{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.055955012170215146,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691998958587646,
      "epoch": 0.0002797750608510757,
      "grad_norm": 15.75,
      "learning_rate": 2e-06,
      "loss": 10.7863,
      "mean_token_accuracy": 0.0001396648003719747,
      "num_tokens": 8714.0,
      "step": 5
    },
    {
      "entropy": 10.691917991638183,
      "epoch": 0.0005595501217021514,
      "grad_norm": 12.6875,
      "learning_rate": 4.5e-06,
      "loss": 10.7427,
      "mean_token_accuracy": 6.393861840479076e-05,
      "num_tokens": 18630.0,
      "step": 10
    },
    {
      "entropy": 10.690977668762207,
      "epoch": 0.0008393251825532272,
      "grad_norm": 9.8125,
      "learning_rate": 7e-06,
      "loss": 10.4865,
      "mean_token_accuracy": 0.026551677379757166,
      "num_tokens": 28134.0,
      "step": 15
    },
    {
      "entropy": 10.678603363037109,
      "epoch": 0.0011191002434043028,
      "grad_norm": 6.4375,
      "learning_rate": 9.5e-06,
      "loss": 10.188,
      "mean_token_accuracy": 0.04146423395723105,
      "num_tokens": 37326.0,
      "step": 20
    },
    {
      "entropy": 10.619599628448487,
      "epoch": 0.0013988753042553786,
      "grad_norm": 4.15625,
      "learning_rate": 1.2e-05,
      "loss": 9.8709,
      "mean_token_accuracy": 0.04364684820175171,
      "num_tokens": 45813.0,
      "step": 25
    },
    {
      "entropy": 10.597146987915039,
      "epoch": 0.0016786503651064545,
      "grad_norm": 3.265625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7719,
      "mean_token_accuracy": 0.03872578963637352,
      "num_tokens": 54746.0,
      "step": 30
    },
    {
      "entropy": 10.614226627349854,
      "epoch": 0.0019584254259575303,
      "grad_norm": 2.734375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6616,
      "mean_token_accuracy": 0.04101139903068542,
      "num_tokens": 64386.0,
      "step": 35
    },
    {
      "entropy": 10.594823837280273,
      "epoch": 0.0022382004868086057,
      "grad_norm": 2.71875,
      "learning_rate": 1.95e-05,
      "loss": 9.6762,
      "mean_token_accuracy": 0.03911035992205143,
      "num_tokens": 73648.0,
      "step": 40
    },
    {
      "entropy": 10.59369068145752,
      "epoch": 0.0025179755476596815,
      "grad_norm": 2.546875,
      "learning_rate": 2.2e-05,
      "loss": 9.5708,
      "mean_token_accuracy": 0.04052439667284489,
      "num_tokens": 82108.0,
      "step": 45
    },
    {
      "entropy": 10.583232975006103,
      "epoch": 0.0027977506085107573,
      "grad_norm": 2.546875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5017,
      "mean_token_accuracy": 0.047141891345381735,
      "num_tokens": 92109.0,
      "step": 50
    },
    {
      "entropy": 10.565110969543458,
      "epoch": 0.003077525669361833,
      "grad_norm": 2.703125,
      "learning_rate": 2.7e-05,
      "loss": 9.4323,
      "mean_token_accuracy": 0.052879580110311505,
      "num_tokens": 101365.0,
      "step": 55
    },
    {
      "entropy": 10.425063991546631,
      "epoch": 0.003357300730212909,
      "grad_norm": 2.546875,
      "learning_rate": 2.95e-05,
      "loss": 9.3995,
      "mean_token_accuracy": 0.05232988931238651,
      "num_tokens": 110432.0,
      "step": 60
    },
    {
      "entropy": 10.407742786407471,
      "epoch": 0.0036370757910639847,
      "grad_norm": 2.34375,
      "learning_rate": 3.2e-05,
      "loss": 9.3122,
      "mean_token_accuracy": 0.051964789256453515,
      "num_tokens": 120191.0,
      "step": 65
    },
    {
      "entropy": 10.429958152770997,
      "epoch": 0.0039168508519150606,
      "grad_norm": 2.3125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.1579,
      "mean_token_accuracy": 0.06370341181755065,
      "num_tokens": 129580.0,
      "step": 70
    },
    {
      "entropy": 10.423612117767334,
      "epoch": 0.004196625912766136,
      "grad_norm": 2.375,
      "learning_rate": 3.7e-05,
      "loss": 9.1589,
      "mean_token_accuracy": 0.05433790199458599,
      "num_tokens": 139083.0,
      "step": 75
    },
    {
