
    {
      "entropy": 5.355503177642822,
      "epoch": 8.47052460789616,
      "grad_norm": 1.2421875,
      "learning_rate": 8.156049536632629e-05,
      "loss": 4.5575,
      "mean_token_accuracy": 0.26188210397958755,
      "num_tokens": 17585571.0,
      "step": 7835
    },
    {
      "entropy": 5.185672998428345,
      "epoch": 8.475932936722552,
      "grad_norm": 1.3203125,
      "learning_rate": 8.13417779864955e-05,
      "loss": 4.3962,
      "mean_token_accuracy": 0.26869903355836866,
      "num_tokens": 17597187.0,
      "step": 7840
    },
    {
      "entropy": 5.292581844329834,
      "epoch": 8.481341265548945,
      "grad_norm": 1.34375,
      "learning_rate": 8.112376436109107e-05,
      "loss": 4.4818,
      "mean_token_accuracy": 0.26242540776729584,
      "num_tokens": 17608017.0,
      "step": 7845
    },
    {
      "entropy": 5.270896577835083,
      "epoch": 8.486749594375338,
      "grad_norm": 1.3828125,
      "learning_rate": 8.090645528237493e-05,
      "loss": 4.4799,
      "mean_token_accuracy": 0.2663352757692337,
      "num_tokens": 17619001.0,
      "step": 7850
    },
    {
      "entropy": 5.355388355255127,
      "epoch": 8.49215792320173,
      "grad_norm": 1.421875,
      "learning_rate": 8.068985154004879e-05,
      "loss": 4.5607,
      "mean_token_accuracy": 0.25888815224170686,
      "num_tokens": 17628925.0,
      "step": 7855
    },
    {
      "entropy": 5.221790790557861,
      "epoch": 8.497566252028124,
      "grad_norm": 1.4609375,
      "learning_rate": 8.047395392125126e-05,
      "loss": 4.3545,
      "mean_token_accuracy": 0.27896223962306976,
      "num_tokens": 17639266.0,
      "step": 7860
    },
    {
      "entropy": 5.2215910911560055,
      "epoch": 8.502974580854517,
      "grad_norm": 1.3125,
      "learning_rate": 8.02587632105547e-05,
      "loss": 4.4063,
      "mean_token_accuracy": 0.2767242968082428,
      "num_tokens": 17650028.0,
      "step": 7865
    },
    {
      "entropy": 5.224254846572876,
      "epoch": 8.50838290968091,
      "grad_norm": 1.296875,
      "learning_rate": 8.004428018996264e-05,
      "loss": 4.4415,
      "mean_token_accuracy": 0.2647675424814224,
      "num_tokens": 17662150.0,
      "step": 7870
    },
    {
      "entropy": 5.278802871704102,
      "epoch": 8.5137912385073,
      "grad_norm": 1.3515625,
      "learning_rate": 7.983050563890692e-05,
      "loss": 4.4375,
      "mean_token_accuracy": 0.2695013627409935,
      "num_tokens": 17672820.0,
      "step": 7875
    },
    {
      "entropy": 5.298423194885254,
      "epoch": 8.519199567333693,
      "grad_norm": 1.5078125,
      "learning_rate": 7.961744033424473e-05,
      "loss": 4.5383,
      "mean_token_accuracy": 0.25984006077051164,
      "num_tokens": 17683071.0,
      "step": 7880
    },
    {
      "entropy": 5.244650077819824,
      "epoch": 8.524607896160086,
      "grad_norm": 1.3828125,
      "learning_rate": 7.940508505025586e-05,
      "loss": 4.4387,
      "mean_token_accuracy": 0.26784766614437105,
      "num_tokens": 17694352.0,
      "step": 7885
    },
    {
      "entropy": 5.273040866851806,
      "epoch": 8.530016224986479,
      "grad_norm": 1.15625,
      "learning_rate": 7.919344055863997e-05,
      "loss": 4.4534,
      "mean_token_accuracy": 0.26143433451652526,
      "num_tokens": 17706613.0,
      "step": 7890
    },
    {
      "entropy": 5.298657178878784,
      "epoch": 8.535424553812872,
      "grad_norm": 1.4921875,
      "learning_rate": 7.89825076285135e-05,
      "loss": 4.4305,
      "mean_token_accuracy": 0.26394264549016955,
      "num_tokens": 17717171.0,
      "step": 7895
    },
    {
      "entropy": 5.26628942489624,
      "epoch": 8.540832882639265,
      "grad_norm": 1.4296875,
      "learning_rate": 7.877228702640739e-05,
      "loss": 4.4612,
      "mean_token_accuracy": 0.2650628849864006,
      "num_tokens": 17728183.0,
      "step": 7900
    },
    {
      "entropy": 5.271047782897949,
      "epoch": 8.546241211465658,
      "grad_norm": 1.3984375,
      "learning_rate": 7.856277951626363e-05,
      "loss": 4.4445,
      "mean_token_accuracy": 0.2690369367599487,
      "num_tokens": 17740478.0,
      "step": 7905
    },
    {
      "entropy": 5.185982418060303,
      "epoch": 8.55164954029205,
      "grad_norm": 1.1484375,
      "learning_rate": 7.835398585943307e-05,
      "loss": 4.354,
      "mean_token_accuracy": 0.2689744636416435,
      "num_tokens": 17752183.0,
      "step": 7910
    },
    {
      "entropy": 5.273795509338379,
      "epoch": 8.557057869118442,
      "grad_norm": 1.3828125,
      "learning_rate": 7.814590681467235e-05,
      "loss": 4.4972,
      "mean_token_accuracy": 0.26827667355537416,
      "num_tokens": 17762858.0,
      "step": 7915
    },
    {
      "entropy": 5.313104581832886,
      "epoch": 8.562466197944834,
      "grad_norm": 1.40625,
      "learning_rate": 7.793854313814114e-05,
      "loss": 4.5176,
      "mean_token_accuracy": 0.2640592038631439,
      "num_tokens": 17774307.0,
      "step": 7920
    },
    {
      "entropy": 5.258284282684326,
      "epoch": 8.567874526771227,
      "grad_norm": 1.265625,
      "learning_rate": 7.773189558339956e-05,
      "loss": 4.3747,
      "mean_token_accuracy": 0.2768690586090088,
      "num_tokens": 17786506.0,
      "step": 7925
    },
    {
      "entropy": 5.29032940864563,
      "epoch": 8.57328285559762,
      "grad_norm": 1.390625,
      "learning_rate": 7.752596490140537e-05,
      "loss": 4.434,
      "mean_token_accuracy": 0.26769240498542785,
      "num_tokens": 17796764.0,
      "step": 7930
    },
    {
      "entropy": 5.2506941795349125,
      "epoch": 8.578691184424013,
      "grad_norm": 1.34375,
      "learning_rate": 7.732075184051098e-05,
      "loss": 4.4214,
      "mean_token_accuracy": 0.26963243931531905,
      "num_tokens": 17807922.0,
      "step": 7935
    },
    {
      "entropy": 5.273650360107422,
      "epoch": 8.584099513250406,
      "grad_norm": 1.1796875,
      "learning_rate": 7.711625714646127e-05,
      "loss": 4.5034,
      "mean_token_accuracy": 0.256424143910408,
      "num_tokens": 17820000.0,
      "step": 7940
    },
    {
      "entropy": 5.2234714984893795,
      "epoch": 8.589507842076799,
      "grad_norm": 1.3359375,
      "learning_rate": 7.691248156239041e-05,
      "loss": 4.4129,
      "mean_token_accuracy": 0.27300242632627486,
      "num_tokens": 17832584.0,
      "step": 7945
    },
    {
      "entropy": 5.278550052642823,
      "epoch": 8.594916170903192,
      "grad_norm": 1.3203125,
      "learning_rate": 7.670942582881931e-05,
      "loss": 4.4258,
      "mean_token_accuracy": 0.2664855748414993,
      "num_tokens": 17843978.0,
      "step": 7950
    },
    {
      "entropy": 5.292669773101807,
      "epoch": 8.600324499729584,
      "grad_norm": 1.34375,
      "learning_rate": 7.650709068365295e-05,
      "loss": 4.4925,
      "mean_token_accuracy": 0.26284020990133283,
      "num_tokens": 17854820.0,
      "step": 7955
    },
    {
      "entropy": 5.274573183059692,
      "epoch": 8.605732828555976,
      "grad_norm": 1.2265625,
      "learning_rate": 7.630547686217774e-05,
      "loss": 4.5117,
      "mean_token_accuracy": 0.26464353799819945,
      "num_tokens": 17867988.0,
      "step": 7960
    },
    {
      "entropy": 5.271131753921509,
      "epoch": 8.611141157382368,
      "grad_norm": 1.2265625,
      "learning_rate": 7.61045850970588e-05,
      "loss": 4.466,
      "mean_token_accuracy": 0.26941823661327363,
      "num_tokens": 17880727.0,
      "step": 7965
    },
    {
      "entropy": 5.3468616008758545,
      "epoch": 8.616549486208761,
      "grad_norm": 1.4140625,
      "learning_rate": 7.59044161183373e-05,
      "loss": 4.5025,
      "mean_token_accuracy": 0.26479560136795044,
      "num_tokens": 17891949.0,
      "step": 7970
    },
    {
      "entropy": 5.2756554126739506,
      "epoch": 8.621957815035154,
      "grad_norm": 1.15625,
      "learning_rate": 7.570497065342762e-05,
      "loss": 4.4752,
      "mean_token_accuracy": 0.2645415022969246,
      "num_tokens": 17904436.0,
      "step": 7975
    },
    {
      "entropy": 5.257435035705567,
      "epoch": 8.627366143861547,
      "grad_norm": 1.328125,
      "learning_rate": 7.550624942711517e-05,
      "loss": 4.4786,
