{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0811249323958896,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.692032432556152,
      "epoch": 0.005408328826392645,
      "grad_norm": 13.4375,
      "learning_rate": 2e-06,
      "loss": 10.8497,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10948.0,
      "step": 5
    },
    {
      "entropy": 10.691972923278808,
      "epoch": 0.01081665765278529,
      "grad_norm": 13.4375,
      "learning_rate": 4.5e-06,
      "loss": 10.7725,
      "mean_token_accuracy": 0.00017152659129351377,
      "num_tokens": 21455.0,
      "step": 10
    },
    {
      "entropy": 10.691221618652344,
      "epoch": 0.016224986479177934,
      "grad_norm": 10.125,
      "learning_rate": 7e-06,
      "loss": 10.508,
      "mean_token_accuracy": 0.017251307470723988,
      "num_tokens": 32377.0,
      "step": 15
    },
    {
      "entropy": 10.68411626815796,
      "epoch": 0.02163331530557058,
      "grad_norm": 6.6875,
      "learning_rate": 9.5e-06,
      "loss": 10.1253,
      "mean_token_accuracy": 0.036476511135697366,
      "num_tokens": 42276.0,
      "step": 20
    },
    {
      "entropy": 10.645281887054443,
      "epoch": 0.02704164413196322,
      "grad_norm": 4.21875,
      "learning_rate": 1.2e-05,
      "loss": 9.8249,
      "mean_token_accuracy": 0.03992591369897127,
      "num_tokens": 55049.0,
      "step": 25
    },
    {
      "entropy": 10.584981632232665,
      "epoch": 0.03244997295835587,
      "grad_norm": 3.203125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.6624,
      "mean_token_accuracy": 0.03540853708982468,
      "num_tokens": 66792.0,
      "step": 30
    },
    {
      "entropy": 10.57096061706543,
      "epoch": 0.03785830178474851,
      "grad_norm": 2.9375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.5971,
      "mean_token_accuracy": 0.03990415818989277,
      "num_tokens": 78208.0,
      "step": 35
    },
    {
      "entropy": 10.563229370117188,
      "epoch": 0.04326663061114116,
      "grad_norm": 2.703125,
      "learning_rate": 1.95e-05,
      "loss": 9.5045,
      "mean_token_accuracy": 0.039873688109219076,
      "num_tokens": 88871.0,
      "step": 40
    },
    {
      "entropy": 10.554053497314452,
      "epoch": 0.048674959437533805,
      "grad_norm": 2.71875,
      "learning_rate": 2.2e-05,
      "loss": 9.509,
      "mean_token_accuracy": 0.0394243435934186,
      "num_tokens": 100118.0,
      "step": 45
    },
    {
      "entropy": 10.538364601135253,
      "epoch": 0.05408328826392644,
      "grad_norm": 2.390625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4423,
      "mean_token_accuracy": 0.048855461180210114,
      "num_tokens": 113120.0,
      "step": 50
    },
    {
      "entropy": 10.51687936782837,
      "epoch": 0.05949161709031909,
      "grad_norm": 2.6875,
      "learning_rate": 2.7e-05,
      "loss": 9.3524,
      "mean_token_accuracy": 0.046342721208930016,
      "num_tokens": 123220.0,
      "step": 55
    },
    {
      "entropy": 10.486348628997803,
      "epoch": 0.06489994591671173,
      "grad_norm": 2.296875,
      "learning_rate": 2.95e-05,
      "loss": 9.3335,
      "mean_token_accuracy": 0.057121139764785764,
      "num_tokens": 134479.0,
      "step": 60
    },
    {
      "entropy": 10.347621631622314,
      "epoch": 0.07030827474310439,
      "grad_norm": 2.453125,
      "learning_rate": 3.2e-05,
      "loss": 9.1695,
      "mean_token_accuracy": 0.06469983011484146,
      "num_tokens": 145189.0,
      "step": 65
    },
    {
      "entropy": 10.380140495300292,
      "epoch": 0.07571660356949703,
      "grad_norm": 2.21875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.0849,
      "mean_token_accuracy": 0.0679698098450899,
      "num_tokens": 157354.0,
      "step": 70
    },
    {
      "entropy": 10.407375621795655,
      "epoch": 0.08112493239588967,
      "grad_norm": 2.15625,
      "learning_rate": 3.7e-05,
      "loss": 9.0362,
      "mean_token_accuracy": 0.06854141913354397,
      "num_tokens": 168821.0,
      "step": 75
    },
    {
      "entropy": 10.368484210968017,
      "epoch": 0.08653326122228232,
      "grad_norm": 2.421875,
      "learning_rate": 3.95e-05,
      "loss": 8.8889,
      "mean_token_accuracy": 0.06982190646231175,
      "num_tokens": 180554.0,
      "step": 80
    },
    {
      "entropy": 10.238024425506591,
      "epoch": 0.09194159004867496,
      "grad_norm": 2.421875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.7914,
      "mean_token_accuracy": 0.07617409080266953,
      "num_tokens": 191229.0,
      "step": 85
    },
    {
      "entropy": 10.228119373321533,
      "epoch": 0.09734991887506761,
      "grad_norm": 2.28125,
      "learning_rate": 4.45e-05,
      "loss": 8.6865,
      "mean_token_accuracy": 0.07864613682031632,
      "num_tokens": 201903.0,
      "step": 90
    },
    {
      "entropy": 10.117883205413818,
      "epoch": 0.10275824770146025,
      "grad_norm": 2.15625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6671,
      "mean_token_accuracy": 0.07588070593774318,
      "num_tokens": 214255.0,
      "step": 95
    },
    {
      "entropy": 10.088370513916015,
      "epoch": 0.10816657652785289,
      "grad_norm": 2.265625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.4926,
      "mean_token_accuracy": 0.08170793950557709,
      "num_tokens": 225423.0,
      "step": 100
    },
    {
      "entropy": 10.01626272201538,
      "epoch": 0.11357490535424554,
      "grad_norm": 2.765625,
      "learning_rate": 5.2e-05,
      "loss": 8.4216,
      "mean_token_accuracy": 0.0811295285820961,
      "num_tokens": 237801.0,
      "step": 105
    },
    {
      "entropy": 9.888921546936036,
      "epoch": 0.11898323418063818,
      "grad_norm": 2.0625,
      "learning_rate": 5.45e-05,
      "loss": 8.2161,
      "mean_token_accuracy": 0.08590030111372471,
      "num_tokens": 249195.0,
      "step": 110
    },
    {
      "entropy": 9.866813278198242,
      "epoch": 0.12439156300703083,
      "grad_norm": 1.921875,
      "learning_rate": 5.7e-05,
      "loss": 8.0932,
      "mean_token_accuracy": 0.09252363666892052,
      "num_tokens": 260129.0,
      "step": 115
    },
    {
      "entropy": 9.781723117828369,
      "epoch": 0.12979989183342347,
      "grad_norm": 1.8359375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1482,
      "mean_token_accuracy": 0.07961882352828979,
      "num_tokens": 271220.0,
      "step": 120
    },
    {
      "entropy": 9.653579044342042,
      "epoch": 0.1352082206598161,
      "grad_norm": 1.7421875,
