{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06052719184093454,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691960716247559,
      "epoch": 0.0003026359592046727,
      "grad_norm": 13.5,
      "learning_rate": 2e-06,
      "loss": 10.8365,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10868.0,
      "step": 5
    },
    {
      "entropy": 10.69189395904541,
      "epoch": 0.0006052719184093454,
      "grad_norm": 11.75,
      "learning_rate": 4.5e-06,
      "loss": 10.7763,
      "mean_token_accuracy": 6.540222093462945e-05,
      "num_tokens": 22589.0,
      "step": 10
    },
    {
      "entropy": 10.6910231590271,
      "epoch": 0.0009079078776140181,
      "grad_norm": 9.5,
      "learning_rate": 7e-06,
      "loss": 10.5057,
      "mean_token_accuracy": 0.022316107933875173,
      "num_tokens": 32471.0,
      "step": 15
    },
    {
      "entropy": 10.678193473815918,
      "epoch": 0.0012105438368186909,
      "grad_norm": 6.09375,
      "learning_rate": 9.5e-06,
      "loss": 10.1537,
      "mean_token_accuracy": 0.044866344705224034,
      "num_tokens": 42231.0,
      "step": 20
    },
    {
      "entropy": 10.613947486877441,
      "epoch": 0.0015131797960233635,
      "grad_norm": 4.0,
      "learning_rate": 1.2e-05,
      "loss": 9.8674,
      "mean_token_accuracy": 0.04661851041018963,
      "num_tokens": 52966.0,
      "step": 25
    },
    {
      "entropy": 10.58952293395996,
      "epoch": 0.0018158157552280363,
      "grad_norm": 3.125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7212,
      "mean_token_accuracy": 0.047293811663985254,
      "num_tokens": 63711.0,
      "step": 30
    },
    {
      "entropy": 10.600228118896485,
      "epoch": 0.002118451714432709,
      "grad_norm": 2.703125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6714,
      "mean_token_accuracy": 0.03934945203363895,
      "num_tokens": 74046.0,
      "step": 35
    },
    {
      "entropy": 10.61097412109375,
      "epoch": 0.0024210876736373817,
      "grad_norm": 2.765625,
      "learning_rate": 1.95e-05,
      "loss": 9.5759,
      "mean_token_accuracy": 0.042743779718875885,
      "num_tokens": 84275.0,
      "step": 40
    },
    {
      "entropy": 10.5778733253479,
      "epoch": 0.0027237236328420543,
      "grad_norm": 2.46875,
      "learning_rate": 2.2e-05,
      "loss": 9.5666,
      "mean_token_accuracy": 0.04388550743460655,
      "num_tokens": 93302.0,
      "step": 45
    },
    {
      "entropy": 10.563697338104248,
      "epoch": 0.003026359592046727,
      "grad_norm": 2.546875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5216,
      "mean_token_accuracy": 0.0417548893019557,
      "num_tokens": 103414.0,
      "step": 50
    },
    {
      "entropy": 10.595334911346436,
      "epoch": 0.0033289955512513995,
      "grad_norm": 2.375,
      "learning_rate": 2.7e-05,
      "loss": 9.452,
      "mean_token_accuracy": 0.048485443368554114,
      "num_tokens": 113749.0,
      "step": 55
    },
    {
      "entropy": 10.54566011428833,
      "epoch": 0.0036316315104560726,
      "grad_norm": 2.3125,
      "learning_rate": 2.95e-05,
      "loss": 9.3969,
      "mean_token_accuracy": 0.05398624651134014,
      "num_tokens": 123922.0,
      "step": 60
    },
    {
      "entropy": 10.41702938079834,
      "epoch": 0.003934267469660745,
      "grad_norm": 2.5625,
      "learning_rate": 3.2e-05,
      "loss": 9.3086,
      "mean_token_accuracy": 0.05944342501461506,
      "num_tokens": 134720.0,
      "step": 65
    },
    {
      "entropy": 10.440918159484863,
      "epoch": 0.004236903428865418,
      "grad_norm": 2.484375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.1562,
      "mean_token_accuracy": 0.06198657713830471,
      "num_tokens": 145025.0,
      "step": 70
    },
    {
      "entropy": 10.4688570022583,
      "epoch": 0.00453953938807009,
      "grad_norm": 2.40625,
      "learning_rate": 3.7e-05,
      "loss": 9.0631,
      "mean_token_accuracy": 0.06703252233564853,
      "num_tokens": 155182.0,
      "step": 75
    },
    {
      "entropy": 10.363926982879638,
      "epoch": 0.004842175347274763,
      "grad_norm": 2.421875,
      "learning_rate": 3.95e-05,
      "loss": 9.0084,
      "mean_token_accuracy": 0.0727278657257557,
      "num_tokens": 166231.0,
      "step": 80
    },
    {
      "entropy": 10.344818210601806,
      "epoch": 0.005144811306479436,
      "grad_norm": 2.828125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.8422,
      "mean_token_accuracy": 0.07659510225057602,
      "num_tokens": 175799.0,
      "step": 85
    },
    {
      "entropy": 10.192126178741455,
      "epoch": 0.005447447265684109,
      "grad_norm": 2.75,
      "learning_rate": 4.45e-05,
      "loss": 8.7595,
      "mean_token_accuracy": 0.07438088804483414,
      "num_tokens": 185455.0,
      "step": 90
    },
    {
      "entropy": 10.227859497070312,
      "epoch": 0.005750083224888782,
      "grad_norm": 2.15625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.7451,
      "mean_token_accuracy": 0.072508579865098,
      "num_tokens": 196210.0,
      "step": 95
    },
    {
      "entropy": 10.18924322128296,
      "epoch": 0.006052719184093454,
      "grad_norm": 1.953125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.6534,
      "mean_token_accuracy": 0.072960215061903,
      "num_tokens": 206618.0,
      "step": 100
    },
    {
      "entropy": 10.125890731811523,
      "epoch": 0.006355355143298127,
      "grad_norm": 2.109375,
      "learning_rate": 5.2e-05,
      "loss": 8.5252,
      "mean_token_accuracy": 0.07648448273539543,
      "num_tokens": 217020.0,
      "step": 105
    },
    {
      "entropy": 10.093370056152343,
      "epoch": 0.006657991102502799,
      "grad_norm": 1.7421875,
      "learning_rate": 5.45e-05,
      "loss": 8.3649,
      "mean_token_accuracy": 0.07982658669352531,
      "num_tokens": 227195.0,
      "step": 110
    },
    {
      "entropy": 9.942719650268554,
      "epoch": 0.006960627061707472,
      "grad_norm": 2.046875,
      "learning_rate": 5.7e-05,
      "loss": 8.2903,
      "mean_token_accuracy": 0.07775866836309434,
      "num_tokens": 238473.0,
      "step": 115
    },
    {
      "entropy": 9.829155445098877,
      "epoch": 0.007263263020912145,
      "grad_norm": 1.8125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.0937,
      "mean_token_accuracy": 0.08428319320082664,
      "num_tokens": 247778.0,
      "step": 120
    },
    {
      "entropy": 9.750654411315917,
      "epoch": 0.007565898980116817,