      "mean_token_accuracy": 0.263490591943264,
      "num_tokens": 17917051.0,
      "step": 7980
    },
    {
      "entropy": 5.2463866710662845,
      "epoch": 8.63277447268794,
      "grad_norm": 1.4453125,
      "learning_rate": 7.530825316155337e-05,
      "loss": 4.4279,
      "mean_token_accuracy": 0.2613639160990715,
      "num_tokens": 17928331.0,
      "step": 7985
    },
    {
      "entropy": 5.252237176895141,
      "epoch": 8.638182801514333,
      "grad_norm": 1.359375,
      "learning_rate": 7.511098257626101e-05,
      "loss": 4.4224,
      "mean_token_accuracy": 0.2620156958699226,
      "num_tokens": 17939212.0,
      "step": 7990
    },
    {
      "entropy": 5.232524967193603,
      "epoch": 8.643591130340724,
      "grad_norm": 1.421875,
      "learning_rate": 7.491443838811991e-05,
      "loss": 4.4203,
      "mean_token_accuracy": 0.26305613219738005,
      "num_tokens": 17949366.0,
      "step": 7995
    },
    {
      "entropy": 5.2436666011810305,
      "epoch": 8.648999459167117,
      "grad_norm": 1.2265625,
      "learning_rate": 7.47186213113721e-05,
      "loss": 4.4241,
      "mean_token_accuracy": 0.2639735236763954,
      "num_tokens": 17959983.0,
      "step": 8000
    },
    {
      "epoch": 8.648999459167117,
      "eval_entropy": 5.184521691365675,
      "eval_loss": 5.8559346199035645,
      "eval_mean_token_accuracy": 0.18259730042720382,
      "eval_num_tokens": 17959983.0,
      "eval_runtime": 1.9398,
      "eval_samples_per_second": 1812.054,
      "eval_steps_per_second": 226.829,
      "step": 8000
    },
    {
      "entropy": 5.180384397506714,
      "epoch": 8.65440778799351,
      "grad_norm": 1.25,
      "learning_rate": 7.452353205761725e-05,
      "loss": 4.3416,
      "mean_token_accuracy": 0.2822924882173538,
      "num_tokens": 17970710.0,
      "step": 8005
    },
    {
      "entropy": 5.275919580459595,
      "epoch": 8.659816116819902,
      "grad_norm": 1.28125,
      "learning_rate": 7.432917133581019e-05,
      "loss": 4.5221,
      "mean_token_accuracy": 0.25931547582149506,
      "num_tokens": 17982291.0,
      "step": 8010
    },
    {
      "entropy": 5.207006883621216,
      "epoch": 8.665224445646295,
      "grad_norm": 1.4375,
      "learning_rate": 7.413553985225824e-05,
      "loss": 4.4238,
      "mean_token_accuracy": 0.27249220460653306,
      "num_tokens": 17993650.0,
      "step": 8015
    },
    {
      "entropy": 5.204651403427124,
      "epoch": 8.670632774472688,
      "grad_norm": 1.3203125,
      "learning_rate": 7.394263831061865e-05,
      "loss": 4.4395,
      "mean_token_accuracy": 0.2713017761707306,
      "num_tokens": 18004913.0,
      "step": 8020
    },
    {
      "entropy": 5.248039436340332,
      "epoch": 8.676041103299081,
      "grad_norm": 1.328125,
      "learning_rate": 7.375046741189611e-05,
      "loss": 4.4412,
      "mean_token_accuracy": 0.26687727570533754,
      "num_tokens": 18015821.0,
      "step": 8025
    },
    {
      "entropy": 5.2869916439056395,
      "epoch": 8.681449432125474,
      "grad_norm": 1.390625,
      "learning_rate": 7.355902785444002e-05,
      "loss": 4.4329,
      "mean_token_accuracy": 0.27145728319883344,
      "num_tokens": 18027237.0,
      "step": 8030
    },
    {
      "entropy": 5.350145578384399,
      "epoch": 8.686857760951867,
      "grad_norm": 1.171875,
      "learning_rate": 7.336832033394217e-05,
      "loss": 4.5621,
      "mean_token_accuracy": 0.2583585143089294,
      "num_tokens": 18039003.0,
      "step": 8035
    },
    {
      "entropy": 5.210152101516724,
      "epoch": 8.692266089778258,
      "grad_norm": 1.4375,
      "learning_rate": 7.317834554343429e-05,
      "loss": 4.4031,
      "mean_token_accuracy": 0.2710442364215851,
      "num_tokens": 18049588.0,
      "step": 8040
    },
    {
      "entropy": 5.240427827835083,
      "epoch": 8.69767441860465,
      "grad_norm": 1.1328125,
      "learning_rate": 7.298910417328503e-05,
      "loss": 4.4159,
      "mean_token_accuracy": 0.2695280656218529,
      "num_tokens": 18061938.0,
      "step": 8045
    },
    {
      "entropy": 5.2455394744873045,
      "epoch": 8.703082747431043,
      "grad_norm": 1.4921875,
      "learning_rate": 7.280059691119806e-05,
      "loss": 4.4317,
      "mean_token_accuracy": 0.2637948840856552,
      "num_tokens": 18073148.0,
      "step": 8050
    },
    {
      "entropy": 5.210743141174317,
      "epoch": 8.708491076257436,
      "grad_norm": 1.3125,
      "learning_rate": 7.261282444220917e-05,
      "loss": 4.4503,
      "mean_token_accuracy": 0.26324902921915055,
      "num_tokens": 18084826.0,
      "step": 8055
    },
    {
      "entropy": 5.212789154052734,
      "epoch": 8.71389940508383,
      "grad_norm": 1.2578125,
      "learning_rate": 7.242578744868395e-05,
      "loss": 4.371,
      "mean_token_accuracy": 0.27760329097509384,
      "num_tokens": 18095299.0,
      "step": 8060
    },
    {
      "entropy": 5.234632158279419,
      "epoch": 8.719307733910222,
      "grad_norm": 1.28125,
      "learning_rate": 7.223948661031534e-05,
      "loss": 4.3982,
      "mean_token_accuracy": 0.28325620889663694,
      "num_tokens": 18106904.0,
      "step": 8065
    },
    {
      "entropy": 5.255150222778321,
      "epoch": 8.724716062736615,
      "grad_norm": 1.3515625,
      "learning_rate": 7.205392260412088e-05,
      "loss": 4.3845,
      "mean_token_accuracy": 0.27937616407871246,
      "num_tokens": 18117967.0,
      "step": 8070
    },
    {
      "entropy": 5.211847019195557,
      "epoch": 8.730124391563008,
      "grad_norm": 1.328125,
      "learning_rate": 7.186909610444065e-05,
      "loss": 4.4366,
      "mean_token_accuracy": 0.268591533601284,
      "num_tokens": 18129067.0,
      "step": 8075
    },
    {
      "entropy": 5.261353206634522,
      "epoch": 8.735532720389399,
      "grad_norm": 1.328125,
      "learning_rate": 7.168500778293466e-05,
      "loss": 4.444,
      "mean_token_accuracy": 0.27177583128213884,
      "num_tokens": 18139104.0,
      "step": 8080
    },
    {
      "entropy": 5.241223382949829,
      "epoch": 8.740941049215792,
      "grad_norm": 1.2734375,
      "learning_rate": 7.15016583085802e-05,
      "loss": 4.4124,
      "mean_token_accuracy": 0.2636688783764839,
      "num_tokens": 18149986.0,
      "step": 8085
    },
    {
      "entropy": 5.2680816650390625,
      "epoch": 8.746349378042185,
      "grad_norm": 1.28125,
      "learning_rate": 7.131904834766973e-05,
      "loss": 4.5283,
      "mean_token_accuracy": 0.25626952201128006,
      "num_tokens": 18162639.0,
      "step": 8090
    },
    {
      "entropy": 5.300923109054565,
      "epoch": 8.751757706868577,
      "grad_norm": 1.1328125,
      "learning_rate": 7.113717856380824e-05,
      "loss": 4.5539,
      "mean_token_accuracy": 0.25542216151952746,
      "num_tokens": 18175807.0,
      "step": 8095
    },
    {
      "entropy": 5.244313764572143,
      "epoch": 8.75716603569497,
      "grad_norm": 1.3359375,
      "learning_rate": 7.095604961791095e-05,
      "loss": 4.499,
      "mean_token_accuracy": 0.26093165278434755,
      "num_tokens": 18188150.0,
      "step": 8100
    },
    {
      "entropy": 5.2729137420654295,
      "epoch": 8.762574364521363,
      "grad_norm": 1.3125,
      "learning_rate": 7.077566216820089e-05,
      "loss": 4.3998,
      "mean_token_accuracy": 0.27213784456253054,
      "num_tokens": 18198676.0,
      "step": 8105
    },
    {
      "entropy": 5.248143196105957,
      "epoch": 8.767982693347756,
      "grad_norm": 1.265625,
      "learning_rate": 7.059601687020647e-05,
      "loss": 4.4057,
      "mean_token_accuracy": 0.2706846743822098,
      "num_tokens": 18209424.0,
      "step": 8110
    },
    {
      "entropy": 5.221116828918457,
      "epoch": 8.773391022174149,
      "grad_norm": 1.25,
      "learning_rate": 7.041711437675908e-05,
      "loss": 4.3951,
      "mean_token_accuracy": 0.27077078968286517,
      "num_tokens": 18221143.0,
      "step": 8115
    },
    {
      "entropy": 5.222454357147217,
      "epoch": 8.77879935100054,
      "grad_norm": 1.3359375,
      "learning_rate": 7.023895533799087e-05,
      "loss": 4.4508,
      "mean_token_accuracy": 0.2738396689295769,
      "num_tokens": 18231812.0,
      "step": 8120