      "entropy": 10.42996587753296,
      "epoch": 0.004476400973617211,
      "grad_norm": 2.359375,
      "learning_rate": 3.95e-05,
      "loss": 8.9628,
      "mean_token_accuracy": 0.05919392295181751,
      "num_tokens": 148183.0,
      "step": 80
    },
    {
      "entropy": 10.377639865875244,
      "epoch": 0.004756176034468288,
      "grad_norm": 2.359375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.8812,
      "mean_token_accuracy": 0.060510285571217536,
      "num_tokens": 157015.0,
      "step": 85
    },
    {
      "entropy": 10.363342475891113,
      "epoch": 0.005035951095319363,
      "grad_norm": 2.921875,
      "learning_rate": 4.45e-05,
      "loss": 8.8142,
      "mean_token_accuracy": 0.06536102406680584,
      "num_tokens": 166314.0,
      "step": 90
    },
    {
      "entropy": 10.275917053222656,
      "epoch": 0.005315726156170439,
      "grad_norm": 2.203125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6269,
      "mean_token_accuracy": 0.07039711400866508,
      "num_tokens": 175975.0,
      "step": 95
    },
    {
      "entropy": 10.235202980041503,
      "epoch": 0.005595501217021515,
      "grad_norm": 2.734375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.574,
      "mean_token_accuracy": 0.06645569279789924,
      "num_tokens": 185976.0,
      "step": 100
    },
    {
      "entropy": 10.220190620422363,
      "epoch": 0.005875276277872591,
      "grad_norm": 2.171875,
      "learning_rate": 5.2e-05,
      "loss": 8.4802,
      "mean_token_accuracy": 0.06686681807041168,
      "num_tokens": 195112.0,
      "step": 105
    },
    {
      "entropy": 10.113536167144776,
      "epoch": 0.006155051338723666,
      "grad_norm": 1.984375,
      "learning_rate": 5.45e-05,
      "loss": 8.4636,
      "mean_token_accuracy": 0.05900593213737011,
      "num_tokens": 205951.0,
      "step": 110
    },
    {
      "entropy": 10.075279331207275,
      "epoch": 0.006434826399574742,
      "grad_norm": 2.25,
      "learning_rate": 5.7e-05,
      "loss": 8.294,
      "mean_token_accuracy": 0.07148804180324078,
      "num_tokens": 214744.0,
      "step": 115
    },
    {
      "entropy": 9.955604076385498,
      "epoch": 0.006714601460425818,
      "grad_norm": 2.25,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1343,
      "mean_token_accuracy": 0.0751000139862299,
      "num_tokens": 223279.0,
      "step": 120
    },
    {
      "entropy": 9.898994541168213,
      "epoch": 0.006994376521276893,
      "grad_norm": 1.796875,
      "learning_rate": 6.2e-05,
      "loss": 8.0915,
      "mean_token_accuracy": 0.07411009185016156,
      "num_tokens": 232677.0,
      "step": 125
    },
    {
      "entropy": 9.685633563995362,
      "epoch": 0.0072741515821279695,
      "grad_norm": 2.015625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9322,
      "mean_token_accuracy": 0.07425446137785911,
      "num_tokens": 242890.0,
      "step": 130
    },
    {
      "entropy": 9.561828994750977,
      "epoch": 0.007553926642979045,
      "grad_norm": 1.78125,
      "learning_rate": 6.7e-05,
      "loss": 7.7431,
      "mean_token_accuracy": 0.07636365555226803,
      "num_tokens": 252490.0,
      "step": 135
    },
    {
      "entropy": 9.356704235076904,
      "epoch": 0.007833701703830121,
      "grad_norm": 1.5703125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.6948,
      "mean_token_accuracy": 0.08176706209778786,
      "num_tokens": 261280.0,
      "step": 140
    },
    {
      "entropy": 9.167966556549072,
      "epoch": 0.008113476764681197,
      "grad_norm": 1.65625,
      "learning_rate": 7.2e-05,
      "loss": 7.65,
      "mean_token_accuracy": 0.08069443143904209,
      "num_tokens": 270000.0,
      "step": 145
    },
    {
      "entropy": 8.9812726020813,
      "epoch": 0.008393251825532272,
      "grad_norm": 2.453125,
      "learning_rate": 7.45e-05,
      "loss": 7.458,
      "mean_token_accuracy": 0.08502581343054771,
      "num_tokens": 279709.0,
      "step": 150
    },
    {
      "entropy": 8.729528427124023,
      "epoch": 0.008673026886383347,
      "grad_norm": 1.4453125,
      "learning_rate": 7.7e-05,