      "learning_rate": 6.2e-05,
      "loss": 7.9073,
      "mean_token_accuracy": 0.08841332867741585,
      "num_tokens": 281560.0,
      "step": 125
    },
    {
      "entropy": 9.472657203674316,
      "epoch": 0.14061654948620878,
      "grad_norm": 1.5703125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.7576,
      "mean_token_accuracy": 0.08678923547267914,
      "num_tokens": 292518.0,
      "step": 130
    },
    {
      "entropy": 9.37501802444458,
      "epoch": 0.1460248783126014,
      "grad_norm": 1.453125,
      "learning_rate": 6.7e-05,
      "loss": 7.7439,
      "mean_token_accuracy": 0.08266892358660698,
      "num_tokens": 304865.0,
      "step": 135
    },
    {
      "entropy": 9.135330295562744,
      "epoch": 0.15143320713899405,
      "grad_norm": 1.5625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.5576,
      "mean_token_accuracy": 0.0914510041475296,
      "num_tokens": 315951.0,
      "step": 140
    },
    {
      "entropy": 8.899916648864746,
      "epoch": 0.1568415359653867,
      "grad_norm": 1.4765625,
      "learning_rate": 7.2e-05,
      "loss": 7.4698,
      "mean_token_accuracy": 0.0939729243516922,
      "num_tokens": 327094.0,
      "step": 145
    },
    {
      "entropy": 8.730432033538818,
      "epoch": 0.16224986479177933,
      "grad_norm": 1.4140625,
      "learning_rate": 7.45e-05,
      "loss": 7.4175,
      "mean_token_accuracy": 0.09195762872695923,
      "num_tokens": 337801.0,
      "step": 150
    },
    {
      "entropy": 8.471641349792481,
      "epoch": 0.167658193618172,
      "grad_norm": 1.140625,
      "learning_rate": 7.7e-05,
      "loss": 7.3256,
      "mean_token_accuracy": 0.09636828005313873,
      "num_tokens": 349039.0,
      "step": 155
    },
    {
      "entropy": 8.36181583404541,
      "epoch": 0.17306652244456464,
      "grad_norm": 1.1328125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.3141,
      "mean_token_accuracy": 0.09097177907824516,
      "num_tokens": 360639.0,
      "step": 160
    },
    {
      "entropy": 8.211484432220459,
      "epoch": 0.17847485127095727,
      "grad_norm": 1.4765625,
      "learning_rate": 8.2e-05,
      "loss": 7.3216,
      "mean_token_accuracy": 0.09984223693609237,
      "num_tokens": 371005.0,
      "step": 165
    },
    {
      "entropy": 8.06982226371765,
      "epoch": 0.1838831800973499,
      "grad_norm": 2.078125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.1972,
      "mean_token_accuracy": 0.09530351907014847,
      "num_tokens": 383340.0,
      "step": 170
    },
    {
      "entropy": 7.89442892074585,
      "epoch": 0.18929150892374255,
      "grad_norm": 1.375,
      "learning_rate": 8.7e-05,
      "loss": 7.1349,
      "mean_token_accuracy": 0.10384258776903152,
      "num_tokens": 394682.0,
      "step": 175
    },
    {
      "entropy": 7.894894552230835,
      "epoch": 0.19469983775013522,
      "grad_norm": 1.2265625,
      "learning_rate": 8.95e-05,
      "loss": 7.1506,
      "mean_token_accuracy": 0.09678633958101272,
      "num_tokens": 405713.0,
      "step": 180
    },
    {
      "entropy": 7.920239067077636,
      "epoch": 0.20010816657652786,
      "grad_norm": 1.3203125,
      "learning_rate": 9.2e-05,
      "loss": 7.0992,
      "mean_token_accuracy": 0.09813495352864265,
      "num_tokens": 416208.0,
      "step": 185
    },
    {
      "entropy": 7.751987409591675,
      "epoch": 0.2055164954029205,
      "grad_norm": 1.2890625,
      "learning_rate": 9.45e-05,
      "loss": 7.0093,
      "mean_token_accuracy": 0.0977411150932312,
      "num_tokens": 427595.0,
      "step": 190
    },
    {
      "entropy": 7.748253011703492,
      "epoch": 0.21092482422931313,
      "grad_norm": 1.546875,
      "learning_rate": 9.7e-05,
      "loss": 7.1353,
      "mean_token_accuracy": 0.10629882290959358,
      "num_tokens": 438273.0,
      "step": 195
    },
    {
      "entropy": 7.726998567581177,
      "epoch": 0.21633315305570577,
      "grad_norm": 1.34375,
      "learning_rate": 9.95e-05,
      "loss": 7.0456,
      "mean_token_accuracy": 0.09703748449683189,
      "num_tokens": 448948.0,
      "step": 200
    },
    {
      "entropy": 7.69028058052063,
      "epoch": 0.22174148188209844,
      "grad_norm": 1.421875,
      "learning_rate": 0.000102,
      "loss": 7.0949,
      "mean_token_accuracy": 0.09859743788838386,
      "num_tokens": 460838.0,
      "step": 205
    },
    {
      "entropy": 7.614819478988648,
      "epoch": 0.22714981070849108,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0979,
      "mean_token_accuracy": 0.09938543438911437,
      "num_tokens": 473555.0,
      "step": 210
    },
    {
      "entropy": 7.5858986377716064,
      "epoch": 0.23255813953488372,
      "grad_norm": 1.625,
      "learning_rate": 0.000107,
      "loss": 7.0348,
      "mean_token_accuracy": 0.10209831669926643,
      "num_tokens": 485221.0,
      "step": 215
    },
    {
      "entropy": 7.542715787887573,
      "epoch": 0.23796646836127636,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001095,
      "loss": 6.9358,
      "mean_token_accuracy": 0.10402886867523194,
      "num_tokens": 496552.0,
      "step": 220
    },
    {
      "entropy": 7.515826463699341,
      "epoch": 0.24337479718766902,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000112,
      "loss": 6.9993,
      "mean_token_accuracy": 0.10345738977193833,
      "num_tokens": 509760.0,
      "step": 225
    },
    {
      "entropy": 7.480216789245605,
      "epoch": 0.24878312601406166,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001145,
      "loss": 6.8336,
      "mean_token_accuracy": 0.11249553188681602,
      "num_tokens": 521175.0,
      "step": 230
    },
    {
      "entropy": 7.457189130783081,
      "epoch": 0.2541914548404543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9476,
      "mean_token_accuracy": 0.1091655321419239,
      "num_tokens": 532597.0,
      "step": 235
    },
    {
      "entropy": 7.4506001472473145,
      "epoch": 0.25959978366684694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0133,
      "mean_token_accuracy": 0.10449624136090278,
      "num_tokens": 545661.0,
      "step": 240
    },
    {
      "entropy": 7.514426422119141,
      "epoch": 0.2650081124932396,
      "grad_norm": 1.34375,
      "learning_rate": 0.000122,
      "loss": 6.979,
      "mean_token_accuracy": 0.10253533497452735,
      "num_tokens": 557336.0,
      "step": 245
    },
    {
      "entropy": 7.388407945632935,
      "epoch": 0.2704164413196322,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001245,