    },
    {
      "entropy": 5.2647288799285885,
      "epoch": 8.784207679826933,
      "grad_norm": 1.3671875,
      "learning_rate": 7.006154040133211e-05,
      "loss": 4.5376,
      "mean_token_accuracy": 0.25880497694015503,
      "num_tokens": 18244117.0,
      "step": 8125
    },
    {
      "entropy": 5.211640214920044,
      "epoch": 8.789616008653326,
      "grad_norm": 1.46875,
      "learning_rate": 6.98848702115091e-05,
      "loss": 4.4404,
      "mean_token_accuracy": 0.2720267608761787,
      "num_tokens": 18254751.0,
      "step": 8130
    },
    {
      "entropy": 5.310969829559326,
      "epoch": 8.795024337479719,
      "grad_norm": 1.3203125,
      "learning_rate": 6.970894541054182e-05,
      "loss": 4.5032,
      "mean_token_accuracy": 0.2615102991461754,
      "num_tokens": 18265759.0,
      "step": 8135
    },
    {
      "entropy": 5.282863664627075,
      "epoch": 8.800432666306111,
      "grad_norm": 1.4375,
      "learning_rate": 6.953376663774129e-05,
      "loss": 4.4588,
      "mean_token_accuracy": 0.266700042784214,
      "num_tokens": 18275887.0,
      "step": 8140
    },
    {
      "entropy": 5.23297905921936,
      "epoch": 8.805840995132504,
      "grad_norm": 1.09375,
      "learning_rate": 6.935933452970762e-05,
      "loss": 4.4796,
      "mean_token_accuracy": 0.2671940267086029,
      "num_tokens": 18288455.0,
      "step": 8145
    },
    {
      "entropy": 5.295588064193725,
      "epoch": 8.811249323958897,
      "grad_norm": 1.234375,
      "learning_rate": 6.918564972032745e-05,
      "loss": 4.4742,
      "mean_token_accuracy": 0.26502630263566973,
      "num_tokens": 18299135.0,
      "step": 8150
    },
    {
      "entropy": 5.342410755157471,
      "epoch": 8.81665765278529,
      "grad_norm": 1.296875,
      "learning_rate": 6.901271284077188e-05,
      "loss": 4.5882,
      "mean_token_accuracy": 0.2515240773558617,
      "num_tokens": 18310383.0,
      "step": 8155
    },
    {
      "entropy": 5.266007280349731,
      "epoch": 8.822065981611683,
      "grad_norm": 1.3203125,
      "learning_rate": 6.884052451949382e-05,
      "loss": 4.4293,
      "mean_token_accuracy": 0.26586702316999433,
      "num_tokens": 18321931.0,
      "step": 8160
    },
    {
      "entropy": 5.211520671844482,
      "epoch": 8.827474310438074,
      "grad_norm": 1.390625,
      "learning_rate": 6.866908538222611e-05,
      "loss": 4.3069,
      "mean_token_accuracy": 0.2854560285806656,
      "num_tokens": 18332122.0,
      "step": 8165
    },
    {
      "entropy": 5.220061254501343,
      "epoch": 8.832882639264467,
      "grad_norm": 1.390625,
      "learning_rate": 6.84983960519789e-05,
      "loss": 4.4276,
      "mean_token_accuracy": 0.26214383244514466,
      "num_tokens": 18343262.0,
      "step": 8170
    },
    {
      "entropy": 5.2289591312408445,
      "epoch": 8.83829096809086,
      "grad_norm": 1.34375,
      "learning_rate": 6.83284571490377e-05,
      "loss": 4.4266,
      "mean_token_accuracy": 0.2736802875995636,
      "num_tokens": 18353723.0,
      "step": 8175
    },
    {
      "entropy": 5.293781566619873,
      "epoch": 8.843699296917253,
      "grad_norm": 1.5078125,
      "learning_rate": 6.815926929096077e-05,
      "loss": 4.4709,
      "mean_token_accuracy": 0.25810123383998873,
      "num_tokens": 18365258.0,
      "step": 8180
    },
    {
      "entropy": 5.239980840682984,
      "epoch": 8.849107625743645,
      "grad_norm": 1.2265625,
      "learning_rate": 6.799083309257726e-05,
      "loss": 4.5078,
      "mean_token_accuracy": 0.2589730069041252,
      "num_tokens": 18378037.0,
      "step": 8185
    },
    {
      "entropy": 5.215294551849365,
      "epoch": 8.854515954570038,
      "grad_norm": 1.1953125,
      "learning_rate": 6.782314916598465e-05,
      "loss": 4.4103,
      "mean_token_accuracy": 0.26488734781742096,
      "num_tokens": 18390215.0,
      "step": 8190
    },
    {
      "entropy": 5.245328807830811,
      "epoch": 8.859924283396431,
      "grad_norm": 1.46875,
      "learning_rate": 6.765621812054675e-05,
      "loss": 4.3891,
      "mean_token_accuracy": 0.2729858785867691,
      "num_tokens": 18401462.0,
      "step": 8195
    },
    {
      "entropy": 5.269210577011108,
      "epoch": 8.865332612222822,
      "grad_norm": 1.421875,
      "learning_rate": 6.749004056289132e-05,
      "loss": 4.4278,
      "mean_token_accuracy": 0.2688741058111191,
      "num_tokens": 18412157.0,
      "step": 8200
    },
    {
      "entropy": 5.21815767288208,
      "epoch": 8.870740941049215,
      "grad_norm": 1.4140625,
      "learning_rate": 6.732461709690811e-05,
      "loss": 4.3508,
      "mean_token_accuracy": 0.27718335539102557,
      "num_tokens": 18423074.0,
      "step": 8205
    },
    {
      "entropy": 5.218471384048462,
      "epoch": 8.876149269875608,
      "grad_norm": 1.28125,
      "learning_rate": 6.715994832374619e-05,
      "loss": 4.4341,
      "mean_token_accuracy": 0.2639485985040665,
      "num_tokens": 18434377.0,
      "step": 8210
    },
    {
      "entropy": 5.267412948608398,
      "epoch": 8.881557598702,
      "grad_norm": 1.390625,
      "learning_rate": 6.699603484181245e-05,
      "loss": 4.4966,
      "mean_token_accuracy": 0.26500020921230316,
      "num_tokens": 18444890.0,
      "step": 8215
    },
    {
      "entropy": 5.254890584945679,
      "epoch": 8.886965927528394,
      "grad_norm": 1.09375,
      "learning_rate": 6.683287724676874e-05,
      "loss": 4.4684,
      "mean_token_accuracy": 0.2713591828942299,
      "num_tokens": 18457485.0,
      "step": 8220
    },
    {
      "entropy": 5.280308532714844,
      "epoch": 8.892374256354787,
      "grad_norm": 1.1953125,
      "learning_rate": 6.667047613153018e-05,
      "loss": 4.474,
      "mean_token_accuracy": 0.26754706650972365,
      "num_tokens": 18468628.0,
      "step": 8225
    },
    {
      "entropy": 5.1966087341308596,
      "epoch": 8.89778258518118,
      "grad_norm": 1.3203125,
      "learning_rate": 6.650883208626275e-05,
      "loss": 4.4303,
      "mean_token_accuracy": 0.26456423550844194,
      "num_tokens": 18480413.0,
      "step": 8230
    },
    {
      "entropy": 5.289001035690307,
      "epoch": 8.903190914007572,
      "grad_norm": 1.3984375,
      "learning_rate": 6.634794569838136e-05,
      "loss": 4.4647,
      "mean_token_accuracy": 0.26172349601984024,
      "num_tokens": 18491154.0,
      "step": 8235
    },
    {
      "entropy": 5.223420333862305,
      "epoch": 8.908599242833965,
      "grad_norm": 1.4375,
      "learning_rate": 6.618781755254749e-05,
      "loss": 4.4427,
      "mean_token_accuracy": 0.2713198721408844,
      "num_tokens": 18500924.0,
      "step": 8240
    },
    {
      "entropy": 5.310253429412842,
      "epoch": 8.914007571660356,
      "grad_norm": 1.3984375,
      "learning_rate": 6.60284482306672e-05,
      "loss": 4.469,
      "mean_token_accuracy": 0.2668606534600258,
      "num_tokens": 18511835.0,
      "step": 8245
    },
    {
      "entropy": 5.276546669006348,
      "epoch": 8.919415900486749,
      "grad_norm": 1.375,
      "learning_rate": 6.586983831188888e-05,
      "loss": 4.5068,
      "mean_token_accuracy": 0.26456409096717837,
      "num_tokens": 18522079.0,
      "step": 8250
    },
    {
      "entropy": 5.257075452804566,
      "epoch": 8.924824229313142,
      "grad_norm": 1.4921875,
      "learning_rate": 6.571198837260148e-05,
      "loss": 4.4272,
      "mean_token_accuracy": 0.26700678169727327,
      "num_tokens": 18533749.0,
      "step": 8255
    },
    {
      "entropy": 5.264775896072388,
      "epoch": 8.930232558139535,
      "grad_norm": 1.390625,
      "learning_rate": 6.555489898643201e-05,
      "loss": 4.4268,
      "mean_token_accuracy": 0.2764290124177933,
      "num_tokens": 18545994.0,
      "step": 8260
    },
    {
      "entropy": 5.21644549369812,
      "epoch": 8.935640886965928,
      "grad_norm": 1.3125,
      "learning_rate": 6.539857072424357e-05,
      "loss": 4.3984,
      "mean_token_accuracy": 0.267026923596859,
      "num_tokens": 18556260.0,
      "step": 8265
    },
    {
      "entropy": 5.317139339447022,
      "epoch": 8.94104921579232,
      "grad_norm": 1.2265625,
      "learning_rate": 6.524300415413354e-05,
      "loss": 4.5004,