      "grad_norm": 1.8125,
      "learning_rate": 6.2e-05,
      "loss": 8.1057,
      "mean_token_accuracy": 0.07914027571678162,
      "num_tokens": 258424.0,
      "step": 125
    },
    {
      "entropy": 9.584226131439209,
      "epoch": 0.00786853493932149,
      "grad_norm": 1.984375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9413,
      "mean_token_accuracy": 0.08204369693994522,
      "num_tokens": 267715.0,
      "step": 130
    },
    {
      "entropy": 9.357891654968261,
      "epoch": 0.008171170898526163,
      "grad_norm": 1.4765625,
      "learning_rate": 6.7e-05,
      "loss": 7.8035,
      "mean_token_accuracy": 0.08693078830838204,
      "num_tokens": 277127.0,
      "step": 135
    },
    {
      "entropy": 9.246243381500244,
      "epoch": 0.008473806857730836,
      "grad_norm": 1.6796875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.6874,
      "mean_token_accuracy": 0.08121695443987846,
      "num_tokens": 287275.0,
      "step": 140
    },
    {
      "entropy": 9.074360179901124,
      "epoch": 0.008776442816935508,
      "grad_norm": 1.2734375,
      "learning_rate": 7.2e-05,
      "loss": 7.577,
      "mean_token_accuracy": 0.08432774916291237,
      "num_tokens": 297131.0,
      "step": 145
    },
    {
      "entropy": 8.791367721557616,
      "epoch": 0.00907907877614018,
      "grad_norm": 1.890625,
      "learning_rate": 7.45e-05,
      "loss": 7.5053,
      "mean_token_accuracy": 0.0846397966146469,
      "num_tokens": 307649.0,
      "step": 150
    },
    {
      "entropy": 8.599281787872314,
      "epoch": 0.009381714735344854,
      "grad_norm": 1.3984375,
      "learning_rate": 7.7e-05,
      "loss": 7.3741,
      "mean_token_accuracy": 0.09144526571035386,
      "num_tokens": 317655.0,
      "step": 155
    },
    {
      "entropy": 8.408251190185547,
      "epoch": 0.009684350694549527,
      "grad_norm": 1.34375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.5165,
      "mean_token_accuracy": 0.0804292269051075,
      "num_tokens": 329286.0,
      "step": 160
    },
    {
      "entropy": 8.305608749389648,
      "epoch": 0.0099869866537542,
      "grad_norm": 1.328125,
      "learning_rate": 8.2e-05,
      "loss": 7.3511,
      "mean_token_accuracy": 0.08777917921543121,
      "num_tokens": 339212.0,
      "step": 165
    },
    {
      "entropy": 8.165600776672363,
      "epoch": 0.010289622612958871,
      "grad_norm": 1.265625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3448,
      "mean_token_accuracy": 0.084172061085701,
      "num_tokens": 349205.0,
      "step": 170
    },
    {
      "entropy": 8.088679838180543,
      "epoch": 0.010592258572163544,
      "grad_norm": 1.3203125,
      "learning_rate": 8.7e-05,
      "loss": 7.3862,
      "mean_token_accuracy": 0.09051570817828178,
      "num_tokens": 360548.0,
      "step": 175
    },
    {
      "entropy": 8.012139940261841,
      "epoch": 0.010894894531368217,
      "grad_norm": 1.2734375,
      "learning_rate": 8.95e-05,
      "loss": 7.2984,
      "mean_token_accuracy": 0.08498221635818481,
      "num_tokens": 371049.0,
      "step": 180
    },
    {
      "entropy": 7.907455492019653,
      "epoch": 0.01119753049057289,
      "grad_norm": 1.2578125,
      "learning_rate": 9.2e-05,
      "loss": 7.3813,
      "mean_token_accuracy": 0.08016184419393539,
      "num_tokens": 382334.0,
      "step": 185
    },
    {
      "entropy": 7.951280832290649,
      "epoch": 0.011500166449777563,
      "grad_norm": 1.28125,
      "learning_rate": 9.45e-05,
      "loss": 7.3108,
      "mean_token_accuracy": 0.08266079649329186,
      "num_tokens": 393169.0,
      "step": 190
    },
    {
      "entropy": 7.8983691215515135,
      "epoch": 0.011802802408982235,
      "grad_norm": 1.375,
      "learning_rate": 9.7e-05,
      "loss": 7.1965,
      "mean_token_accuracy": 0.0919338345527649,
      "num_tokens": 403826.0,
      "step": 195
    },
    {
      "entropy": 7.7634564399719235,
      "epoch": 0.012105438368186908,
      "grad_norm": 1.4609375,
      "learning_rate": 9.95e-05,
      "loss": 7.1829,
      "mean_token_accuracy": 0.09122484698891639,
      "num_tokens": 414231.0,
      "step": 200
    },
    {
      "entropy": 7.706010866165161,
      "epoch": 0.01240807432739158,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000102,
      "loss": 7.1705,
      "mean_token_accuracy": 0.09041819274425507,
      "num_tokens": 424532.0,
      "step": 205
    },
    {
      "entropy": 7.7909204959869385,
      "epoch": 0.012710710286596254,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.242,
      "mean_token_accuracy": 0.08969231508672237,
      "num_tokens": 435002.0,
      "step": 210
    },
    {
      "entropy": 7.730073738098144,
      "epoch": 0.013013346245800927,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000107,
      "loss": 7.1687,
      "mean_token_accuracy": 0.09499488323926926,
      "num_tokens": 444954.0,
      "step": 215
    },
    {
      "entropy": 7.597019338607788,
      "epoch": 0.013315982205005598,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001095,
      "loss": 7.1592,
      "mean_token_accuracy": 0.09385055750608444,
      "num_tokens": 455078.0,
      "step": 220
    },
    {
      "entropy": 7.616647529602051,
      "epoch": 0.013618618164210271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000112,
      "loss": 7.1523,
      "mean_token_accuracy": 0.08928861916065216,
      "num_tokens": 464688.0,
      "step": 225
    },
    {
      "entropy": 7.610339879989624,
      "epoch": 0.013921254123414944,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001145,
      "loss": 7.0821,
      "mean_token_accuracy": 0.0953272007405758,
      "num_tokens": 475365.0,
      "step": 230
    },
    {
      "entropy": 7.643180513381958,
      "epoch": 0.014223890082619617,
      "grad_norm": 1.65625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1841,
      "mean_token_accuracy": 0.08852923288941383,
      "num_tokens": 487238.0,
      "step": 235
    },
    {
      "entropy": 7.571054601669312,
      "epoch": 0.01452652604182429,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0531,
      "mean_token_accuracy": 0.09381508678197861,
      "num_tokens": 497485.0,
      "step": 240
    },
    {
      "entropy": 7.497907066345215,
      "epoch": 0.014829162001028962,
      "grad_norm": 1.390625,
      "learning_rate": 0.000122,
      "loss": 7.1308,
      "mean_token_accuracy": 0.09702477678656578,
      "num_tokens": 508228.0,
      "step": 245
    },