      "loss": 7.5225,
      "mean_token_accuracy": 0.082467582821846,
      "num_tokens": 288343.0,
      "step": 155
    },
    {
      "entropy": 8.595497226715088,
      "epoch": 0.008952801947234423,
      "grad_norm": 1.5234375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.3566,
      "mean_token_accuracy": 0.08212127052247524,
      "num_tokens": 297387.0,
      "step": 160
    },
    {
      "entropy": 8.476513195037843,
      "epoch": 0.0092325770080855,
      "grad_norm": 1.625,
      "learning_rate": 8.2e-05,
      "loss": 7.3103,
      "mean_token_accuracy": 0.08278237357735634,
      "num_tokens": 307175.0,
      "step": 165
    },
    {
      "entropy": 8.29503517150879,
      "epoch": 0.009512352068936575,
      "grad_norm": 1.53125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.2308,
      "mean_token_accuracy": 0.08270871713757515,
      "num_tokens": 316033.0,
      "step": 170
    },
    {
      "entropy": 8.15572385787964,
      "epoch": 0.00979212712978765,
      "grad_norm": 1.3984375,
      "learning_rate": 8.7e-05,
      "loss": 7.3076,
      "mean_token_accuracy": 0.08210290595889091,
      "num_tokens": 325740.0,
      "step": 175
    },
    {
      "entropy": 8.114148664474488,
      "epoch": 0.010071902190638726,
      "grad_norm": 1.375,
      "learning_rate": 8.95e-05,
      "loss": 7.2221,
      "mean_token_accuracy": 0.09068928360939026,
      "num_tokens": 334637.0,
      "step": 180
    },
    {
      "entropy": 8.006746578216553,
      "epoch": 0.010351677251489803,
      "grad_norm": 1.765625,
      "learning_rate": 9.2e-05,
      "loss": 7.2562,
      "mean_token_accuracy": 0.0831844873726368,
      "num_tokens": 344020.0,
      "step": 185
    },
    {
      "entropy": 7.939169883728027,
      "epoch": 0.010631452312340878,
      "grad_norm": 1.5703125,
      "learning_rate": 9.45e-05,
      "loss": 7.1292,
      "mean_token_accuracy": 0.09121446087956428,
      "num_tokens": 352924.0,
      "step": 190
    },
    {
      "entropy": 7.8939591407775875,
      "epoch": 0.010911227373191954,
      "grad_norm": 1.390625,
      "learning_rate": 9.7e-05,
      "loss": 7.149,
      "mean_token_accuracy": 0.08797023706138134,
      "num_tokens": 361755.0,
      "step": 195
    },
    {
      "entropy": 7.845557069778442,
      "epoch": 0.01119100243404303,
      "grad_norm": 1.2265625,
      "learning_rate": 9.95e-05,
      "loss": 7.1833,
      "mean_token_accuracy": 0.08474454060196876,
      "num_tokens": 370605.0,
      "step": 200
    },
    {
      "entropy": 7.757431316375732,
      "epoch": 0.011470777494894105,
      "grad_norm": 1.8125,
      "learning_rate": 0.000102,
      "loss": 7.0648,
      "mean_token_accuracy": 0.08956050127744675,
      "num_tokens": 379723.0,
      "step": 205
    },
    {
      "entropy": 7.683956050872803,
      "epoch": 0.011750552555745182,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0111,
      "mean_token_accuracy": 0.09118850380182267,
      "num_tokens": 389502.0,
      "step": 210
    },
    {
      "entropy": 7.613946056365966,
      "epoch": 0.012030327616596257,
      "grad_norm": 1.375,
      "learning_rate": 0.000107,
      "loss": 6.9865,
      "mean_token_accuracy": 0.09210970029234886,
      "num_tokens": 399116.0,
      "step": 215
    },
    {
      "entropy": 7.6888175964355465,
      "epoch": 0.012310102677447332,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001095,
      "loss": 7.0522,
      "mean_token_accuracy": 0.08312384895980358,
      "num_tokens": 408239.0,
      "step": 220
    },
    {
      "entropy": 7.582952213287354,
      "epoch": 0.012589877738298408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000112,
      "loss": 7.0558,
      "mean_token_accuracy": 0.08601429760456085,
      "num_tokens": 418801.0,
      "step": 225
    },
    {
      "entropy": 7.63437852859497,
      "epoch": 0.012869652799149483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001145,
      "loss": 7.0975,
      "mean_token_accuracy": 0.08165879175066948,
      "num_tokens": 427584.0,
      "step": 230
    },
    {