      "mean_token_accuracy": 0.2684118703007698,
      "num_tokens": 18567341.0,
      "step": 8270
    },
    {
      "entropy": 5.279985952377319,
      "epoch": 8.946457544618713,
      "grad_norm": 1.28125,
      "learning_rate": 6.508819984143111e-05,
      "loss": 4.5058,
      "mean_token_accuracy": 0.2568607360124588,
      "num_tokens": 18579203.0,
      "step": 8275
    },
    {
      "entropy": 5.257149124145508,
      "epoch": 8.951865873445106,
      "grad_norm": 1.265625,
      "learning_rate": 6.493415834869555e-05,
      "loss": 4.5299,
      "mean_token_accuracy": 0.2594893991947174,
      "num_tokens": 18592194.0,
      "step": 8280
    },
    {
      "entropy": 5.259315109252929,
      "epoch": 8.957274202271497,
      "grad_norm": 1.4140625,
      "learning_rate": 6.478088023571403e-05,
      "loss": 4.4463,
      "mean_token_accuracy": 0.2619550883769989,
      "num_tokens": 18603988.0,
      "step": 8285
    },
    {
      "entropy": 5.315274286270141,
      "epoch": 8.96268253109789,
      "grad_norm": 1.3828125,
      "learning_rate": 6.462836605949945e-05,
      "loss": 4.4424,
      "mean_token_accuracy": 0.26160885840654374,
      "num_tokens": 18614324.0,
      "step": 8290
    },
    {
      "entropy": 5.27551498413086,
      "epoch": 8.968090859924283,
      "grad_norm": 1.2578125,
      "learning_rate": 6.447661637428882e-05,
      "loss": 4.5017,
      "mean_token_accuracy": 0.26732158809900286,
      "num_tokens": 18626499.0,
      "step": 8295
    },
    {
      "entropy": 5.268152236938477,
      "epoch": 8.973499188750676,
      "grad_norm": 1.2578125,
      "learning_rate": 6.432563173154082e-05,
      "loss": 4.4522,
      "mean_token_accuracy": 0.267254339158535,
      "num_tokens": 18637755.0,
      "step": 8300
    },
    {
      "entropy": 5.305533695220947,
      "epoch": 8.978907517577069,
      "grad_norm": 1.40625,
      "learning_rate": 6.417541267993391e-05,
      "loss": 4.4878,
      "mean_token_accuracy": 0.25812672525644303,
      "num_tokens": 18647971.0,
      "step": 8305
    },
    {
      "entropy": 5.298321676254273,
      "epoch": 8.984315846403462,
      "grad_norm": 1.2734375,
      "learning_rate": 6.402595976536451e-05,
      "loss": 4.4639,
      "mean_token_accuracy": 0.27111134976148604,
      "num_tokens": 18660396.0,
      "step": 8310
    },
    {
      "entropy": 5.282906532287598,
      "epoch": 8.989724175229854,
      "grad_norm": 1.3203125,
      "learning_rate": 6.387727353094492e-05,
      "loss": 4.4586,
      "mean_token_accuracy": 0.2672571614384651,
      "num_tokens": 18671017.0,
      "step": 8315
    },
    {
      "entropy": 5.292878580093384,
      "epoch": 8.995132504056247,
      "grad_norm": 1.265625,
      "learning_rate": 6.372935451700119e-05,
      "loss": 4.4686,
      "mean_token_accuracy": 0.2600624248385429,
      "num_tokens": 18683229.0,
      "step": 8320
    },
    {
      "entropy": 5.228705671098497,
      "epoch": 9.0,
      "grad_norm": 2.828125,
      "learning_rate": 6.358220326107135e-05,
      "loss": 4.5196,
      "mean_token_accuracy": 0.25854048132896423,
      "num_tokens": 18692001.0,
      "step": 8325
    },
    {
      "entropy": 5.233514928817749,
      "epoch": 9.005408328826393,
      "grad_norm": 1.2890625,
      "learning_rate": 6.343582029790345e-05,
      "loss": 4.363,
      "mean_token_accuracy": 0.2730303943157196,
      "num_tokens": 18703192.0,
      "step": 8330
    },
    {
      "entropy": 5.252371072769165,
      "epoch": 9.010816657652786,
      "grad_norm": 1.28125,
      "learning_rate": 6.32902061594535e-05,
      "loss": 4.4333,
      "mean_token_accuracy": 0.27003188580274584,
      "num_tokens": 18713962.0,
      "step": 8335
    },
    {
      "entropy": 5.317679691314697,
      "epoch": 9.016224986479179,
      "grad_norm": 1.15625,
      "learning_rate": 6.314536137488364e-05,
      "loss": 4.4629,
      "mean_token_accuracy": 0.26359549164772034,
      "num_tokens": 18726267.0,
      "step": 8340
    },
    {
      "entropy": 5.2380163192749025,
      "epoch": 9.021633315305571,
      "grad_norm": 1.3203125,
      "learning_rate": 6.300128647056007e-05,
      "loss": 4.3936,
      "mean_token_accuracy": 0.2596570193767548,
      "num_tokens": 18737178.0,
      "step": 8345
    },
    {
      "entropy": 5.353796768188476,
      "epoch": 9.027041644131963,
      "grad_norm": 1.3515625,
      "learning_rate": 6.285798197005136e-05,
      "loss": 4.4832,
      "mean_token_accuracy": 0.26296883523464204,
      "num_tokens": 18747773.0,
      "step": 8350
    },
    {
      "entropy": 5.256224632263184,
      "epoch": 9.032449972958355,
      "grad_norm": 1.453125,
      "learning_rate": 6.271544839412647e-05,
      "loss": 4.372,
      "mean_token_accuracy": 0.2663264662027359,
      "num_tokens": 18758641.0,
      "step": 8355
    },
    {
      "entropy": 5.217123937606812,
      "epoch": 9.037858301784748,
      "grad_norm": 1.265625,
      "learning_rate": 6.257368626075266e-05,
      "loss": 4.3848,
      "mean_token_accuracy": 0.26678973585367205,
      "num_tokens": 18769262.0,
      "step": 8360
    },
    {
      "entropy": 5.276500225067139,
      "epoch": 9.043266630611141,
      "grad_norm": 1.5703125,
      "learning_rate": 6.243269608509388e-05,
      "loss": 4.4508,
      "mean_token_accuracy": 0.2740554228425026,
      "num_tokens": 18780660.0,
      "step": 8365
    },
    {
      "entropy": 5.197865676879883,
      "epoch": 9.048674959437534,
      "grad_norm": 1.25,
      "learning_rate": 6.229247837950878e-05,
      "loss": 4.345,
      "mean_token_accuracy": 0.2760092779994011,
      "num_tokens": 18791998.0,
      "step": 8370
    },
    {
      "entropy": 5.261660385131836,
      "epoch": 9.054083288263927,
      "grad_norm": 1.296875,
      "learning_rate": 6.215303365354875e-05,
      "loss": 4.3985,
      "mean_token_accuracy": 0.2690773010253906,
      "num_tokens": 18802856.0,
      "step": 8375
    },
    {
      "entropy": 5.23825101852417,
      "epoch": 9.05949161709032,
      "grad_norm": 1.4375,
      "learning_rate": 6.201436241395634e-05,
      "loss": 4.401,
      "mean_token_accuracy": 0.2655333742499352,
      "num_tokens": 18813110.0,
      "step": 8380
    },
    {
      "entropy": 5.294359636306763,
      "epoch": 9.064899945916713,
      "grad_norm": 1.2890625,
      "learning_rate": 6.187646516466311e-05,
      "loss": 4.5177,
      "mean_token_accuracy": 0.2550377696752548,
      "num_tokens": 18824982.0,
      "step": 8385
    },
    {
      "entropy": 5.233408498764038,
      "epoch": 9.070308274743104,
      "grad_norm": 1.3125,
      "learning_rate": 6.173934240678794e-05,
      "loss": 4.4012,
      "mean_token_accuracy": 0.2690296649932861,
      "num_tokens": 18836272.0,
      "step": 8390
    },
    {
      "entropy": 5.2167035102844235,
      "epoch": 9.075716603569496,
      "grad_norm": 1.328125,
      "learning_rate": 6.160299463863535e-05,
      "loss": 4.3799,
      "mean_token_accuracy": 0.2737075611948967,
      "num_tokens": 18846904.0,
      "step": 8395
    },
    {
      "entropy": 5.254539871215821,
      "epoch": 9.08112493239589,
      "grad_norm": 1.375,
      "learning_rate": 6.146742235569336e-05,
      "loss": 4.4334,
      "mean_token_accuracy": 0.2698467135429382,
      "num_tokens": 18857384.0,
      "step": 8400
    },
    {
      "entropy": 5.267408609390259,
      "epoch": 9.086533261222282,
      "grad_norm": 1.234375,
      "learning_rate": 6.133262605063199e-05,
      "loss": 4.4461,
      "mean_token_accuracy": 0.26529874205589293,
      "num_tokens": 18869358.0,
      "step": 8405
    },
    {
      "entropy": 5.276978302001953,
      "epoch": 9.091941590048675,
      "grad_norm": 1.296875,
      "learning_rate": 6.119860621330131e-05,
      "loss": 4.4573,
      "mean_token_accuracy": 0.2647540494799614,
      "num_tokens": 18879863.0,
      "step": 8410
    },
    {
      "entropy": 5.328768062591553,
      "epoch": 9.097349918875068,
      "grad_norm": 1.3671875,
      "learning_rate": 6.106536333072973e-05,
      "loss": 4.4526,
      "mean_token_accuracy": 0.2775487914681435,
      "num_tokens": 18891075.0,
      "step": 8415
    },
    {
      "entropy": 5.233926773071289,
      "epoch": 9.10275824770146,
      "grad_norm": 1.3359375,