    {
      "entropy": 7.578845882415772,
      "epoch": 0.015131797960233635,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001245,
      "loss": 7.0308,
      "mean_token_accuracy": 0.09686913788318634,
      "num_tokens": 518645.0,
      "step": 250
    },
    {
      "entropy": 7.445795106887817,
      "epoch": 0.015434433919438308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000127,
      "loss": 7.0034,
      "mean_token_accuracy": 0.09973393902182578,
      "num_tokens": 528151.0,
      "step": 255
    },
    {
      "entropy": 7.537321949005127,
      "epoch": 0.01573706987864298,
      "grad_norm": 1.125,
      "learning_rate": 0.0001295,
      "loss": 7.0661,
      "mean_token_accuracy": 0.09172890931367875,
      "num_tokens": 539160.0,
      "step": 260
    },
    {
      "entropy": 7.48092999458313,
      "epoch": 0.016039705837847654,
      "grad_norm": 1.296875,
      "learning_rate": 0.000132,
      "loss": 7.1637,
      "mean_token_accuracy": 0.09009059146046638,
      "num_tokens": 550053.0,
      "step": 265
    },
    {
      "entropy": 7.453875637054443,
      "epoch": 0.016342341797052327,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9476,
      "mean_token_accuracy": 0.0969300739467144,
      "num_tokens": 560229.0,
      "step": 270
    },
    {
      "entropy": 7.485111570358276,
      "epoch": 0.016644977756257,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.0325,
      "mean_token_accuracy": 0.09286703020334244,
      "num_tokens": 570630.0,
      "step": 275
    },
    {
      "entropy": 7.580011320114136,
      "epoch": 0.016947613715461673,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001395,
      "loss": 7.0944,
      "mean_token_accuracy": 0.09208111613988876,
      "num_tokens": 581208.0,
      "step": 280
    },
    {
      "entropy": 7.438995409011841,
      "epoch": 0.017250249674666342,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.969,
      "mean_token_accuracy": 0.10797585695981979,
      "num_tokens": 591352.0,
      "step": 285
    },
    {
      "entropy": 7.403115940093994,
      "epoch": 0.017552885633871015,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001445,
      "loss": 7.0468,
      "mean_token_accuracy": 0.09680219516158103,
      "num_tokens": 600670.0,
      "step": 290
    },
    {
      "entropy": 7.429439353942871,
      "epoch": 0.01785552159307569,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000147,
      "loss": 7.0626,
      "mean_token_accuracy": 0.09904387444257737,
      "num_tokens": 610625.0,
      "step": 295
    },
    {
      "entropy": 7.462025833129883,
      "epoch": 0.01815815755228036,
      "grad_norm": 1.671875,
      "learning_rate": 0.0001495,
      "loss": 7.0209,
      "mean_token_accuracy": 0.09903855994343758,
      "num_tokens": 621096.0,
      "step": 300
    },
    {
      "entropy": 7.307780361175537,
      "epoch": 0.018460793511485035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000152,
      "loss": 6.9771,
      "mean_token_accuracy": 0.09788332656025886,
      "num_tokens": 632168.0,
      "step": 305
    },
    {
      "entropy": 7.444544649124145,
      "epoch": 0.018763429470689708,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.1017,
      "mean_token_accuracy": 0.09367784410715103,
      "num_tokens": 642277.0,
      "step": 310
    },
    {
      "entropy": 7.421035623550415,
      "epoch": 0.01906606542989438,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000157,
      "loss": 6.9797,
      "mean_token_accuracy": 0.09828720614314079,
      "num_tokens": 652429.0,
      "step": 315
    },
    {
      "entropy": 7.280771350860595,
      "epoch": 0.019368701389099054,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001595,
      "loss": 6.8647,
      "mean_token_accuracy": 0.09724124744534493,
      "num_tokens": 662805.0,
      "step": 320
    },
    {
      "entropy": 7.281081628799439,
      "epoch": 0.019671337348303727,
      "grad_norm": 1.3125,
      "learning_rate": 0.000162,
      "loss": 6.8711,
      "mean_token_accuracy": 0.10219163969159126,
      "num_tokens": 673015.0,
      "step": 325
    },
    {
      "entropy": 7.247241401672364,
      "epoch": 0.0199739733075084,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7959,
      "mean_token_accuracy": 0.09649019688367844,
      "num_tokens": 683357.0,
      "step": 330
    },
    {
      "entropy": 7.275740766525269,
      "epoch": 0.02027660926671307,
      "grad_norm": 1.1875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9325,
      "mean_token_accuracy": 0.09565900787711143,
      "num_tokens": 693310.0,
      "step": 335
    },
    {
      "entropy": 7.2511467933654785,
      "epoch": 0.020579245225917742,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.908,
      "mean_token_accuracy": 0.09649437665939331,
      "num_tokens": 703189.0,
      "step": 340
    },
    {
      "entropy": 7.20253620147705,
      "epoch": 0.020881881185122415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8688,
      "mean_token_accuracy": 0.0983615130186081,
      "num_tokens": 714359.0,
      "step": 345
    },
    {
      "entropy": 7.277911758422851,
      "epoch": 0.02118451714432709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9297,
      "mean_token_accuracy": 0.09489808529615402,
      "num_tokens": 724991.0,
      "step": 350
    },
    {
      "entropy": 7.230852937698364,
      "epoch": 0.02148715310353176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000177,
      "loss": 6.7761,
      "mean_token_accuracy": 0.09346158728003502,
      "num_tokens": 735879.0,
      "step": 355
    },
    {
      "entropy": 7.207417583465576,
      "epoch": 0.021789789062736434,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001795,
      "loss": 6.8486,
      "mean_token_accuracy": 0.10134509652853012,
      "num_tokens": 745887.0,
      "step": 360
    },
    {
      "entropy": 7.293970727920533,
      "epoch": 0.022092425021941108,
      "grad_norm": 1.3125,
      "learning_rate": 0.000182,
      "loss": 6.8689,
      "mean_token_accuracy": 0.10263044536113738,
      "num_tokens": 755822.0,
      "step": 365
    },
    {
      "entropy": 7.115125322341919,
      "epoch": 0.02239506098114578,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001845,
      "loss": 6.8332,
      "mean_token_accuracy": 0.10288432836532593,
      "num_tokens": 765330.0,
      "step": 370
    },
    {
      "entropy": 7.188446855545044,
      "epoch": 0.022697696940350454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000187,