      "loss": 6.8425,
      "mean_token_accuracy": 0.11243968978524208,
      "num_tokens": 567398.0,
      "step": 250
    },
    {
      "entropy": 7.362952089309692,
      "epoch": 0.27582477014602486,
      "grad_norm": 1.28125,
      "learning_rate": 0.000127,
      "loss": 6.9312,
      "mean_token_accuracy": 0.10559455379843712,
      "num_tokens": 579543.0,
      "step": 255
    },
    {
      "entropy": 7.340886926651001,
      "epoch": 0.28123309897241755,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001295,
      "loss": 6.753,
      "mean_token_accuracy": 0.11401380077004433,
      "num_tokens": 589218.0,
      "step": 260
    },
    {
      "entropy": 7.349085855484009,
      "epoch": 0.2866414277988102,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000132,
      "loss": 6.8123,
      "mean_token_accuracy": 0.11433196514844894,
      "num_tokens": 600327.0,
      "step": 265
    },
    {
      "entropy": 7.245560216903686,
      "epoch": 0.2920497566252028,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.7717,
      "mean_token_accuracy": 0.10991677045822143,
      "num_tokens": 610923.0,
      "step": 270
    },
    {
      "entropy": 7.36246280670166,
      "epoch": 0.29745808545159547,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8509,
      "mean_token_accuracy": 0.10712052285671234,
      "num_tokens": 621138.0,
      "step": 275
    },
    {
      "entropy": 7.193227529525757,
      "epoch": 0.3028664142779881,
      "grad_norm": 1.953125,
      "learning_rate": 0.0001395,
      "loss": 6.797,
      "mean_token_accuracy": 0.11277740374207497,
      "num_tokens": 631440.0,
      "step": 280
    },
    {
      "entropy": 7.308440208435059,
      "epoch": 0.30827474310438074,
      "grad_norm": 1.625,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8392,
      "mean_token_accuracy": 0.11829005107283593,
      "num_tokens": 642328.0,
      "step": 285
    },
    {
      "entropy": 7.2388709545135494,
      "epoch": 0.3136830719307734,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001445,
      "loss": 6.7809,
      "mean_token_accuracy": 0.11447241306304931,
      "num_tokens": 652875.0,
      "step": 290
    },
    {
      "entropy": 7.163767290115357,
      "epoch": 0.319091400757166,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000147,
      "loss": 6.717,
      "mean_token_accuracy": 0.1139206700026989,
      "num_tokens": 663726.0,
      "step": 295
    },
    {
      "entropy": 7.29945330619812,
      "epoch": 0.32449972958355866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001495,
      "loss": 6.9178,
      "mean_token_accuracy": 0.11128493323922158,
      "num_tokens": 675638.0,
      "step": 300
    },
    {
      "entropy": 7.235377931594849,
      "epoch": 0.3299080584099513,
      "grad_norm": 1.09375,
      "learning_rate": 0.000152,
      "loss": 6.7798,
      "mean_token_accuracy": 0.11265585124492646,
      "num_tokens": 687239.0,
      "step": 305
    },
    {
      "entropy": 7.244372606277466,
      "epoch": 0.335316387236344,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8244,
      "mean_token_accuracy": 0.10671208873391151,
      "num_tokens": 698905.0,
      "step": 310
    },
    {
      "entropy": 7.157739448547363,
      "epoch": 0.34072471606273663,
      "grad_norm": 1.40625,
      "learning_rate": 0.000157,
      "loss": 6.8115,
      "mean_token_accuracy": 0.11196564212441444,
      "num_tokens": 710602.0,
      "step": 315
    },
    {
      "entropy": 7.112873458862305,
      "epoch": 0.34613304488912927,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001595,
      "loss": 6.7691,
      "mean_token_accuracy": 0.11123333275318145,
      "num_tokens": 721326.0,
      "step": 320
    },
    {
      "entropy": 7.238367700576783,
      "epoch": 0.3515413737155219,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000162,
      "loss": 6.743,
      "mean_token_accuracy": 0.11625269651412964,
      "num_tokens": 732972.0,
      "step": 325
    },
    {
      "entropy": 7.11000337600708,
      "epoch": 0.35694970254191455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7174,
      "mean_token_accuracy": 0.11805651634931565,
      "num_tokens": 743624.0,
      "step": 330
    },
    {
      "entropy": 7.168763875961304,
      "epoch": 0.3623580313683072,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8101,
      "mean_token_accuracy": 0.11515277996659279,
      "num_tokens": 756384.0,
      "step": 335
    },
    {
      "entropy": 7.154894590377808,
      "epoch": 0.3677663601946998,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7909,
      "mean_token_accuracy": 0.11520344465970993,
      "num_tokens": 767256.0,
      "step": 340
    },
    {
      "entropy": 7.1156007766723635,
      "epoch": 0.37317468902109246,
      "grad_norm": 1.25,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7043,
      "mean_token_accuracy": 0.11901735439896584,
      "num_tokens": 779059.0,
      "step": 345
    },
    {
      "entropy": 7.027617883682251,
      "epoch": 0.3785830178474851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.8016,
      "mean_token_accuracy": 0.11899174749851227,
      "num_tokens": 790291.0,
      "step": 350
    },
    {
      "entropy": 7.1015932083129885,
      "epoch": 0.3839913466738778,
      "grad_norm": 1.40625,
      "learning_rate": 0.000177,
      "loss": 6.7063,
      "mean_token_accuracy": 0.11784811541438103,
      "num_tokens": 801450.0,
      "step": 355
    },
    {
      "entropy": 7.017753171920776,
      "epoch": 0.38939967550027044,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001795,
      "loss": 6.625,
      "mean_token_accuracy": 0.1219206415116787,
      "num_tokens": 812428.0,
      "step": 360
    },
    {
      "entropy": 7.003644132614136,
      "epoch": 0.3948080043266631,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000182,
      "loss": 6.7244,
      "mean_token_accuracy": 0.1133291907608509,
      "num_tokens": 823644.0,
      "step": 365
    },
    {
      "entropy": 7.151744842529297,
      "epoch": 0.4002163331530557,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001845,
      "loss": 6.7116,
      "mean_token_accuracy": 0.12677086964249612,
      "num_tokens": 834064.0,
      "step": 370
    },
    {
      "entropy": 7.033484983444214,
      "epoch": 0.40562466197944835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000187,
      "loss": 6.6546,
      "mean_token_accuracy": 0.11925762817263603,
      "num_tokens": 845207.0,