      "learning_rate": 6.093289788712218e-05,
      "loss": 4.4598,
      "mean_token_accuracy": 0.26452367305755614,
      "num_tokens": 18902372.0,
      "step": 8420
    },
    {
      "entropy": 5.228367280960083,
      "epoch": 9.108166576527854,
      "grad_norm": 1.34375,
      "learning_rate": 6.080121036385845e-05,
      "loss": 4.4455,
      "mean_token_accuracy": 0.2711497455835342,
      "num_tokens": 18912283.0,
      "step": 8425
    },
    {
      "entropy": 5.283715391159058,
      "epoch": 9.113574905354245,
      "grad_norm": 1.34375,
      "learning_rate": 6.067030123949119e-05,
      "loss": 4.5088,
      "mean_token_accuracy": 0.25499771237373353,
      "num_tokens": 18923743.0,
      "step": 8430
    },
    {
      "entropy": 5.310470008850098,
      "epoch": 9.118983234180638,
      "grad_norm": 1.3046875,
      "learning_rate": 6.0540170989744566e-05,
      "loss": 4.3741,
      "mean_token_accuracy": 0.2815611720085144,
      "num_tokens": 18935348.0,
      "step": 8435
    },
    {
      "entropy": 5.197281455993652,
      "epoch": 9.12439156300703,
      "grad_norm": 1.484375,
      "learning_rate": 6.041082008751215e-05,
      "loss": 4.3542,
      "mean_token_accuracy": 0.2816265895962715,
      "num_tokens": 18945168.0,
      "step": 8440
    },
    {
      "entropy": 5.230170392990113,
      "epoch": 9.129799891833423,
      "grad_norm": 1.25,
      "learning_rate": 6.028224900285545e-05,
      "loss": 4.317,
      "mean_token_accuracy": 0.2724506095051765,
      "num_tokens": 18956814.0,
      "step": 8445
    },
    {
      "entropy": 5.3020414352417,
      "epoch": 9.135208220659816,
      "grad_norm": 1.3828125,
      "learning_rate": 6.015445820300207e-05,
      "loss": 4.4052,
      "mean_token_accuracy": 0.2717366203665733,
      "num_tokens": 18967414.0,
      "step": 8450
    },
    {
      "entropy": 5.208000612258911,
      "epoch": 9.140616549486209,
      "grad_norm": 1.3046875,
      "learning_rate": 6.002744815234405e-05,
      "loss": 4.3439,
      "mean_token_accuracy": 0.2807419687509537,
      "num_tokens": 18978624.0,
      "step": 8455
    },
    {
      "entropy": 5.257434511184693,
      "epoch": 9.146024878312602,
      "grad_norm": 1.2578125,
      "learning_rate": 5.990121931243623e-05,
      "loss": 4.3971,
      "mean_token_accuracy": 0.2747312068939209,
      "num_tokens": 18989669.0,
      "step": 8460
    },
    {
      "entropy": 5.2072014808654785,
      "epoch": 9.151433207138995,
      "grad_norm": 1.3359375,
      "learning_rate": 5.977577214199456e-05,
      "loss": 4.405,
      "mean_token_accuracy": 0.26832211911678316,
      "num_tokens": 19000076.0,
      "step": 8465
    },
    {
      "entropy": 5.207018041610718,
      "epoch": 9.156841535965386,
      "grad_norm": 1.2265625,
      "learning_rate": 5.9651107096894214e-05,
      "loss": 4.3624,
      "mean_token_accuracy": 0.27644550502300264,
      "num_tokens": 19012273.0,
      "step": 8470
    },
    {
      "entropy": 5.259827756881714,
      "epoch": 9.162249864791779,
      "grad_norm": 1.484375,
      "learning_rate": 5.952722463016836e-05,
      "loss": 4.4266,
      "mean_token_accuracy": 0.2678874537348747,
      "num_tokens": 19023379.0,
      "step": 8475
    },
    {
      "entropy": 5.249720811843872,
      "epoch": 9.167658193618172,
      "grad_norm": 1.2578125,
      "learning_rate": 5.940412519200617e-05,
      "loss": 4.39,
      "mean_token_accuracy": 0.27513250410556794,
      "num_tokens": 19034174.0,
      "step": 8480
    },
    {
      "entropy": 5.275164031982422,
      "epoch": 9.173066522444564,
      "grad_norm": 1.203125,
      "learning_rate": 5.928180922975124e-05,
      "loss": 4.4494,
      "mean_token_accuracy": 0.2576396957039833,
      "num_tokens": 19045891.0,
      "step": 8485
    },
    {
      "entropy": 5.253449487686157,
      "epoch": 9.178474851270957,
      "grad_norm": 1.2734375,
      "learning_rate": 5.916027718790007e-05,
      "loss": 4.3976,
      "mean_token_accuracy": 0.27672222852706907,
      "num_tokens": 19057120.0,
      "step": 8490
    },
    {
      "entropy": 5.330633449554443,
      "epoch": 9.18388318009735,
      "grad_norm": 1.4921875,
      "learning_rate": 5.903952950810038e-05,
      "loss": 4.4531,
      "mean_token_accuracy": 0.2764663890004158,
      "num_tokens": 19068631.0,
      "step": 8495
    },
    {
      "entropy": 5.210518503189087,
      "epoch": 9.189291508923743,
      "grad_norm": 1.2734375,
      "learning_rate": 5.8919566629149535e-05,
      "loss": 4.3416,
      "mean_token_accuracy": 0.2763280510902405,
      "num_tokens": 19079304.0,
      "step": 8500
    },
    {
      "entropy": 5.241615247726441,
      "epoch": 9.194699837750136,
      "grad_norm": 1.3515625,
      "learning_rate": 5.880038898699291e-05,
      "loss": 4.4252,
      "mean_token_accuracy": 0.2763049840927124,
      "num_tokens": 19090838.0,
      "step": 8505
    },
    {
      "entropy": 5.2076560974121096,
      "epoch": 9.200108166576527,
      "grad_norm": 1.2421875,
      "learning_rate": 5.8681997014722303e-05,
      "loss": 4.3558,
      "mean_token_accuracy": 0.27298678159713746,
      "num_tokens": 19102632.0,
      "step": 8510
    },
    {
      "entropy": 5.25959620475769,
      "epoch": 9.20551649540292,
      "grad_norm": 1.34375,
      "learning_rate": 5.856439114257446e-05,
      "loss": 4.4213,
      "mean_token_accuracy": 0.2639960676431656,
      "num_tokens": 19113173.0,
      "step": 8515
    },
    {
      "entropy": 5.216385555267334,
      "epoch": 9.210924824229313,
      "grad_norm": 1.3203125,
      "learning_rate": 5.84475717979294e-05,
      "loss": 4.3719,
      "mean_token_accuracy": 0.26903262734413147,
      "num_tokens": 19124501.0,
      "step": 8520
    },
    {
      "entropy": 5.2307061672210695,
      "epoch": 9.216333153055706,
      "grad_norm": 1.3046875,
      "learning_rate": 5.833153940530882e-05,
      "loss": 4.3934,
      "mean_token_accuracy": 0.2668346047401428,
      "num_tokens": 19136317.0,
      "step": 8525
    },
    {
      "entropy": 5.242311096191406,
      "epoch": 9.221741481882098,
      "grad_norm": 1.2734375,
      "learning_rate": 5.8216294386374727e-05,
      "loss": 4.3607,
      "mean_token_accuracy": 0.27337653934955597,
      "num_tokens": 19146895.0,
      "step": 8530
    },
    {
      "entropy": 5.284078931808471,
      "epoch": 9.227149810708491,
      "grad_norm": 1.21875,
      "learning_rate": 5.8101837159927845e-05,
      "loss": 4.4422,
      "mean_token_accuracy": 0.2740187615156174,
      "num_tokens": 19157974.0,
      "step": 8535
    },
    {
      "entropy": 5.226920795440674,
      "epoch": 9.232558139534884,
      "grad_norm": 1.28125,
      "learning_rate": 5.798816814190595e-05,
      "loss": 4.3971,
      "mean_token_accuracy": 0.27572468519210813,
      "num_tokens": 19169515.0,
      "step": 8540
    },
    {
      "entropy": 5.309182214736938,
      "epoch": 9.237966468361277,
      "grad_norm": 1.2890625,
      "learning_rate": 5.7875287745382504e-05,
      "loss": 4.4321,
      "mean_token_accuracy": 0.2715494215488434,
      "num_tokens": 19180511.0,
      "step": 8545
    },
    {
      "entropy": 5.220964670181274,
      "epoch": 9.24337479718767,
      "grad_norm": 1.2734375,
      "learning_rate": 5.776319638056516e-05,
      "loss": 4.4121,
      "mean_token_accuracy": 0.27345471978187563,
      "num_tokens": 19191913.0,
      "step": 8550
    },
    {
      "entropy": 5.269768142700196,
      "epoch": 9.248783126014061,
      "grad_norm": 1.3203125,
      "learning_rate": 5.76518944547942e-05,
      "loss": 4.4837,
      "mean_token_accuracy": 0.26473542898893354,
      "num_tokens": 19204306.0,
      "step": 8555
    },
    {
      "entropy": 5.349148559570312,
      "epoch": 9.254191454840454,
      "grad_norm": 1.2890625,
      "learning_rate": 5.754138237254108e-05,
      "loss": 4.5056,
      "mean_token_accuracy": 0.26274736523628234,
      "num_tokens": 19215466.0,
      "step": 8560
    },
    {
      "entropy": 5.182336235046387,
      "epoch": 9.259599783666847,
      "grad_norm": 1.3828125,
      "learning_rate": 5.743166053540692e-05,
      "loss": 4.3379,
      "mean_token_accuracy": 0.2780485197901726,
      "num_tokens": 19225787.0,