      "loss": 6.7761,
      "mean_token_accuracy": 0.10406340062618255,
      "num_tokens": 775577.0,
      "step": 375
    },
    {
      "entropy": 7.159714698791504,
      "epoch": 0.023000332899555127,
      "grad_norm": 1.25,
      "learning_rate": 0.0001895,
      "loss": 6.8354,
      "mean_token_accuracy": 0.10109718143939972,
      "num_tokens": 784999.0,
      "step": 380
    },
    {
      "entropy": 7.176281404495239,
      "epoch": 0.023302968858759796,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000192,
      "loss": 6.7913,
      "mean_token_accuracy": 0.0978355996310711,
      "num_tokens": 794917.0,
      "step": 385
    },
    {
      "entropy": 7.06991753578186,
      "epoch": 0.02360560481796447,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001945,
      "loss": 6.7832,
      "mean_token_accuracy": 0.10112489834427833,
      "num_tokens": 805772.0,
      "step": 390
    },
    {
      "entropy": 7.272248983383179,
      "epoch": 0.023908240777169142,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.004,
      "mean_token_accuracy": 0.0911038339138031,
      "num_tokens": 816326.0,
      "step": 395
    },
    {
      "entropy": 7.135561180114746,
      "epoch": 0.024210876736373815,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7879,
      "mean_token_accuracy": 0.09880353137850761,
      "num_tokens": 825723.0,
      "step": 400
    },
    {
      "entropy": 7.084798860549927,
      "epoch": 0.02451351269557849,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000202,
      "loss": 6.7648,
      "mean_token_accuracy": 0.1035990685224533,
      "num_tokens": 834939.0,
      "step": 405
    },
    {
      "entropy": 7.045407962799072,
      "epoch": 0.02481614865478316,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7017,
      "mean_token_accuracy": 0.10783247500658036,
      "num_tokens": 844093.0,
      "step": 410
    },
    {
      "entropy": 7.1281506538391115,
      "epoch": 0.025118784613987834,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000207,
      "loss": 6.7839,
      "mean_token_accuracy": 0.10164490938186646,
      "num_tokens": 853879.0,
      "step": 415
    },
    {
      "entropy": 7.0675661087036135,
      "epoch": 0.025421420573192507,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002095,
      "loss": 6.7924,
      "mean_token_accuracy": 0.1014258049428463,
      "num_tokens": 864013.0,
      "step": 420
    },
    {
      "entropy": 7.088807916641235,
      "epoch": 0.02572405653239718,
      "grad_norm": 1.40625,
      "learning_rate": 0.000212,
      "loss": 6.7624,
      "mean_token_accuracy": 0.10403250828385353,
      "num_tokens": 874295.0,
      "step": 425
    },
    {
      "entropy": 7.082110023498535,
      "epoch": 0.026026692491601854,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002145,
      "loss": 6.8044,
      "mean_token_accuracy": 0.09894282519817352,
      "num_tokens": 884748.0,
      "step": 430
    },
    {
      "entropy": 7.119795227050782,
      "epoch": 0.026329328450806523,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8236,
      "mean_token_accuracy": 0.09981633573770524,
      "num_tokens": 895809.0,
      "step": 435
    },
    {
      "entropy": 7.022919845581055,
      "epoch": 0.026631964410011196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002195,
      "loss": 6.7971,
      "mean_token_accuracy": 0.1037395603954792,
      "num_tokens": 905252.0,
      "step": 440
    },
    {
      "entropy": 7.170399904251099,
      "epoch": 0.02693460036921587,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000222,
      "loss": 6.7273,
      "mean_token_accuracy": 0.10618472471833229,
      "num_tokens": 913876.0,
      "step": 445
    },
    {
      "entropy": 6.956621217727661,
      "epoch": 0.027237236328420542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002245,
      "loss": 6.7072,
      "mean_token_accuracy": 0.10306966602802277,
      "num_tokens": 924255.0,
      "step": 450
    },
    {
      "entropy": 7.0447276592254635,
      "epoch": 0.027539872287625215,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7086,
      "mean_token_accuracy": 0.1057076096534729,
      "num_tokens": 934799.0,
      "step": 455
    },
    {
      "entropy": 7.002421140670776,
      "epoch": 0.02784250824682989,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8042,
      "mean_token_accuracy": 0.10550158098340034,
      "num_tokens": 945757.0,
      "step": 460
    },
    {
      "entropy": 7.067310953140259,
      "epoch": 0.02814514420603456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.799,
      "mean_token_accuracy": 0.09717833548784256,
      "num_tokens": 956143.0,
      "step": 465
    },
    {
      "entropy": 7.0194652557373045,
      "epoch": 0.028447780165239234,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7151,
      "mean_token_accuracy": 0.10694214552640915,
      "num_tokens": 966437.0,
      "step": 470
    },
    {
      "entropy": 6.994701957702636,
      "epoch": 0.028750416124443907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000237,
      "loss": 6.7536,
      "mean_token_accuracy": 0.09864981994032859,
      "num_tokens": 977350.0,
      "step": 475
    },
    {
      "entropy": 7.053237247467041,
      "epoch": 0.02905305208364858,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0002395,
      "loss": 6.8339,
      "mean_token_accuracy": 0.10132090672850609,
      "num_tokens": 988340.0,
      "step": 480
    },
    {
      "entropy": 7.013916683197022,
      "epoch": 0.029355688042853254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000242,
      "loss": 6.6967,
      "mean_token_accuracy": 0.10258661732077598,
      "num_tokens": 998323.0,
      "step": 485
    },
    {
      "entropy": 6.890393209457398,
      "epoch": 0.029658324002057923,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002445,
      "loss": 6.5934,
      "mean_token_accuracy": 0.11049276739358901,
      "num_tokens": 1007731.0,
      "step": 490
    },
    {
      "entropy": 6.860795974731445,
      "epoch": 0.029960959961262596,
      "grad_norm": 1.359375,
      "learning_rate": 0.000247,
      "loss": 6.5649,
      "mean_token_accuracy": 0.11042860373854638,
      "num_tokens": 1017543.0,
      "step": 495
    },
    {
      "entropy": 6.819127559661865,
      "epoch": 0.03026359592046727,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002495,
      "loss": 6.6613,
      "mean_token_accuracy": 0.10852760300040246,