      "entropy": 7.647107696533203,
      "epoch": 0.01314942786000056,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.0398,
      "mean_token_accuracy": 0.08437106013298035,
      "num_tokens": 437345.0,
      "step": 235
    },
    {
      "entropy": 7.595585727691651,
      "epoch": 0.013429202920851636,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0693,
      "mean_token_accuracy": 0.09033490121364593,
      "num_tokens": 446550.0,
      "step": 240
    },
    {
      "entropy": 7.530496215820312,
      "epoch": 0.013708977981702711,
      "grad_norm": 1.1875,
      "learning_rate": 0.000122,
      "loss": 7.0332,
      "mean_token_accuracy": 0.08355908021330834,
      "num_tokens": 456394.0,
      "step": 245
    },
    {
      "entropy": 7.545537424087525,
      "epoch": 0.013988753042553786,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001245,
      "loss": 6.9254,
      "mean_token_accuracy": 0.09105066359043121,
      "num_tokens": 465467.0,
      "step": 250
    },
    {
      "entropy": 7.4903782367706295,
      "epoch": 0.014268528103404862,
      "grad_norm": 1.296875,
      "learning_rate": 0.000127,
      "loss": 6.9994,
      "mean_token_accuracy": 0.09321118667721748,
      "num_tokens": 474838.0,
      "step": 255
    },
    {
      "entropy": 7.407280015945434,
      "epoch": 0.014548303164255939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001295,
      "loss": 6.8929,
      "mean_token_accuracy": 0.09139523953199387,
      "num_tokens": 483414.0,
      "step": 260
    },
    {
      "entropy": 7.481622314453125,
      "epoch": 0.014828078225107014,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000132,
      "loss": 6.9248,
      "mean_token_accuracy": 0.08975184708833694,
      "num_tokens": 491802.0,
      "step": 265
    },
    {
      "entropy": 7.453138542175293,
      "epoch": 0.01510785328595809,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9211,
      "mean_token_accuracy": 0.08993491381406785,
      "num_tokens": 502007.0,
      "step": 270
    },
    {
      "entropy": 7.3567157745361325,
      "epoch": 0.015387628346809165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.9517,
      "mean_token_accuracy": 0.09103975892066955,
      "num_tokens": 510806.0,
      "step": 275
    },
    {
      "entropy": 7.387433815002441,
      "epoch": 0.015667403407660242,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001395,
      "loss": 6.8927,
      "mean_token_accuracy": 0.09123391136527062,
      "num_tokens": 519526.0,
      "step": 280
    },
    {
      "entropy": 7.356798839569092,
      "epoch": 0.015947178468511316,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8069,
      "mean_token_accuracy": 0.09702679142355919,
      "num_tokens": 529036.0,
      "step": 285
    },
    {
      "entropy": 7.348542070388794,
      "epoch": 0.016226953529362393,
      "grad_norm": 1.53125,
      "learning_rate": 0.0001445,
      "loss": 6.8877,
      "mean_token_accuracy": 0.09220745116472244,
      "num_tokens": 537758.0,
      "step": 290
    },
    {
      "entropy": 7.252509832382202,
      "epoch": 0.016506728590213467,
      "grad_norm": 1.25,
      "learning_rate": 0.000147,
      "loss": 6.797,
      "mean_token_accuracy": 0.0996977336704731,
      "num_tokens": 546401.0,
      "step": 295
    },
    {
      "entropy": 7.282312059402466,
      "epoch": 0.016786503651064544,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001495,
      "loss": 6.8229,
      "mean_token_accuracy": 0.09258124381303787,
      "num_tokens": 555839.0,
      "step": 300
    },
    {
      "entropy": 7.447148847579956,
      "epoch": 0.01706627871191562,
      "grad_norm": 1.6875,
      "learning_rate": 0.000152,
      "loss": 6.9703,
      "mean_token_accuracy": 0.09544998183846473,
      "num_tokens": 565150.0,
      "step": 305
    },
    {
      "entropy": 7.174850606918335,
      "epoch": 0.017346053772766695,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8185,
      "mean_token_accuracy": 0.09337743893265724,
      "num_tokens": 574376.0,
      "step": 310
    },
    {
      "entropy": 7.367434978485107,