      "step": 8565
    },
    {
      "entropy": 5.304765129089356,
      "epoch": 9.26500811249324,
      "grad_norm": 1.34375,
      "learning_rate": 5.732272934212113e-05,
      "loss": 4.4459,
      "mean_token_accuracy": 0.2689804807305336,
      "num_tokens": 19237078.0,
      "step": 8570
    },
    {
      "entropy": 5.280612134933472,
      "epoch": 9.270416441319632,
      "grad_norm": 1.21875,
      "learning_rate": 5.721458918854001e-05,
      "loss": 4.4528,
      "mean_token_accuracy": 0.2652651697397232,
      "num_tokens": 19249071.0,
      "step": 8575
    },
    {
      "entropy": 5.277837610244751,
      "epoch": 9.275824770146025,
      "grad_norm": 1.3671875,
      "learning_rate": 5.7107240467645024e-05,
      "loss": 4.4354,
      "mean_token_accuracy": 0.2710088148713112,
      "num_tokens": 19259210.0,
      "step": 8580
    },
    {
      "entropy": 5.259544563293457,
      "epoch": 9.281233098972418,
      "grad_norm": 1.296875,
      "learning_rate": 5.700068356954172e-05,
      "loss": 4.4716,
      "mean_token_accuracy": 0.26590579450130464,
      "num_tokens": 19270803.0,
      "step": 8585
    },
    {
      "entropy": 5.23133864402771,
      "epoch": 9.286641427798811,
      "grad_norm": 1.3828125,
      "learning_rate": 5.689491888145811e-05,
      "loss": 4.3614,
      "mean_token_accuracy": 0.27088024020195006,
      "num_tokens": 19280883.0,
      "step": 8590
    },
    {
      "entropy": 5.297468662261963,
      "epoch": 9.292049756625202,
      "grad_norm": 1.2578125,
      "learning_rate": 5.67899467877433e-05,
      "loss": 4.4108,
      "mean_token_accuracy": 0.26513769626617434,
      "num_tokens": 19292399.0,
      "step": 8595
    },
    {
      "entropy": 5.252052307128906,
      "epoch": 9.297458085451595,
      "grad_norm": 1.3203125,
      "learning_rate": 5.668576766986611e-05,
      "loss": 4.4239,
      "mean_token_accuracy": 0.2683245688676834,
      "num_tokens": 19302386.0,
      "step": 8600
    },
    {
      "entropy": 5.302150821685791,
      "epoch": 9.302866414277988,
      "grad_norm": 1.2734375,
      "learning_rate": 5.658238190641378e-05,
      "loss": 4.4928,
      "mean_token_accuracy": 0.2607463806867599,
      "num_tokens": 19314774.0,
      "step": 8605
    },
    {
      "entropy": 5.274616718292236,
      "epoch": 9.30827474310438,
      "grad_norm": 1.2421875,
      "learning_rate": 5.6479789873090273e-05,
      "loss": 4.4579,
      "mean_token_accuracy": 0.2596276491880417,
      "num_tokens": 19326013.0,
      "step": 8610
    },
    {
      "entropy": 5.290752792358399,
      "epoch": 9.313683071930773,
      "grad_norm": 1.3203125,
      "learning_rate": 5.637799194271541e-05,
      "loss": 4.4067,
      "mean_token_accuracy": 0.27421183586120607,
      "num_tokens": 19337107.0,
      "step": 8615
    },
    {
      "entropy": 5.238031959533691,
      "epoch": 9.319091400757166,
      "grad_norm": 1.40625,
      "learning_rate": 5.6276988485223004e-05,
      "loss": 4.359,
      "mean_token_accuracy": 0.2718453943729401,
      "num_tokens": 19347667.0,
      "step": 8620
    },
    {
      "entropy": 5.222658920288086,
      "epoch": 9.32449972958356,
      "grad_norm": 1.2890625,
      "learning_rate": 5.617677986765987e-05,
      "loss": 4.3745,
      "mean_token_accuracy": 0.26994083523750306,
      "num_tokens": 19358071.0,
      "step": 8625
    },
    {
      "entropy": 5.222465515136719,
      "epoch": 9.329908058409952,
      "grad_norm": 1.2890625,
      "learning_rate": 5.6077366454184446e-05,
      "loss": 4.4164,
      "mean_token_accuracy": 0.26124546974897384,
      "num_tokens": 19370835.0,
      "step": 8630
    },
    {
      "entropy": 5.262303352355957,
      "epoch": 9.335316387236343,
      "grad_norm": 1.2734375,
      "learning_rate": 5.597874860606521e-05,
      "loss": 4.4184,
      "mean_token_accuracy": 0.2650756731629372,
      "num_tokens": 19382286.0,
      "step": 8635
    },
    {
      "entropy": 5.331569576263428,
      "epoch": 9.340724716062736,
      "grad_norm": 1.171875,
      "learning_rate": 5.5880926681679695e-05,
      "loss": 4.5139,
      "mean_token_accuracy": 0.2633565112948418,
      "num_tokens": 19393721.0,
      "step": 8640
    },
    {
      "entropy": 5.238211584091187,
      "epoch": 9.346133044889129,
      "grad_norm": 1.359375,
      "learning_rate": 5.578390103651297e-05,
      "loss": 4.3741,
      "mean_token_accuracy": 0.2740733250975609,
      "num_tokens": 19406199.0,
      "step": 8645
    },
    {
      "entropy": 5.229563808441162,
      "epoch": 9.351541373715522,
      "grad_norm": 1.40625,
      "learning_rate": 5.568767202315647e-05,
      "loss": 4.4302,
      "mean_token_accuracy": 0.2684115290641785,
      "num_tokens": 19416700.0,
      "step": 8650
    },
    {
      "entropy": 5.2509904384613035,
      "epoch": 9.356949702541915,
      "grad_norm": 1.3125,
      "learning_rate": 5.559223999130667e-05,
      "loss": 4.4992,
      "mean_token_accuracy": 0.2624796897172928,
      "num_tokens": 19427739.0,
      "step": 8655
    },
    {
      "entropy": 5.189073085784912,
      "epoch": 9.362358031368307,
      "grad_norm": 1.3359375,
      "learning_rate": 5.549760528776376e-05,
      "loss": 4.3313,
      "mean_token_accuracy": 0.278664630651474,
      "num_tokens": 19437575.0,
      "step": 8660
    },
    {
      "entropy": 5.288852405548096,
      "epoch": 9.3677663601947,
      "grad_norm": 1.3515625,
      "learning_rate": 5.5403768256430475e-05,
      "loss": 4.4793,
      "mean_token_accuracy": 0.26517258286476136,
      "num_tokens": 19449342.0,
      "step": 8665
    },
    {
      "entropy": 5.265920686721802,
      "epoch": 9.373174689021093,
      "grad_norm": 1.421875,
      "learning_rate": 5.5310729238310854e-05,
      "loss": 4.4104,
      "mean_token_accuracy": 0.2692974582314491,
      "num_tokens": 19460002.0,
      "step": 8670
    },
    {
      "entropy": 5.260612106323242,
      "epoch": 9.378583017847484,
      "grad_norm": 1.3515625,
      "learning_rate": 5.521848857150884e-05,
      "loss": 4.3969,
      "mean_token_accuracy": 0.26888851821422577,
      "num_tokens": 19471271.0,
      "step": 8675
    },
    {
      "entropy": 5.239969635009766,
      "epoch": 9.383991346673877,
      "grad_norm": 1.359375,
      "learning_rate": 5.512704659122725e-05,
      "loss": 4.4224,
      "mean_token_accuracy": 0.2731319785118103,
      "num_tokens": 19482959.0,
      "step": 8680
    },
    {
      "entropy": 5.22791018486023,
      "epoch": 9.38939967550027,
      "grad_norm": 1.3984375,
      "learning_rate": 5.503640362976646e-05,
      "loss": 4.3668,
      "mean_token_accuracy": 0.273055724799633,
      "num_tokens": 19493640.0,
      "step": 8685
    },
    {
      "entropy": 5.32327299118042,
      "epoch": 9.394808004326663,
      "grad_norm": 1.390625,
      "learning_rate": 5.4946560016523184e-05,
      "loss": 4.5007,
      "mean_token_accuracy": 0.2687953218817711,
      "num_tokens": 19505501.0,
      "step": 8690
    },
    {
      "entropy": 5.2029634475708,
      "epoch": 9.400216333153056,
      "grad_norm": 1.2265625,
      "learning_rate": 5.485751607798937e-05,
      "loss": 4.3848,
      "mean_token_accuracy": 0.2718505963683128,
      "num_tokens": 19517731.0,
      "step": 8695
    },
    {
      "entropy": 5.286900424957276,
      "epoch": 9.405624661979449,
      "grad_norm": 1.390625,
      "learning_rate": 5.4769272137750855e-05,
      "loss": 4.5139,
      "mean_token_accuracy": 0.2627735689282417,
      "num_tokens": 19529319.0,
      "step": 8700
    },
    {
      "entropy": 5.3168271541595455,
      "epoch": 9.411032990805841,
      "grad_norm": 1.375,
      "learning_rate": 5.468182851648629e-05,
      "loss": 4.4843,
      "mean_token_accuracy": 0.2634874388575554,
      "num_tokens": 19541124.0,
      "step": 8705
    },
    {
      "entropy": 5.208156585693359,
      "epoch": 9.416441319632234,
      "grad_norm": 1.3046875,
      "learning_rate": 5.4595185531966074e-05,
      "loss": 4.4067,
      "mean_token_accuracy": 0.2772697746753693,
      "num_tokens": 19552664.0,
      "step": 8710
    },
    {
      "entropy": 5.2800140380859375,
      "epoch": 9.421849648458625,
      "grad_norm": 1.28125,
      "learning_rate": 5.450934349905094e-05,