      "step": 375
    },
    {
      "entropy": 6.970419311523438,
      "epoch": 0.411032990805841,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001895,
      "loss": 6.7181,
      "mean_token_accuracy": 0.11609353423118592,
      "num_tokens": 855564.0,
      "step": 380
    },
    {
      "entropy": 7.12083477973938,
      "epoch": 0.41644131963223363,
      "grad_norm": 1.203125,
      "learning_rate": 0.000192,
      "loss": 6.7377,
      "mean_token_accuracy": 0.12403716966509819,
      "num_tokens": 866959.0,
      "step": 385
    },
    {
      "entropy": 6.928013324737549,
      "epoch": 0.42184964845862627,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001945,
      "loss": 6.6585,
      "mean_token_accuracy": 0.125290846824646,
      "num_tokens": 877319.0,
      "step": 390
    },
    {
      "entropy": 7.049751853942871,
      "epoch": 0.4272579772850189,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7489,
      "mean_token_accuracy": 0.11747724786400796,
      "num_tokens": 888037.0,
      "step": 395
    },
    {
      "entropy": 6.912725925445557,
      "epoch": 0.43266630611141155,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.4906,
      "mean_token_accuracy": 0.12297609969973564,
      "num_tokens": 898775.0,
      "step": 400
    },
    {
      "entropy": 7.0705005645751955,
      "epoch": 0.43807463493780424,
      "grad_norm": 1.421875,
      "learning_rate": 0.000202,
      "loss": 6.6433,
      "mean_token_accuracy": 0.12098436206579208,
      "num_tokens": 909513.0,
      "step": 405
    },
    {
      "entropy": 6.938232803344727,
      "epoch": 0.4434829637641969,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6959,
      "mean_token_accuracy": 0.11758999228477478,
      "num_tokens": 920730.0,
      "step": 410
    },
    {
      "entropy": 6.924176979064941,
      "epoch": 0.4488912925905895,
      "grad_norm": 1.3125,
      "learning_rate": 0.000207,
      "loss": 6.6151,
      "mean_token_accuracy": 0.12237995192408561,
      "num_tokens": 931457.0,
      "step": 415
    },
    {
      "entropy": 6.865782737731934,
      "epoch": 0.45429962141698216,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002095,
      "loss": 6.6026,
      "mean_token_accuracy": 0.12269196733832359,
      "num_tokens": 942572.0,
      "step": 420
    },
    {
      "entropy": 6.9836249351501465,
      "epoch": 0.4597079502433748,
      "grad_norm": 1.40625,
      "learning_rate": 0.000212,
      "loss": 6.607,
      "mean_token_accuracy": 0.11696747690439224,
      "num_tokens": 953132.0,
      "step": 425
    },
    {
      "entropy": 6.915006303787232,
      "epoch": 0.46511627906976744,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002145,
      "loss": 6.6414,
      "mean_token_accuracy": 0.11870383992791175,
      "num_tokens": 963551.0,
      "step": 430
    },
    {
      "entropy": 6.957869148254394,
      "epoch": 0.4705246078961601,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.599,
      "mean_token_accuracy": 0.12203558534383774,
      "num_tokens": 975297.0,
      "step": 435
    },
    {
      "entropy": 6.861082458496094,
      "epoch": 0.4759329367225527,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002195,
      "loss": 6.6076,
      "mean_token_accuracy": 0.12166279256343841,
      "num_tokens": 986910.0,
      "step": 440
    },
    {
      "entropy": 6.963198518753051,
      "epoch": 0.48134126554894535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000222,
      "loss": 6.7303,
      "mean_token_accuracy": 0.12663792818784714,
      "num_tokens": 997058.0,
      "step": 445
    },
    {
      "entropy": 6.887674808502197,
      "epoch": 0.48674959437533805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002245,
      "loss": 6.6118,
      "mean_token_accuracy": 0.11577775105834007,
      "num_tokens": 1008261.0,
      "step": 450
    },
    {
      "entropy": 6.85930438041687,
      "epoch": 0.4921579232017307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.527,
      "mean_token_accuracy": 0.1269462674856186,
      "num_tokens": 1019193.0,
      "step": 455
    },
    {
      "entropy": 6.86927375793457,
      "epoch": 0.4975662520281233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6448,
      "mean_token_accuracy": 0.12487371489405633,
      "num_tokens": 1031227.0,
      "step": 460
    },
    {
      "entropy": 6.828432273864746,
      "epoch": 0.502974580854516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.4702,
      "mean_token_accuracy": 0.1266406588256359,
      "num_tokens": 1042963.0,
      "step": 465
    },
    {
      "entropy": 6.750533151626587,
      "epoch": 0.5083829096809086,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5397,
      "mean_token_accuracy": 0.11756750047206879,
      "num_tokens": 1053709.0,
      "step": 470
    },
    {
      "entropy": 6.890934610366822,
      "epoch": 0.5137912385073012,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000237,
      "loss": 6.5681,
      "mean_token_accuracy": 0.13536556139588357,
      "num_tokens": 1064527.0,
      "step": 475
    },
    {
      "entropy": 6.778618907928466,
      "epoch": 0.5191995673336939,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002395,
      "loss": 6.6317,
      "mean_token_accuracy": 0.12472611665725708,
      "num_tokens": 1075127.0,
      "step": 480
    },
    {
      "entropy": 6.814949655532837,
      "epoch": 0.5246078961600865,
      "grad_norm": 1.15625,
      "learning_rate": 0.000242,
      "loss": 6.6266,
      "mean_token_accuracy": 0.11958301439881325,
      "num_tokens": 1086581.0,
      "step": 485
    },
    {
      "entropy": 6.7194966793060305,
      "epoch": 0.5300162249864792,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002445,
      "loss": 6.55,
      "mean_token_accuracy": 0.13253298550844192,
      "num_tokens": 1097818.0,
      "step": 490
    },
    {
      "entropy": 6.829859018325806,
      "epoch": 0.5354245538128718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000247,
      "loss": 6.5252,
      "mean_token_accuracy": 0.12295517474412918,
      "num_tokens": 1108043.0,
      "step": 495
    },
    {
      "entropy": 6.82882833480835,
      "epoch": 0.5408328826392644,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002495,
      "loss": 6.592,
      "mean_token_accuracy": 0.12201916500926017,
      "num_tokens": 1119875.0,
      "step": 500
    },
    {
      "entropy": 6.7380537509918215,
      "epoch": 0.5462412114656571,
      "grad_norm": 1.390625,