      "num_tokens": 1028558.0,
      "step": 500
    },
    {
      "entropy": 6.864386653900146,
      "epoch": 0.030566231879671942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000252,
      "loss": 6.6768,
      "mean_token_accuracy": 0.10810114666819573,
      "num_tokens": 1038904.0,
      "step": 505
    },
    {
      "entropy": 6.9760339736938475,
      "epoch": 0.030868867838876615,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002545,
      "loss": 6.695,
      "mean_token_accuracy": 0.10194359943270684,
      "num_tokens": 1049143.0,
      "step": 510
    },
    {
      "entropy": 6.881354522705078,
      "epoch": 0.03117150379808129,
      "grad_norm": 1.078125,
      "learning_rate": 0.000257,
      "loss": 6.6833,
      "mean_token_accuracy": 0.1067008651793003,
      "num_tokens": 1059914.0,
      "step": 515
    },
    {
      "entropy": 6.937298679351807,
      "epoch": 0.03147413975728596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002595,
      "loss": 6.7412,
      "mean_token_accuracy": 0.10159787982702255,
      "num_tokens": 1071007.0,
      "step": 520
    },
    {
      "entropy": 6.9824971675872805,
      "epoch": 0.03177677571649063,
      "grad_norm": 1.296875,
      "learning_rate": 0.000262,
      "loss": 6.7196,
      "mean_token_accuracy": 0.10421358048915863,
      "num_tokens": 1081983.0,
      "step": 525
    },
    {
      "entropy": 6.946443557739258,
      "epoch": 0.03207941167569531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7629,
      "mean_token_accuracy": 0.10414092987775803,
      "num_tokens": 1093200.0,
      "step": 530
    },
    {
      "entropy": 6.906464385986328,
      "epoch": 0.03238204763489998,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6426,
      "mean_token_accuracy": 0.1078506626188755,
      "num_tokens": 1103831.0,
      "step": 535
    },
    {
      "entropy": 6.812680625915528,
      "epoch": 0.03268468359410465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6863,
      "mean_token_accuracy": 0.10473831668496132,
      "num_tokens": 1114949.0,
      "step": 540
    },
    {
      "entropy": 6.873226833343506,
      "epoch": 0.03298731955330932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.5394,
      "mean_token_accuracy": 0.10754107609391213,
      "num_tokens": 1125079.0,
      "step": 545
    },
    {
      "entropy": 6.884418582916259,
      "epoch": 0.033289955512514,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002745,
      "loss": 6.6943,
      "mean_token_accuracy": 0.10482205227017402,
      "num_tokens": 1136097.0,
      "step": 550
    },
    {
      "entropy": 6.832566213607788,
      "epoch": 0.03359259147171867,
      "grad_norm": 1.390625,
      "learning_rate": 0.000277,
      "loss": 6.5888,
      "mean_token_accuracy": 0.10584642738103867,
      "num_tokens": 1146960.0,
      "step": 555
    },
    {
      "entropy": 6.708227252960205,
      "epoch": 0.033895227430923346,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002795,
      "loss": 6.622,
      "mean_token_accuracy": 0.1061646118760109,
      "num_tokens": 1156752.0,
      "step": 560
    },
    {
      "entropy": 6.7955772399902346,
      "epoch": 0.034197863390128015,
      "grad_norm": 1.25,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5806,
      "mean_token_accuracy": 0.10415633767843246,
      "num_tokens": 1166831.0,
      "step": 565
    },
    {
      "entropy": 6.92924394607544,
      "epoch": 0.034500499349332685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002845,
      "loss": 6.665,
      "mean_token_accuracy": 0.11003877967596054,
      "num_tokens": 1178318.0,
      "step": 570
    },
    {
      "entropy": 6.801376008987427,
      "epoch": 0.03480313530853736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000287,
      "loss": 6.6321,
      "mean_token_accuracy": 0.10529007613658906,
      "num_tokens": 1187822.0,
      "step": 575
    },
    {
      "entropy": 6.865694141387939,
      "epoch": 0.03510577126774203,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002895,
      "loss": 6.653,
      "mean_token_accuracy": 0.1032220371067524,
      "num_tokens": 1198243.0,
      "step": 580
    },
    {
      "entropy": 6.825288724899292,
      "epoch": 0.03540840722694671,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000292,
      "loss": 6.5809,
      "mean_token_accuracy": 0.11235557571053505,
      "num_tokens": 1208688.0,
      "step": 585
    },
    {
      "entropy": 6.768784046173096,
      "epoch": 0.03571104318615138,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002945,
      "loss": 6.6436,
      "mean_token_accuracy": 0.10551055148243904,
      "num_tokens": 1218928.0,
      "step": 590
    },
    {
      "entropy": 6.84692416191101,
      "epoch": 0.03601367914535605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000297,
      "loss": 6.718,
      "mean_token_accuracy": 0.10656947270035744,
      "num_tokens": 1229582.0,
      "step": 595
    },
    {
      "entropy": 6.706984806060791,
      "epoch": 0.03631631510456072,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002995,
      "loss": 6.5104,
      "mean_token_accuracy": 0.10828106105327606,
      "num_tokens": 1239887.0,
      "step": 600
    },
    {
      "entropy": 6.820476007461548,
      "epoch": 0.0366189510637654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000302,
      "loss": 6.6014,
      "mean_token_accuracy": 0.11197339668869973,
      "num_tokens": 1249278.0,
      "step": 605
    },
    {
      "entropy": 6.728797483444214,
      "epoch": 0.03692158702297007,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003045,
      "loss": 6.5345,
      "mean_token_accuracy": 0.11249430924654007,
      "num_tokens": 1259931.0,
      "step": 610
    },
    {
      "entropy": 6.744882392883301,
      "epoch": 0.03722422298217474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000307,
      "loss": 6.5856,
      "mean_token_accuracy": 0.10611726269125939,
      "num_tokens": 1269299.0,
      "step": 615
    },
    {
      "entropy": 6.845881986618042,
      "epoch": 0.037526858941379415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003095,
      "loss": 6.5987,
      "mean_token_accuracy": 0.11280884444713593,
      "num_tokens": 1279662.0,
      "step": 620
    },
    {
      "entropy": 6.697132110595703,
      "epoch": 0.037829494900584085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000312,
      "loss": 6.6368,
      "mean_token_accuracy": 0.1129900798201561,
      "num_tokens": 1289513.0,