      "loss": 4.4423,
      "mean_token_accuracy": 0.27446315735578536,
      "num_tokens": 19564903.0,
      "step": 8715
    },
    {
      "entropy": 5.212539291381836,
      "epoch": 9.427257977285018,
      "grad_norm": 1.3359375,
      "learning_rate": 5.44243027296911e-05,
      "loss": 4.3744,
      "mean_token_accuracy": 0.26784237921237947,
      "num_tokens": 19575148.0,
      "step": 8720
    },
    {
      "entropy": 5.199701404571533,
      "epoch": 9.432666306111411,
      "grad_norm": 1.390625,
      "learning_rate": 5.434006353292485e-05,
      "loss": 4.3539,
      "mean_token_accuracy": 0.27157516181468966,
      "num_tokens": 19585676.0,
      "step": 8725
    },
    {
      "entropy": 5.309289836883545,
      "epoch": 9.438074634937804,
      "grad_norm": 1.359375,
      "learning_rate": 5.425662621487766e-05,
      "loss": 4.4564,
      "mean_token_accuracy": 0.27129481732845306,
      "num_tokens": 19597089.0,
      "step": 8730
    },
    {
      "entropy": 5.2756492614746096,
      "epoch": 9.443482963764197,
      "grad_norm": 1.28125,
      "learning_rate": 5.417399107876092e-05,
      "loss": 4.4592,
      "mean_token_accuracy": 0.2651286855340004,
      "num_tokens": 19607990.0,
      "step": 8735
    },
    {
      "entropy": 5.318877506256103,
      "epoch": 9.44889129259059,
      "grad_norm": 1.453125,
      "learning_rate": 5.4092158424870936e-05,
      "loss": 4.4607,
      "mean_token_accuracy": 0.25489178150892255,
      "num_tokens": 19618469.0,
      "step": 8740
    },
    {
      "entropy": 5.211070966720581,
      "epoch": 9.454299621416983,
      "grad_norm": 1.4296875,
      "learning_rate": 5.40111285505877e-05,
      "loss": 4.353,
      "mean_token_accuracy": 0.2766400188207626,
      "num_tokens": 19630674.0,
      "step": 8745
    },
    {
      "entropy": 5.296444940567016,
      "epoch": 9.459707950243375,
      "grad_norm": 1.328125,
      "learning_rate": 5.393090175037399e-05,
      "loss": 4.4544,
      "mean_token_accuracy": 0.2788611724972725,
      "num_tokens": 19642132.0,
      "step": 8750
    },
    {
      "entropy": 5.317198657989502,
      "epoch": 9.465116279069768,
      "grad_norm": 1.28125,
      "learning_rate": 5.385147831577424e-05,
      "loss": 4.4376,
      "mean_token_accuracy": 0.27303552329540254,
      "num_tokens": 19653553.0,
      "step": 8755
    },
    {
      "entropy": 5.2411082744598385,
      "epoch": 9.47052460789616,
      "grad_norm": 1.4765625,
      "learning_rate": 5.377285853541332e-05,
      "loss": 4.4481,
      "mean_token_accuracy": 0.26680942475795744,
      "num_tokens": 19665419.0,
      "step": 8760
    },
    {
      "entropy": 5.279013156890869,
      "epoch": 9.475932936722552,
      "grad_norm": 1.3984375,
      "learning_rate": 5.369504269499574e-05,
      "loss": 4.3956,
      "mean_token_accuracy": 0.2730926305055618,
      "num_tokens": 19676047.0,
      "step": 8765
    },
    {
      "entropy": 5.247904062271118,
      "epoch": 9.481341265548945,
      "grad_norm": 1.3984375,
      "learning_rate": 5.3618031077304406e-05,
      "loss": 4.4199,
      "mean_token_accuracy": 0.2647079661488533,
      "num_tokens": 19687399.0,
      "step": 8770
    },
    {
      "entropy": 5.331304121017456,
      "epoch": 9.486749594375338,
      "grad_norm": 1.203125,
      "learning_rate": 5.354182396219974e-05,
      "loss": 4.5036,
      "mean_token_accuracy": 0.26295833140611646,
      "num_tokens": 19698851.0,
      "step": 8775
    },
    {
      "entropy": 5.283445262908936,
      "epoch": 9.49215792320173,
      "grad_norm": 1.2578125,
      "learning_rate": 5.3466421626618573e-05,
      "loss": 4.4192,
      "mean_token_accuracy": 0.2698570266366005,
      "num_tokens": 19710523.0,
      "step": 8780
    },
    {
      "entropy": 5.271654224395752,
      "epoch": 9.497566252028124,
      "grad_norm": 1.390625,
      "learning_rate": 5.339182434457314e-05,
      "loss": 4.3989,
      "mean_token_accuracy": 0.26870272159576414,
      "num_tokens": 19723366.0,
      "step": 8785
    },
    {
      "entropy": 5.382745695114136,
      "epoch": 9.502974580854517,
      "grad_norm": 1.2421875,
      "learning_rate": 5.3318032387150165e-05,
      "loss": 4.5285,
      "mean_token_accuracy": 0.26733721494674684,
      "num_tokens": 19734320.0,
      "step": 8790
    },
    {
      "entropy": 5.283776378631591,
      "epoch": 9.50838290968091,
      "grad_norm": 1.3359375,
      "learning_rate": 5.324504602250981e-05,
      "loss": 4.4258,
      "mean_token_accuracy": 0.2688348487019539,
      "num_tokens": 19745366.0,
      "step": 8795
    },
    {
      "entropy": 5.262878179550171,
      "epoch": 9.5137912385073,
      "grad_norm": 1.46875,
      "learning_rate": 5.317286551588464e-05,
      "loss": 4.4524,
      "mean_token_accuracy": 0.27158365547657015,
      "num_tokens": 19755709.0,
      "step": 8800
    },
    {
      "entropy": 5.260160207748413,
      "epoch": 9.519199567333693,
      "grad_norm": 1.3515625,
      "learning_rate": 5.310149112957882e-05,
      "loss": 4.4604,
      "mean_token_accuracy": 0.26207941621541975,
      "num_tokens": 19765998.0,
      "step": 8805
    },
    {
      "entropy": 5.247035360336303,
      "epoch": 9.524607896160086,
      "grad_norm": 1.3828125,
      "learning_rate": 5.3030923122967036e-05,
      "loss": 4.4222,
      "mean_token_accuracy": 0.273430198431015,
      "num_tokens": 19777172.0,
      "step": 8810
    },
    {
      "entropy": 5.218124866485596,
      "epoch": 9.530016224986479,
      "grad_norm": 1.421875,
      "learning_rate": 5.296116175249357e-05,
      "loss": 4.4238,
      "mean_token_accuracy": 0.27391154766082765,
      "num_tokens": 19788807.0,
      "step": 8815
    },
    {
      "entropy": 5.185249519348145,
      "epoch": 9.535424553812872,
      "grad_norm": 1.328125,
      "learning_rate": 5.289220727167139e-05,
      "loss": 4.3938,
      "mean_token_accuracy": 0.26936565041542054,
      "num_tokens": 19800570.0,
      "step": 8820
    },
    {
      "entropy": 5.2898516178131105,
      "epoch": 9.540832882639265,
      "grad_norm": 1.4609375,
      "learning_rate": 5.282405993108127e-05,
      "loss": 4.4777,
      "mean_token_accuracy": 0.2633599013090134,
      "num_tokens": 19812961.0,
      "step": 8825
    },
    {
      "entropy": 5.234461641311645,
      "epoch": 9.546241211465658,
      "grad_norm": 1.3125,
      "learning_rate": 5.2756719978370794e-05,
      "loss": 4.4279,
      "mean_token_accuracy": 0.27458291947841645,
      "num_tokens": 19824123.0,
      "step": 8830
    },
    {
      "entropy": 5.216208219528198,
      "epoch": 9.55164954029205,
      "grad_norm": 1.4140625,
      "learning_rate": 5.2690187658253525e-05,
      "loss": 4.3593,
      "mean_token_accuracy": 0.27112574577331544,
      "num_tokens": 19834745.0,
      "step": 8835
    },
    {
      "entropy": 5.241052532196045,
      "epoch": 9.557057869118442,
      "grad_norm": 1.25,
      "learning_rate": 5.262446321250803e-05,
      "loss": 4.3953,
      "mean_token_accuracy": 0.2744523838162422,
      "num_tokens": 19845943.0,
      "step": 8840
    },
    {
      "entropy": 5.231638813018799,
      "epoch": 9.562466197944834,
      "grad_norm": 1.5703125,
      "learning_rate": 5.2559546879977084e-05,
      "loss": 4.4124,
      "mean_token_accuracy": 0.2687002465128899,
      "num_tokens": 19856914.0,
      "step": 8845
    },
    {
      "entropy": 5.2921693325042725,
      "epoch": 9.567874526771227,
      "grad_norm": 1.234375,
      "learning_rate": 5.2495438896566854e-05,
      "loss": 4.4708,
      "mean_token_accuracy": 0.2651283875107765,
      "num_tokens": 19867631.0,
      "step": 8850
    },
    {
      "entropy": 5.289273309707641,
      "epoch": 9.57328285559762,
      "grad_norm": 1.3203125,
      "learning_rate": 5.2432139495245855e-05,
      "loss": 4.463,
      "mean_token_accuracy": 0.2584497734904289,
      "num_tokens": 19877937.0,
      "step": 8855
    },
    {
      "entropy": 5.290476369857788,
      "epoch": 9.578691184424013,
      "grad_norm": 1.3515625,
      "learning_rate": 5.236964890604426e-05,
      "loss": 4.4517,
      "mean_token_accuracy": 0.26506915837526324,
      "num_tokens": 19889158.0,
      "step": 8860
    },
    {