      "learning_rate": 0.000252,
      "loss": 6.4926,
      "mean_token_accuracy": 0.1267081081867218,
      "num_tokens": 1132406.0,
      "step": 505
    },
    {
      "entropy": 6.946386671066284,
      "epoch": 0.5516495402920497,
      "grad_norm": 1.25,
      "learning_rate": 0.0002545,
      "loss": 6.6676,
      "mean_token_accuracy": 0.11859372109174729,
      "num_tokens": 1143828.0,
      "step": 510
    },
    {
      "entropy": 6.720641136169434,
      "epoch": 0.5570578691184424,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000257,
      "loss": 6.4863,
      "mean_token_accuracy": 0.12196403294801712,
      "num_tokens": 1155482.0,
      "step": 515
    },
    {
      "entropy": 6.651684331893921,
      "epoch": 0.5624661979448351,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002595,
      "loss": 6.4324,
      "mean_token_accuracy": 0.13508962318301201,
      "num_tokens": 1166847.0,
      "step": 520
    },
    {
      "entropy": 6.7738018989562985,
      "epoch": 0.5678745267712277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000262,
      "loss": 6.5108,
      "mean_token_accuracy": 0.1282059669494629,
      "num_tokens": 1177419.0,
      "step": 525
    },
    {
      "entropy": 6.822676420211792,
      "epoch": 0.5732828555976204,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6815,
      "mean_token_accuracy": 0.11861670166254043,
      "num_tokens": 1188975.0,
      "step": 530
    },
    {
      "entropy": 6.75103759765625,
      "epoch": 0.578691184424013,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5153,
      "mean_token_accuracy": 0.12072157859802246,
      "num_tokens": 1200252.0,
      "step": 535
    },
    {
      "entropy": 6.676786279678344,
      "epoch": 0.5840995132504057,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.5459,
      "mean_token_accuracy": 0.13504153192043306,
      "num_tokens": 1211864.0,
      "step": 540
    },
    {
      "entropy": 6.7417017936706545,
      "epoch": 0.5895078420767983,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.4194,
      "mean_token_accuracy": 0.12843444123864173,
      "num_tokens": 1222429.0,
      "step": 545
    },
    {
      "entropy": 6.745041131973267,
      "epoch": 0.5949161709031909,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002745,
      "loss": 6.5458,
      "mean_token_accuracy": 0.12370594069361687,
      "num_tokens": 1233782.0,
      "step": 550
    },
    {
      "entropy": 6.790885591506958,
      "epoch": 0.6003244997295836,
      "grad_norm": 1.21875,
      "learning_rate": 0.000277,
      "loss": 6.4923,
      "mean_token_accuracy": 0.12668371945619583,
      "num_tokens": 1244309.0,
      "step": 555
    },
    {
      "entropy": 6.6354663372039795,
      "epoch": 0.6057328285559762,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002795,
      "loss": 6.4497,
      "mean_token_accuracy": 0.12727595791220664,
      "num_tokens": 1256403.0,
      "step": 560
    },
    {
      "entropy": 6.775501108169555,
      "epoch": 0.6111411573823688,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5459,
      "mean_token_accuracy": 0.1229689784348011,
      "num_tokens": 1268758.0,
      "step": 565
    },
    {
      "entropy": 6.60977520942688,
      "epoch": 0.6165494862087615,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002845,
      "loss": 6.4351,
      "mean_token_accuracy": 0.1309477686882019,
      "num_tokens": 1279355.0,
      "step": 570
    },
    {
      "entropy": 6.6845152378082275,
      "epoch": 0.6219578150351541,
      "grad_norm": 1.09375,
      "learning_rate": 0.000287,
      "loss": 6.4289,
      "mean_token_accuracy": 0.12401615604758262,
      "num_tokens": 1290619.0,
      "step": 575
    },
    {
      "entropy": 6.649652814865112,
      "epoch": 0.6273661438615468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002895,
      "loss": 6.3938,
      "mean_token_accuracy": 0.13154794424772262,
      "num_tokens": 1301851.0,
      "step": 580
    },
    {
      "entropy": 6.745173740386963,
      "epoch": 0.6327744726879394,
      "grad_norm": 1.109375,
      "learning_rate": 0.000292,
      "loss": 6.5466,
      "mean_token_accuracy": 0.13009120672941207,
      "num_tokens": 1313260.0,
      "step": 585
    },
    {
      "entropy": 6.597011137008667,
      "epoch": 0.638182801514332,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002945,
      "loss": 6.5331,
      "mean_token_accuracy": 0.12869492620229722,
      "num_tokens": 1324189.0,
      "step": 590
    },
    {
      "entropy": 6.752423334121704,
      "epoch": 0.6435911303407247,
      "grad_norm": 1.109375,
      "learning_rate": 0.000297,
      "loss": 6.5636,
      "mean_token_accuracy": 0.12625337839126588,
      "num_tokens": 1335765.0,
      "step": 595
    },
    {
      "entropy": 6.725863218307495,
      "epoch": 0.6489994591671173,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002995,
      "loss": 6.4908,
      "mean_token_accuracy": 0.12820648178458213,
      "num_tokens": 1347653.0,
      "step": 600
    },
    {
      "entropy": 6.657638025283814,
      "epoch": 0.65440778799351,
      "grad_norm": 1.046875,
      "learning_rate": 0.000302,
      "loss": 6.4944,
      "mean_token_accuracy": 0.12864069044589996,
      "num_tokens": 1360201.0,
      "step": 605
    },
    {
      "entropy": 6.614993000030518,
      "epoch": 0.6598161168199026,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003045,
      "loss": 6.354,
      "mean_token_accuracy": 0.1303904466331005,
      "num_tokens": 1370974.0,
      "step": 610
    },
    {
      "entropy": 6.537722396850586,
      "epoch": 0.6652244456462953,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000307,
      "loss": 6.4304,
      "mean_token_accuracy": 0.1266734465956688,
      "num_tokens": 1383855.0,
      "step": 615
    },
    {
      "entropy": 6.5735701560974125,
      "epoch": 0.670632774472688,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003095,
      "loss": 6.3969,
      "mean_token_accuracy": 0.13598421812057496,
      "num_tokens": 1395041.0,
      "step": 620
    },
    {
      "entropy": 6.759116458892822,
      "epoch": 0.6760411032990806,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000312,
      "loss": 6.5041,
      "mean_token_accuracy": 0.12941347509622575,
      "num_tokens": 1406102.0,
      "step": 625
    },
    {
      "entropy": 6.4543132305145265,
      "epoch": 0.6814494321254733,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003145,
      "loss": 6.3253,