      "step": 625
    },
    {
      "entropy": 6.748120880126953,
      "epoch": 0.03813213085978876,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003145,
      "loss": 6.4993,
      "mean_token_accuracy": 0.11641889214515685,
      "num_tokens": 1299361.0,
      "step": 630
    },
    {
      "entropy": 6.6863518238067625,
      "epoch": 0.03843476681899343,
      "grad_norm": 1.140625,
      "learning_rate": 0.000317,
      "loss": 6.5372,
      "mean_token_accuracy": 0.11358237043023109,
      "num_tokens": 1310263.0,
      "step": 635
    },
    {
      "entropy": 6.787765169143677,
      "epoch": 0.03873740277819811,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195,
      "loss": 6.6145,
      "mean_token_accuracy": 0.10874337628483773,
      "num_tokens": 1320120.0,
      "step": 640
    },
    {
      "entropy": 6.77995982170105,
      "epoch": 0.03904003873740278,
      "grad_norm": 1.3125,
      "learning_rate": 0.000322,
      "loss": 6.7475,
      "mean_token_accuracy": 0.10635755434632302,
      "num_tokens": 1330575.0,
      "step": 645
    },
    {
      "entropy": 6.809815788269043,
      "epoch": 0.03934267469660745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6066,
      "mean_token_accuracy": 0.11191309690475464,
      "num_tokens": 1340416.0,
      "step": 650
    },
    {
      "entropy": 6.645190000534058,
      "epoch": 0.03964531065581212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4977,
      "mean_token_accuracy": 0.11090238913893699,
      "num_tokens": 1350263.0,
      "step": 655
    },
    {
      "entropy": 6.763588619232178,
      "epoch": 0.0399479466150168,
      "grad_norm": 1.03125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.595,
      "mean_token_accuracy": 0.10866745784878731,
      "num_tokens": 1360455.0,
      "step": 660
    },
    {
      "entropy": 6.647847270965576,
      "epoch": 0.04025058257422147,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5816,
      "mean_token_accuracy": 0.11250133812427521,
      "num_tokens": 1370250.0,
      "step": 665
    },
    {
      "entropy": 6.729582405090332,
      "epoch": 0.04055321853342614,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.533,
      "mean_token_accuracy": 0.11229272112250328,
      "num_tokens": 1380625.0,
      "step": 670
    },
    {
      "entropy": 6.658378219604492,
      "epoch": 0.040855854492630815,
      "grad_norm": 0.96875,
      "learning_rate": 0.000337,
      "loss": 6.5428,
      "mean_token_accuracy": 0.11572029441595078,
      "num_tokens": 1391799.0,
      "step": 675
    },
    {
      "entropy": 6.686212539672852,
      "epoch": 0.041158490451835485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003395,
      "loss": 6.4679,
      "mean_token_accuracy": 0.11066331416368484,
      "num_tokens": 1402386.0,
      "step": 680
    },
    {
      "entropy": 6.572569131851196,
      "epoch": 0.04146112641104016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000342,
      "loss": 6.4795,
      "mean_token_accuracy": 0.11622468978166581,
      "num_tokens": 1413076.0,
      "step": 685
    },
    {
      "entropy": 6.615976428985595,
      "epoch": 0.04176376237024483,
      "grad_norm": 1.125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4522,
      "mean_token_accuracy": 0.10912851095199586,
      "num_tokens": 1423406.0,
      "step": 690
    },
    {
      "entropy": 6.554074764251709,
      "epoch": 0.04206639832944951,
      "grad_norm": 1.171875,
      "learning_rate": 0.000347,
      "loss": 6.4843,
      "mean_token_accuracy": 0.11364608034491538,
      "num_tokens": 1434166.0,
      "step": 695
    },
    {
      "entropy": 6.553112173080445,
      "epoch": 0.04236903428865418,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003495,
      "loss": 6.4252,
      "mean_token_accuracy": 0.11718087047338485,
      "num_tokens": 1444492.0,
      "step": 700
    },
    {
      "entropy": 6.63217945098877,
      "epoch": 0.04267167024785885,
      "grad_norm": 1.15625,
      "learning_rate": 0.000352,
      "loss": 6.4765,
      "mean_token_accuracy": 0.11712096258997917,
      "num_tokens": 1454499.0,
      "step": 705
    },
    {
      "entropy": 6.707035875320434,
      "epoch": 0.04297430620706352,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003545,
      "loss": 6.4968,
      "mean_token_accuracy": 0.11697999909520149,
      "num_tokens": 1464854.0,
      "step": 710
    },
    {
      "entropy": 6.561078643798828,
      "epoch": 0.0432769421662682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000357,
      "loss": 6.5045,
      "mean_token_accuracy": 0.11402541399002075,
      "num_tokens": 1475902.0,
      "step": 715
    },
    {
      "entropy": 6.671833848953247,
      "epoch": 0.04357957812547287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003595,
      "loss": 6.5164,
      "mean_token_accuracy": 0.11434905081987382,
      "num_tokens": 1486215.0,
      "step": 720
    },
    {
      "entropy": 6.504771089553833,
      "epoch": 0.04388221408467754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000362,
      "loss": 6.3736,
      "mean_token_accuracy": 0.11950063407421112,
      "num_tokens": 1495981.0,
      "step": 725
    },
    {
      "entropy": 6.6006206512451175,
      "epoch": 0.044184850043882215,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003645,
      "loss": 6.4802,
      "mean_token_accuracy": 0.11169682219624519,
      "num_tokens": 1506298.0,
      "step": 730
    },
    {
      "entropy": 6.528478288650513,
      "epoch": 0.044487486003086885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000367,
      "loss": 6.4969,
      "mean_token_accuracy": 0.11755755618214607,
      "num_tokens": 1516032.0,
      "step": 735
    },
    {
      "entropy": 6.648074007034301,
      "epoch": 0.04479012196229156,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003695,
      "loss": 6.5782,
      "mean_token_accuracy": 0.10931968688964844,
      "num_tokens": 1527335.0,
      "step": 740
    },
    {
      "entropy": 6.554501485824585,
      "epoch": 0.04509275792149623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000372,
      "loss": 6.4557,
      "mean_token_accuracy": 0.1119515359401703,
      "num_tokens": 1538186.0,
      "step": 745
    },
    {
      "entropy": 6.690221738815308,
      "epoch": 0.04539539388070091,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003745,
      "loss": 6.5974,
      "mean_token_accuracy": 0.11424762234091759,
      "num_tokens": 1548197.0,