      "entropy": 5.231437635421753,
      "epoch": 9.584099513250406,
      "grad_norm": 1.2421875,
      "learning_rate": 5.230796735605304e-05,
      "loss": 4.3708,
      "mean_token_accuracy": 0.27195166796445847,
      "num_tokens": 19899673.0,
      "step": 8865
    },
    {
      "entropy": 5.2117900371551515,
      "epoch": 9.589507842076799,
      "grad_norm": 1.3515625,
      "learning_rate": 5.224709506942305e-05,
      "loss": 4.4173,
      "mean_token_accuracy": 0.267379529774189,
      "num_tokens": 19910448.0,
      "step": 8870
    },
    {
      "entropy": 5.299494123458862,
      "epoch": 9.594916170903192,
      "grad_norm": 1.296875,
      "learning_rate": 5.218703226736438e-05,
      "loss": 4.5387,
      "mean_token_accuracy": 0.2529213666915894,
      "num_tokens": 19921867.0,
      "step": 8875
    },
    {
      "entropy": 5.309925985336304,
      "epoch": 9.600324499729584,
      "grad_norm": 1.21875,
      "learning_rate": 5.212777916814533e-05,
      "loss": 4.5049,
      "mean_token_accuracy": 0.26647563427686694,
      "num_tokens": 19933753.0,
      "step": 8880
    },
    {
      "entropy": 5.230911588668823,
      "epoch": 9.605732828555976,
      "grad_norm": 1.28125,
      "learning_rate": 5.206933598709179e-05,
      "loss": 4.3877,
      "mean_token_accuracy": 0.2735194370150566,
      "num_tokens": 19944984.0,
      "step": 8885
    },
    {
      "entropy": 5.2518209457397464,
      "epoch": 9.611141157382368,
      "grad_norm": 1.453125,
      "learning_rate": 5.20117029365865e-05,
      "loss": 4.3587,
      "mean_token_accuracy": 0.28147747218608854,
      "num_tokens": 19955315.0,
      "step": 8890
    },
    {
      "entropy": 5.3051506042480465,
      "epoch": 9.616549486208761,
      "grad_norm": 1.328125,
      "learning_rate": 5.195488022606802e-05,
      "loss": 4.5012,
      "mean_token_accuracy": 0.26427939236164094,
      "num_tokens": 19967162.0,
      "step": 8895
    },
    {
      "entropy": 5.253988790512085,
      "epoch": 9.621957815035154,
      "grad_norm": 1.2734375,
      "learning_rate": 5.189886806203023e-05,
      "loss": 4.3446,
      "mean_token_accuracy": 0.2758065789937973,
      "num_tokens": 19978250.0,
      "step": 8900
    },
    {
      "entropy": 5.287500667572021,
      "epoch": 9.627366143861547,
      "grad_norm": 1.359375,
      "learning_rate": 5.184366664802149e-05,
      "loss": 4.4182,
      "mean_token_accuracy": 0.2699887529015541,
      "num_tokens": 19989794.0,
      "step": 8905
    },
    {
      "entropy": 5.186054849624634,
      "epoch": 9.63277447268794,
      "grad_norm": 1.359375,
      "learning_rate": 5.178927618464384e-05,
      "loss": 4.3212,
      "mean_token_accuracy": 0.27438395768404006,
      "num_tokens": 20000990.0,
      "step": 8910
    },
    {
      "entropy": 5.268441915512085,
      "epoch": 9.638182801514333,
      "grad_norm": 1.3046875,
      "learning_rate": 5.1735696869552364e-05,
      "loss": 4.392,
      "mean_token_accuracy": 0.2736589998006821,
      "num_tokens": 20012939.0,
      "step": 8915
    },
    {
      "entropy": 5.229291677474976,
      "epoch": 9.643591130340724,
      "grad_norm": 1.4296875,
      "learning_rate": 5.1682928897454416e-05,
      "loss": 4.4165,
      "mean_token_accuracy": 0.2637480109930038,
      "num_tokens": 20023553.0,
      "step": 8920
    },
    {
      "entropy": 5.20991268157959,
      "epoch": 9.648999459167117,
      "grad_norm": 1.484375,
      "learning_rate": 5.1630972460108904e-05,
      "loss": 4.3855,
      "mean_token_accuracy": 0.27071274518966676,
      "num_tokens": 20033650.0,
      "step": 8925
    },
    {
      "entropy": 5.265422821044922,
      "epoch": 9.65440778799351,
      "grad_norm": 1.46875,
      "learning_rate": 5.1579827746325694e-05,
      "loss": 4.5349,
      "mean_token_accuracy": 0.2589650586247444,
      "num_tokens": 20044281.0,
      "step": 8930
    },
    {
      "entropy": 5.261706066131592,
      "epoch": 9.659816116819902,
      "grad_norm": 1.296875,
      "learning_rate": 5.152949494196472e-05,
      "loss": 4.4677,
      "mean_token_accuracy": 0.2650755256414413,
      "num_tokens": 20056396.0,
      "step": 8935
    },
    {
      "entropy": 5.270640182495117,
      "epoch": 9.665224445646295,
      "grad_norm": 1.3671875,
      "learning_rate": 5.1479974229935575e-05,
      "loss": 4.5367,
      "mean_token_accuracy": 0.2560572549700737,
      "num_tokens": 20068712.0,
      "step": 8940
    },
    {
      "entropy": 5.286727619171143,
      "epoch": 9.670632774472688,
      "grad_norm": 1.359375,
      "learning_rate": 5.1431265790196583e-05,
      "loss": 4.4341,
      "mean_token_accuracy": 0.27166842967271804,
      "num_tokens": 20082439.0,
      "step": 8945
    },
    {
      "entropy": 5.200519990921021,
      "epoch": 9.676041103299081,
      "grad_norm": 1.34375,
      "learning_rate": 5.138336979975439e-05,
      "loss": 4.4033,
      "mean_token_accuracy": 0.26269145756959916,
      "num_tokens": 20094068.0,
      "step": 8950
    },
    {
      "entropy": 5.223095846176148,
      "epoch": 9.681449432125474,
      "grad_norm": 1.328125,
      "learning_rate": 5.13362864326631e-05,
      "loss": 4.3841,
      "mean_token_accuracy": 0.2819969952106476,
      "num_tokens": 20105166.0,
      "step": 8955
    },
    {
      "entropy": 5.263787508010864,
      "epoch": 9.686857760951867,
      "grad_norm": 1.4453125,
      "learning_rate": 5.12900158600238e-05,
      "loss": 4.444,
      "mean_token_accuracy": 0.26118995249271393,
      "num_tokens": 20116107.0,
      "step": 8960
    },
    {
      "entropy": 5.286194372177124,
      "epoch": 9.692266089778258,
      "grad_norm": 1.3125,
      "learning_rate": 5.12445582499838e-05,
      "loss": 4.45,
      "mean_token_accuracy": 0.26690921634435655,
      "num_tokens": 20126979.0,
      "step": 8965
    },
    {
      "entropy": 5.230099678039551,
      "epoch": 9.69767441860465,
      "grad_norm": 1.2578125,
      "learning_rate": 5.119991376773624e-05,
      "loss": 4.4155,
      "mean_token_accuracy": 0.2703288897871971,
      "num_tokens": 20139122.0,
      "step": 8970
    },
    {
      "entropy": 5.272184228897094,
      "epoch": 9.703082747431043,
      "grad_norm": 1.4296875,
      "learning_rate": 5.115608257551923e-05,
      "loss": 4.5106,
      "mean_token_accuracy": 0.2703643023967743,
      "num_tokens": 20150284.0,
      "step": 8975
    },
    {
      "entropy": 5.213010597229004,
      "epoch": 9.708491076257436,
      "grad_norm": 1.3203125,
      "learning_rate": 5.11130648326154e-05,
      "loss": 4.4745,
      "mean_token_accuracy": 0.26678899526596067,
      "num_tokens": 20161026.0,
      "step": 8980
    },
    {
      "entropy": 5.275950288772583,
      "epoch": 9.71389940508383,
      "grad_norm": 1.234375,
      "learning_rate": 5.107086069535136e-05,
      "loss": 4.4931,
      "mean_token_accuracy": 0.26854248344898224,
      "num_tokens": 20172883.0,
      "step": 8985
    },
    {
      "entropy": 5.283819389343262,
      "epoch": 9.719307733910222,
      "grad_norm": 1.3515625,
      "learning_rate": 5.1029470317097044e-05,
      "loss": 4.4052,
      "mean_token_accuracy": 0.27596860826015474,
      "num_tokens": 20183624.0,
      "step": 8990
    },
    {
      "entropy": 5.276701593399048,
      "epoch": 9.724716062736615,
      "grad_norm": 1.4296875,
      "learning_rate": 5.098889384826519e-05,
      "loss": 4.427,
      "mean_token_accuracy": 0.2674079716205597,
      "num_tokens": 20194200.0,
      "step": 8995
    },
    {
      "entropy": 5.229514503479004,
      "epoch": 9.730124391563008,
      "grad_norm": 1.2109375,
      "learning_rate": 5.094913143631073e-05,
      "loss": 4.4499,
      "mean_token_accuracy": 0.2567264512181282,
      "num_tokens": 20206207.0,
      "step": 9000
    },
    {
      "epoch": 9.730124391563008,
      "eval_entropy": 5.188457922501997,
      "eval_loss": 5.859401702880859,
      "eval_mean_token_accuracy": 0.1827453392811797,
      "eval_num_tokens": 20206207.0,
      "eval_runtime": 1.9473,
      "eval_samples_per_second": 1805.057,
      "eval_steps_per_second": 225.953,
      "step": 9000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4678174147461120.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}