      "mean_token_accuracy": 0.12973413094878197,
      "num_tokens": 1417713.0,
      "step": 630
    },
    {
      "entropy": 6.5458276748657225,
      "epoch": 0.6868577609518659,
      "grad_norm": 1.21875,
      "learning_rate": 0.000317,
      "loss": 6.4712,
      "mean_token_accuracy": 0.13443209007382392,
      "num_tokens": 1428526.0,
      "step": 635
    },
    {
      "entropy": 6.634454774856567,
      "epoch": 0.6922660897782585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003195,
      "loss": 6.4175,
      "mean_token_accuracy": 0.13237082809209824,
      "num_tokens": 1438849.0,
      "step": 640
    },
    {
      "entropy": 6.572102212905884,
      "epoch": 0.6976744186046512,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000322,
      "loss": 6.4857,
      "mean_token_accuracy": 0.12762622609734536,
      "num_tokens": 1450776.0,
      "step": 645
    },
    {
      "entropy": 6.548601722717285,
      "epoch": 0.7030827474310438,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.3406,
      "mean_token_accuracy": 0.12978718280792237,
      "num_tokens": 1462552.0,
      "step": 650
    },
    {
      "entropy": 6.477000045776367,
      "epoch": 0.7084910762574365,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.3594,
      "mean_token_accuracy": 0.1336737334728241,
      "num_tokens": 1473867.0,
      "step": 655
    },
    {
      "entropy": 6.506887197494507,
      "epoch": 0.7138994050838291,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.3755,
      "mean_token_accuracy": 0.13364285305142404,
      "num_tokens": 1485233.0,
      "step": 660
    },
    {
      "entropy": 6.57369909286499,
      "epoch": 0.7193077339102217,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4229,
      "mean_token_accuracy": 0.1384587913751602,
      "num_tokens": 1496919.0,
      "step": 665
    },
    {
      "entropy": 6.569575452804566,
      "epoch": 0.7247160627366144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.3744,
      "mean_token_accuracy": 0.12899189442396164,
      "num_tokens": 1508695.0,
      "step": 670
    },
    {
      "entropy": 6.57951979637146,
      "epoch": 0.730124391563007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000337,
      "loss": 6.4236,
      "mean_token_accuracy": 0.13620161265134811,
      "num_tokens": 1519629.0,
      "step": 675
    },
    {
      "entropy": 6.527429914474487,
      "epoch": 0.7355327203893997,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003395,
      "loss": 6.3362,
      "mean_token_accuracy": 0.13338930681347846,
      "num_tokens": 1530144.0,
      "step": 680
    },
    {
      "entropy": 6.445765209197998,
      "epoch": 0.7409410492157923,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000342,
      "loss": 6.338,
      "mean_token_accuracy": 0.13429746255278588,
      "num_tokens": 1541013.0,
      "step": 685
    },
    {
      "entropy": 6.607530164718628,
      "epoch": 0.7463493780421849,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4094,
      "mean_token_accuracy": 0.13534171059727668,
      "num_tokens": 1553344.0,
      "step": 690
    },
    {
      "entropy": 6.519038200378418,
      "epoch": 0.7517577068685776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000347,
      "loss": 6.4489,
      "mean_token_accuracy": 0.13384861946105958,
      "num_tokens": 1565298.0,
      "step": 695
    },
    {
      "entropy": 6.506717586517334,
      "epoch": 0.7571660356949702,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003495,
      "loss": 6.3857,
      "mean_token_accuracy": 0.14263246059417725,
      "num_tokens": 1577645.0,
      "step": 700
    },
    {
      "entropy": 6.445629692077636,
      "epoch": 0.7625743645213628,
      "grad_norm": 1.203125,
      "learning_rate": 0.000352,
      "loss": 6.3658,
      "mean_token_accuracy": 0.13934621810913086,
      "num_tokens": 1588387.0,
      "step": 705
    },
    {
      "entropy": 6.639528369903564,
      "epoch": 0.7679826933477556,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003545,
      "loss": 6.4506,
      "mean_token_accuracy": 0.1386639840900898,
      "num_tokens": 1599837.0,
      "step": 710
    },
    {
      "entropy": 6.476478481292725,
      "epoch": 0.7733910221741482,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000357,
      "loss": 6.3896,
      "mean_token_accuracy": 0.12935182079672813,
      "num_tokens": 1610401.0,
      "step": 715
    },
    {
      "entropy": 6.4148335456848145,
      "epoch": 0.7787993510005409,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003595,
      "loss": 6.3446,
      "mean_token_accuracy": 0.13272809460759163,
      "num_tokens": 1622688.0,
      "step": 720
    },
    {
      "entropy": 6.624324560165405,
      "epoch": 0.7842076798269335,
      "grad_norm": 1.203125,
      "learning_rate": 0.000362,
      "loss": 6.4146,
      "mean_token_accuracy": 0.12909135445952416,
      "num_tokens": 1633594.0,
      "step": 725
    },
    {
      "entropy": 6.3749419212341305,
      "epoch": 0.7896160086533262,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003645,
      "loss": 6.3181,
      "mean_token_accuracy": 0.13193461969494819,
      "num_tokens": 1645861.0,
      "step": 730
    },
    {
      "entropy": 6.385309505462646,
      "epoch": 0.7950243374797188,
      "grad_norm": 1.21875,
      "learning_rate": 0.000367,
      "loss": 6.2236,
      "mean_token_accuracy": 0.14419274926185607,
      "num_tokens": 1656903.0,
      "step": 735
    },
    {
      "entropy": 6.447410774230957,
      "epoch": 0.8004326663061114,
      "grad_norm": 1.125,
      "learning_rate": 0.0003695,
      "loss": 6.3334,
      "mean_token_accuracy": 0.13758712783455848,
      "num_tokens": 1668130.0,
      "step": 740
    },
    {
      "entropy": 6.309524917602539,
      "epoch": 0.8058409951325041,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000372,
      "loss": 6.2358,
      "mean_token_accuracy": 0.14352481439709663,
      "num_tokens": 1678228.0,
      "step": 745
    },
    {
      "entropy": 6.534902763366699,
      "epoch": 0.8112493239588967,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745,
      "loss": 6.3822,
      "mean_token_accuracy": 0.13744921386241912,
      "num_tokens": 1690339.0,
      "step": 750
    },
    {
      "entropy": 6.490798282623291,
      "epoch": 0.8166576527852893,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000377,
      "loss": 6.3928,
      "mean_token_accuracy": 0.12897668555378913,
      "num_tokens": 1701067.0,
      "step": 755
    },
    {