      "step": 750
    },
    {
      "entropy": 6.628253650665283,
      "epoch": 0.04569802983990558,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000377,
      "loss": 6.5654,
      "mean_token_accuracy": 0.11403838992118835,
      "num_tokens": 1557623.0,
      "step": 755
    },
    {
      "entropy": 6.608108568191528,
      "epoch": 0.04600066579911025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003795,
      "loss": 6.4858,
      "mean_token_accuracy": 0.11169281452894211,
      "num_tokens": 1568205.0,
      "step": 760
    },
    {
      "entropy": 6.592059135437012,
      "epoch": 0.04630330175831492,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000382,
      "loss": 6.5087,
      "mean_token_accuracy": 0.1070365622639656,
      "num_tokens": 1578558.0,
      "step": 765
    },
    {
      "entropy": 6.5310780048370365,
      "epoch": 0.04660593771751959,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003845,
      "loss": 6.4282,
      "mean_token_accuracy": 0.11913914307951927,
      "num_tokens": 1589240.0,
      "step": 770
    },
    {
      "entropy": 6.568341159820557,
      "epoch": 0.04690857367672427,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4832,
      "mean_token_accuracy": 0.11464685276150703,
      "num_tokens": 1600318.0,
      "step": 775
    },
    {
      "entropy": 6.519914770126343,
      "epoch": 0.04721120963592894,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5062,
      "mean_token_accuracy": 0.11620973721146584,
      "num_tokens": 1610606.0,
      "step": 780
    },
    {
      "entropy": 6.740098190307617,
      "epoch": 0.047513845595133615,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.5844,
      "mean_token_accuracy": 0.11021628305315971,
      "num_tokens": 1621517.0,
      "step": 785
    },
    {
      "entropy": 6.446154308319092,
      "epoch": 0.047816481554338285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5337,
      "mean_token_accuracy": 0.11654144525527954,
      "num_tokens": 1631929.0,
      "step": 790
    },
    {
      "entropy": 6.65470027923584,
      "epoch": 0.04811911751354296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4581,
      "mean_token_accuracy": 0.11459191888570786,
      "num_tokens": 1642393.0,
      "step": 795
    },
    {
      "entropy": 6.5919910907745365,
      "epoch": 0.04842175347274763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003995,
      "loss": 6.5776,
      "mean_token_accuracy": 0.10466790720820426,
      "num_tokens": 1653210.0,
      "step": 800
    },
    {
      "entropy": 6.630907821655273,
      "epoch": 0.04872438943195231,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000402,
      "loss": 6.4478,
      "mean_token_accuracy": 0.11859480291604996,
      "num_tokens": 1662984.0,
      "step": 805
    },
    {
      "entropy": 6.475538063049316,
      "epoch": 0.04902702539115698,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004045,
      "loss": 6.4048,
      "mean_token_accuracy": 0.11649263054132461,
      "num_tokens": 1672976.0,
      "step": 810
    },
    {
      "entropy": 6.654109048843384,
      "epoch": 0.04932966135036165,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.525,
      "mean_token_accuracy": 0.11437419354915619,
      "num_tokens": 1682799.0,
      "step": 815
    },
    {
      "entropy": 6.49087176322937,
      "epoch": 0.04963229730956632,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095,
      "loss": 6.3958,
      "mean_token_accuracy": 0.11366481408476829,
      "num_tokens": 1692288.0,
      "step": 820
    },
    {
      "entropy": 6.584628534317017,
      "epoch": 0.04993493326877099,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000412,
      "loss": 6.4682,
      "mean_token_accuracy": 0.10887323394417762,
      "num_tokens": 1703602.0,
      "step": 825
    },
    {
      "entropy": 6.487986660003662,
      "epoch": 0.05023756922797567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004145,
      "loss": 6.471,
      "mean_token_accuracy": 0.11930727735161781,
      "num_tokens": 1713413.0,
      "step": 830
    },
    {
      "entropy": 6.4606489658355715,
      "epoch": 0.05054020518718034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000417,
      "loss": 6.5096,
      "mean_token_accuracy": 0.11392919346690178,
      "num_tokens": 1724142.0,
      "step": 835
    },
    {
      "entropy": 6.569206428527832,
      "epoch": 0.050842841146385015,
      "grad_norm": 1.0,
      "learning_rate": 0.0004195,
      "loss": 6.5431,
      "mean_token_accuracy": 0.11335351839661598,
      "num_tokens": 1734202.0,
      "step": 840
    },
    {
      "entropy": 6.596271944046021,
      "epoch": 0.051145477105589685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000422,
      "loss": 6.4941,
      "mean_token_accuracy": 0.11573994532227516,
      "num_tokens": 1744736.0,
      "step": 845
    },
    {
      "entropy": 6.52998948097229,
      "epoch": 0.05144811306479436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004245,
      "loss": 6.3848,
      "mean_token_accuracy": 0.11664405986666679,
      "num_tokens": 1754142.0,
      "step": 850
    },
    {
      "entropy": 6.4787373542785645,
      "epoch": 0.05175074902399903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000427,
      "loss": 6.405,
      "mean_token_accuracy": 0.1206323929131031,
      "num_tokens": 1764045.0,
      "step": 855
    },
    {
      "entropy": 6.459346961975098,
      "epoch": 0.05205338498320371,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004295,
      "loss": 6.4446,
      "mean_token_accuracy": 0.11291947215795517,
      "num_tokens": 1774580.0,
      "step": 860
    },
    {
      "entropy": 6.44007978439331,
      "epoch": 0.05235602094240838,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000432,
      "loss": 6.4146,
      "mean_token_accuracy": 0.11788542792201043,
      "num_tokens": 1785741.0,
      "step": 865
    },
    {
      "entropy": 6.52412166595459,
      "epoch": 0.052658656901613046,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004345,
      "loss": 6.4581,
      "mean_token_accuracy": 0.11375688314437866,
      "num_tokens": 1796131.0,
      "step": 870
    },
    {
      "entropy": 6.511375808715821,
      "epoch": 0.05296129286081772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000437,
      "loss": 6.4351,
      "mean_token_accuracy": 0.11737742349505424,
      "num_tokens": 1807404.0,
      "step": 875
    },
    {
      "entropy": 6.540281772613525,