      "entropy": 6.508419227600098,
      "epoch": 0.822065981611682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003795,
      "loss": 6.3318,
      "mean_token_accuracy": 0.14198972061276435,
      "num_tokens": 1712835.0,
      "step": 760
    },
    {
      "entropy": 6.4104822158813475,
      "epoch": 0.8274743104380746,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000382,
      "loss": 6.2808,
      "mean_token_accuracy": 0.13734794408082962,
      "num_tokens": 1723432.0,
      "step": 765
    },
    {
      "entropy": 6.467193269729615,
      "epoch": 0.8328826392644673,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003845,
      "loss": 6.3242,
      "mean_token_accuracy": 0.13243861123919487,
      "num_tokens": 1734134.0,
      "step": 770
    },
    {
      "entropy": 6.3534571647644045,
      "epoch": 0.8382909680908599,
      "grad_norm": 1.125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.2744,
      "mean_token_accuracy": 0.1389221429824829,
      "num_tokens": 1744723.0,
      "step": 775
    },
    {
      "entropy": 6.501026391983032,
      "epoch": 0.8436992969172525,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4111,
      "mean_token_accuracy": 0.13507046177983284,
      "num_tokens": 1755269.0,
      "step": 780
    },
    {
      "entropy": 6.521368360519409,
      "epoch": 0.8491076257436452,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4018,
      "mean_token_accuracy": 0.1358663931488991,
      "num_tokens": 1766344.0,
      "step": 785
    },
    {
      "entropy": 6.426554012298584,
      "epoch": 0.8545159545700378,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3736,
      "mean_token_accuracy": 0.12991632744669915,
      "num_tokens": 1779052.0,
      "step": 790
    },
    {
      "entropy": 6.335146522521972,
      "epoch": 0.8599242833964305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.2379,
      "mean_token_accuracy": 0.13654442951083184,
      "num_tokens": 1789754.0,
      "step": 795
    },
    {
      "entropy": 6.265465784072876,
      "epoch": 0.8653326122228231,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003995,
      "loss": 6.1662,
      "mean_token_accuracy": 0.14170663058757782,
      "num_tokens": 1801161.0,
      "step": 800
    },
    {
      "entropy": 6.439857339859008,
      "epoch": 0.8707409410492158,
      "grad_norm": 1.171875,
      "learning_rate": 0.000402,
      "loss": 6.2636,
      "mean_token_accuracy": 0.13717492818832397,
      "num_tokens": 1810892.0,
      "step": 805
    },
    {
      "entropy": 6.293186140060425,
      "epoch": 0.8761492698756085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045,
      "loss": 6.3348,
      "mean_token_accuracy": 0.134286767244339,
      "num_tokens": 1821574.0,
      "step": 810
    },
    {
      "entropy": 6.438879680633545,
      "epoch": 0.8815575987020011,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3697,
      "mean_token_accuracy": 0.13711444959044455,
      "num_tokens": 1833342.0,
      "step": 815
    },
    {
      "entropy": 6.382972478866577,
      "epoch": 0.8869659275283938,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004095,
      "loss": 6.2438,
      "mean_token_accuracy": 0.13821374326944352,
      "num_tokens": 1843768.0,
      "step": 820
    },
    {
      "entropy": 6.487565708160401,
      "epoch": 0.8923742563547864,
      "grad_norm": 1.078125,
      "learning_rate": 0.000412,
      "loss": 6.3156,
      "mean_token_accuracy": 0.13171282932162284,
      "num_tokens": 1855569.0,
      "step": 825
    },
    {
      "entropy": 6.3501959323883055,
      "epoch": 0.897782585181179,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145,
      "loss": 6.3051,
      "mean_token_accuracy": 0.13134763911366462,
      "num_tokens": 1866684.0,
      "step": 830
    },
    {
      "entropy": 6.362168693542481,
      "epoch": 0.9031909140075717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417,
      "loss": 6.2304,
      "mean_token_accuracy": 0.13782247975468637,
      "num_tokens": 1878700.0,
      "step": 835
    },
    {
      "entropy": 6.378588628768921,
      "epoch": 0.9085992428339643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004195,
      "loss": 6.2563,
      "mean_token_accuracy": 0.13868874236941336,
      "num_tokens": 1889970.0,
      "step": 840
    },
    {
      "entropy": 6.256662464141845,
      "epoch": 0.914007571660357,
      "grad_norm": 1.3125,
      "learning_rate": 0.000422,
      "loss": 6.1686,
      "mean_token_accuracy": 0.13953565880656243,
      "num_tokens": 1902008.0,
      "step": 845
    },
    {
      "entropy": 6.318142700195312,
      "epoch": 0.9194159004867496,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004245,
      "loss": 6.3913,
      "mean_token_accuracy": 0.13987977653741837,
      "num_tokens": 1913253.0,
      "step": 850
    },
    {
      "entropy": 6.398606395721435,
      "epoch": 0.9248242293131422,
      "grad_norm": 1.078125,
      "learning_rate": 0.000427,
      "loss": 6.242,
      "mean_token_accuracy": 0.14221547394990922,
      "num_tokens": 1923408.0,
      "step": 855
    },
    {
      "entropy": 6.3646495819091795,
      "epoch": 0.9302325581395349,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004295,
      "loss": 6.299,
      "mean_token_accuracy": 0.1426120236515999,
      "num_tokens": 1935315.0,
      "step": 860
    },
    {
      "entropy": 6.220848560333252,
      "epoch": 0.9356408869659275,
      "grad_norm": 1.1875,
      "learning_rate": 0.000432,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14092365950345992,
      "num_tokens": 1946524.0,
      "step": 865
    },
    {
      "entropy": 6.3639672756195065,
      "epoch": 0.9410492157923201,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004345,
      "loss": 6.2897,
      "mean_token_accuracy": 0.13832299113273622,
      "num_tokens": 1957561.0,
      "step": 870
    },
    {
      "entropy": 6.228976774215698,
      "epoch": 0.9464575446187128,
      "grad_norm": 1.125,
      "learning_rate": 0.000437,
      "loss": 6.2271,
      "mean_token_accuracy": 0.13985422402620315,
      "num_tokens": 1969089.0,
      "step": 875
    },
    {
      "entropy": 6.3985954284667965,
      "epoch": 0.9518658734451054,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004395,
      "loss": 6.2572,
      "mean_token_accuracy": 0.13929264023900031,
      "num_tokens": 1980341.0,
      "step": 880
    },
    {
      "entropy": 6.194135761260986,
      "epoch": 0.9572742022714981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.2152,