      "epoch": 0.05326392882002239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004395,
      "loss": 6.4708,
      "mean_token_accuracy": 0.11675170511007309,
      "num_tokens": 1817579.0,
      "step": 880
    },
    {
      "entropy": 6.431980800628662,
      "epoch": 0.05356656477922707,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000442,
      "loss": 6.3703,
      "mean_token_accuracy": 0.12299558818340302,
      "num_tokens": 1827075.0,
      "step": 885
    },
    {
      "entropy": 6.482553005218506,
      "epoch": 0.05386920073843174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004445,
      "loss": 6.5018,
      "mean_token_accuracy": 0.1127280443906784,
      "num_tokens": 1837556.0,
      "step": 890
    },
    {
      "entropy": 6.577256059646606,
      "epoch": 0.054171836697636415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000447,
      "loss": 6.4601,
      "mean_token_accuracy": 0.11809274181723595,
      "num_tokens": 1847939.0,
      "step": 895
    },
    {
      "entropy": 6.465627002716064,
      "epoch": 0.054474472656841084,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4547,
      "mean_token_accuracy": 0.11632255762815476,
      "num_tokens": 1857837.0,
      "step": 900
    },
    {
      "entropy": 6.447572135925293,
      "epoch": 0.05477710861604576,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.367,
      "mean_token_accuracy": 0.12046080604195594,
      "num_tokens": 1868341.0,
      "step": 905
    },
    {
      "entropy": 6.4407977104187015,
      "epoch": 0.05507974457525043,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4597,
      "mean_token_accuracy": 0.11877358108758926,
      "num_tokens": 1879798.0,
      "step": 910
    },
    {
      "entropy": 6.431598711013794,
      "epoch": 0.05538238053445511,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4364,
      "mean_token_accuracy": 0.12093160301446915,
      "num_tokens": 1890685.0,
      "step": 915
    },
    {
      "entropy": 6.472229862213135,
      "epoch": 0.05568501649365978,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4066,
      "mean_token_accuracy": 0.11710434779524803,
      "num_tokens": 1900870.0,
      "step": 920
    },
    {
      "entropy": 6.36571102142334,
      "epoch": 0.055987652452864446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000462,
      "loss": 6.427,
      "mean_token_accuracy": 0.11282309144735336,
      "num_tokens": 1911562.0,
      "step": 925
    },
    {
      "entropy": 6.390591049194336,
      "epoch": 0.05629028841206912,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004645,
      "loss": 6.4,
      "mean_token_accuracy": 0.11724207326769828,
      "num_tokens": 1922302.0,
      "step": 930
    },
    {
      "entropy": 6.520695447921753,
      "epoch": 0.05659292437127379,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467,
      "loss": 6.4615,
      "mean_token_accuracy": 0.11738158836960792,
      "num_tokens": 1933022.0,
      "step": 935
    },
    {
      "entropy": 6.472976922988892,
      "epoch": 0.05689556033047847,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695,
      "loss": 6.5221,
      "mean_token_accuracy": 0.1196384847164154,
      "num_tokens": 1943630.0,
      "step": 940
    },
    {
      "entropy": 6.467438507080078,
      "epoch": 0.05719819628968314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000472,
      "loss": 6.4745,
      "mean_token_accuracy": 0.11089805960655212,
      "num_tokens": 1954665.0,
      "step": 945
    },
    {
      "entropy": 6.467648077011108,
      "epoch": 0.057500832248887815,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004745,
      "loss": 6.5119,
      "mean_token_accuracy": 0.11348704844713212,
      "num_tokens": 1965383.0,
      "step": 950
    },
    {
      "entropy": 6.465866374969482,
      "epoch": 0.057803468208092484,
      "grad_norm": 1.15625,
      "learning_rate": 0.000477,
      "loss": 6.419,
      "mean_token_accuracy": 0.11720119565725326,
      "num_tokens": 1975087.0,
      "step": 955
    },
    {
      "entropy": 6.458328771591186,
      "epoch": 0.05810610416729716,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004795,
      "loss": 6.4219,
      "mean_token_accuracy": 0.12088532075285911,
      "num_tokens": 1985507.0,
      "step": 960
    },
    {
      "entropy": 6.4001624584198,
      "epoch": 0.05840874012650183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482,
      "loss": 6.3625,
      "mean_token_accuracy": 0.11903072446584702,
      "num_tokens": 1997943.0,
      "step": 965
    },
    {
      "entropy": 6.343327522277832,
      "epoch": 0.05871137608570651,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004845,
      "loss": 6.3109,
      "mean_token_accuracy": 0.12039421126246452,
      "num_tokens": 2008472.0,
      "step": 970
    },
    {
      "entropy": 6.433587121963501,
      "epoch": 0.05901401204491118,
      "grad_norm": 1.140625,
      "learning_rate": 0.000487,
      "loss": 6.4137,
      "mean_token_accuracy": 0.12280701920390129,
      "num_tokens": 2018448.0,
      "step": 975
    },
    {
      "entropy": 6.4722997665405275,
      "epoch": 0.059316648004115846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 6.4388,
      "mean_token_accuracy": 0.11973628625273705,
      "num_tokens": 2029030.0,
      "step": 980
    },
    {
      "entropy": 6.365376329421997,
      "epoch": 0.05961928396332052,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000492,
      "loss": 6.3981,
      "mean_token_accuracy": 0.11999995708465576,
      "num_tokens": 2040111.0,
      "step": 985
    },
    {
      "entropy": 6.450404977798462,
      "epoch": 0.05992191992252519,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004945,
      "loss": 6.4181,
      "mean_token_accuracy": 0.11250131353735923,
      "num_tokens": 2050392.0,
      "step": 990
    },
    {
      "entropy": 6.408387279510498,
      "epoch": 0.06022455588172987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497,
      "loss": 6.4163,
      "mean_token_accuracy": 0.12313750237226487,
      "num_tokens": 2060915.0,
      "step": 995
    },
    {
      "entropy": 6.382927274703979,
      "epoch": 0.06052719184093454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995,
      "loss": 6.4762,
      "mean_token_accuracy": 0.10732885748147965,
      "num_tokens": 2072710.0,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3019507052544000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}