      "mean_token_accuracy": 0.1457873597741127,
      "num_tokens": 1991220.0,
      "step": 885
    },
    {
      "entropy": 6.379308748245239,
      "epoch": 0.9626825310978907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004445,
      "loss": 6.306,
      "mean_token_accuracy": 0.14043606221675872,
      "num_tokens": 2002650.0,
      "step": 890
    },
    {
      "entropy": 6.273330116271973,
      "epoch": 0.9680908599242833,
      "grad_norm": 1.03125,
      "learning_rate": 0.000447,
      "loss": 6.2088,
      "mean_token_accuracy": 0.14419417828321457,
      "num_tokens": 2012884.0,
      "step": 895
    },
    {
      "entropy": 6.3799642562866214,
      "epoch": 0.9734991887506761,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.2861,
      "mean_token_accuracy": 0.13549073413014412,
      "num_tokens": 2023543.0,
      "step": 900
    },
    {
      "entropy": 6.387584972381592,
      "epoch": 0.9789075175770687,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3488,
      "mean_token_accuracy": 0.13450675085186958,
      "num_tokens": 2034789.0,
      "step": 905
    },
    {
      "entropy": 6.239657592773438,
      "epoch": 0.9843158464034614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2558,
      "mean_token_accuracy": 0.1416017398238182,
      "num_tokens": 2046719.0,
      "step": 910
    },
    {
      "entropy": 6.252905702590942,
      "epoch": 0.989724175229854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.1984,
      "mean_token_accuracy": 0.14273628294467927,
      "num_tokens": 2057016.0,
      "step": 915
    },
    {
      "entropy": 6.232679033279419,
      "epoch": 0.9951325040562466,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.193,
      "mean_token_accuracy": 0.1355368375778198,
      "num_tokens": 2067898.0,
      "step": 920
    },
    {
      "entropy": 6.277052084604899,
      "epoch": 1.0,
      "grad_norm": 2.0625,
      "learning_rate": 0.000462,
      "loss": 6.2098,
      "mean_token_accuracy": 0.1415335453218884,
      "num_tokens": 2076889.0,
      "step": 925
    },
    {
      "entropy": 6.290318346023559,
      "epoch": 1.0054083288263926,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004645,
      "loss": 6.1099,
      "mean_token_accuracy": 0.13955417200922965,
      "num_tokens": 2089228.0,
      "step": 930
    },
    {
      "entropy": 6.165477848052978,
      "epoch": 1.0108166576527853,
      "grad_norm": 1.03125,
      "learning_rate": 0.000467,
      "loss": 6.1142,
      "mean_token_accuracy": 0.1436012014746666,
      "num_tokens": 2101098.0,
      "step": 935
    },
    {
      "entropy": 6.25781569480896,
      "epoch": 1.016224986479178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695,
      "loss": 6.0536,
      "mean_token_accuracy": 0.1513492174446583,
      "num_tokens": 2113081.0,
      "step": 940
    },
    {
      "entropy": 6.2349457263946535,
      "epoch": 1.0216333153055706,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000472,
      "loss": 6.0386,
      "mean_token_accuracy": 0.15058111846446992,
      "num_tokens": 2123356.0,
      "step": 945
    },
    {
      "entropy": 6.323075008392334,
      "epoch": 1.0270416441319632,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004745,
      "loss": 6.1321,
      "mean_token_accuracy": 0.14740860909223558,
      "num_tokens": 2134388.0,
      "step": 950
    },
    {
      "entropy": 6.072897148132324,
      "epoch": 1.0324499729583558,
      "grad_norm": 1.125,
      "learning_rate": 0.000477,
      "loss": 6.013,
      "mean_token_accuracy": 0.15088868290185928,
      "num_tokens": 2145209.0,
      "step": 955
    },
    {
      "entropy": 6.271716213226318,
      "epoch": 1.0378583017847485,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004795,
      "loss": 6.0733,
      "mean_token_accuracy": 0.15297747403383255,
      "num_tokens": 2157146.0,
      "step": 960
    },
    {
      "entropy": 6.113353109359741,
      "epoch": 1.043266630611141,
      "grad_norm": 1.046875,
      "learning_rate": 0.000482,
      "loss": 6.0832,
      "mean_token_accuracy": 0.14374596923589705,
      "num_tokens": 2168721.0,
      "step": 965
    },
    {
      "entropy": 6.233334302902222,
      "epoch": 1.0486749594375337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004845,
      "loss": 6.0774,
      "mean_token_accuracy": 0.14971460700035094,
      "num_tokens": 2179213.0,
      "step": 970
    },
    {
      "entropy": 6.177529621124267,
      "epoch": 1.0540832882639264,
      "grad_norm": 1.03125,
      "learning_rate": 0.000487,
      "loss": 6.1019,
      "mean_token_accuracy": 0.14426251277327537,
      "num_tokens": 2191224.0,
      "step": 975
    },
    {
      "entropy": 6.3130913257598875,
      "epoch": 1.059491617090319,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004895,
      "loss": 6.1694,
      "mean_token_accuracy": 0.14404836148023606,
      "num_tokens": 2203665.0,
      "step": 980
    },
    {
      "entropy": 6.18894772529602,
      "epoch": 1.0648999459167117,
      "grad_norm": 1.140625,
      "learning_rate": 0.000492,
      "loss": 6.0744,
      "mean_token_accuracy": 0.14550448656082154,
      "num_tokens": 2215136.0,
      "step": 985
    },
    {
      "entropy": 6.178538846969604,
      "epoch": 1.0703082747431043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004945,
      "loss": 6.1213,
      "mean_token_accuracy": 0.15047272816300392,
      "num_tokens": 2226375.0,
      "step": 990
    },
    {
      "entropy": 6.113742542266846,
      "epoch": 1.075716603569497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497,
      "loss": 5.967,
      "mean_token_accuracy": 0.15593916028738022,
      "num_tokens": 2236097.0,
      "step": 995
    },
    {
      "entropy": 6.344080114364624,
      "epoch": 1.0811249323958896,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995,
      "loss": 6.2159,
      "mean_token_accuracy": 0.14157222136855124,
      "num_tokens": 2247964.0,
      "step": 1000
    },
    {
      "epoch": 1.0811249323958896,
      "eval_entropy": 6.1291101856665176,
      "eval_loss": 6.394670009613037,
      "eval_mean_token_accuracy": 0.1441675969907506,
      "eval_num_tokens": 2247964.0,
      "eval_runtime": 2.5419,
      "eval_samples_per_second": 1382.851,
      "eval_steps_per_second": 173.102,
      "step": 1000
    }
  ],
  "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": 3510135917568000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}