      "epoch": 0.01762582883361777,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000157,
      "loss": 6.9439,
      "mean_token_accuracy": 0.09078950360417366,
      "num_tokens": 583075.0,
      "step": 315
    },
    {
      "entropy": 7.410228586196899,
      "epoch": 0.017905603894468845,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001595,
      "loss": 6.9607,
      "mean_token_accuracy": 0.09146012365818024,
      "num_tokens": 593003.0,
      "step": 320
    },
    {
      "entropy": 7.167853116989136,
      "epoch": 0.018185378955319922,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000162,
      "loss": 6.7824,
      "mean_token_accuracy": 0.10193281322717666,
      "num_tokens": 602273.0,
      "step": 325
    },
    {
      "entropy": 7.272089242935181,
      "epoch": 0.018465154016171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7087,
      "mean_token_accuracy": 0.09642531052231788,
      "num_tokens": 610364.0,
      "step": 330
    },
    {
      "entropy": 7.081447410583496,
      "epoch": 0.018744929077022073,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.7188,
      "mean_token_accuracy": 0.0957895465195179,
      "num_tokens": 618753.0,
      "step": 335
    },
    {
      "entropy": 7.264566278457641,
      "epoch": 0.01902470413787315,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8352,
      "mean_token_accuracy": 0.0967351920902729,
      "num_tokens": 627697.0,
      "step": 340
    },
    {
      "entropy": 7.207143497467041,
      "epoch": 0.019304479198724227,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7869,
      "mean_token_accuracy": 0.093389343470335,
      "num_tokens": 636499.0,
      "step": 345
    },
    {
      "entropy": 7.1959278106689455,
      "epoch": 0.0195842542595753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.8033,
      "mean_token_accuracy": 0.08842543438076973,
      "num_tokens": 646426.0,
      "step": 350
    },
    {
      "entropy": 7.133398866653442,
      "epoch": 0.019864029320426378,
      "grad_norm": 1.125,
      "learning_rate": 0.000177,
      "loss": 6.7289,
      "mean_token_accuracy": 0.09562224373221398,
      "num_tokens": 656593.0,
      "step": 355
    },
    {
      "entropy": 7.104539823532105,
      "epoch": 0.020143804381277452,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001795,
      "loss": 6.7823,
      "mean_token_accuracy": 0.0965128093957901,
      "num_tokens": 665631.0,
      "step": 360
    },
    {
      "entropy": 7.133638668060303,
      "epoch": 0.02042357944212853,
      "grad_norm": 1.21875,
      "learning_rate": 0.000182,
      "loss": 6.6893,
      "mean_token_accuracy": 0.09800055772066116,
      "num_tokens": 674943.0,
      "step": 365
    },
    {
      "entropy": 6.991700029373169,
      "epoch": 0.020703354502979606,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001845,
      "loss": 6.6059,
      "mean_token_accuracy": 0.10655041188001632,
      "num_tokens": 683500.0,
      "step": 370
    },
    {
      "entropy": 7.158477783203125,
      "epoch": 0.02098312956383068,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000187,
      "loss": 6.7946,
      "mean_token_accuracy": 0.09516055136919022,
      "num_tokens": 692147.0,
      "step": 375
    },
    {
      "entropy": 7.127016592025757,
      "epoch": 0.021262904624681757,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001895,
      "loss": 6.7322,
      "mean_token_accuracy": 0.09688446819782257,
      "num_tokens": 701226.0,
      "step": 380
    },
    {
      "entropy": 7.150118589401245,
      "epoch": 0.02154267968553283,
      "grad_norm": 1.421875,
      "learning_rate": 0.000192,
      "loss": 6.7553,
      "mean_token_accuracy": 0.10244472473859786,
      "num_tokens": 710886.0,
      "step": 385
    },
    {
      "entropy": 7.099573230743408,
      "epoch": 0.021822454746383908,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001945,
      "loss": 6.6973,
      "mean_token_accuracy": 0.09484406411647797,
      "num_tokens": 719209.0,
      "step": 390
    },
    {
      "entropy": 6.925067281723022,
      "epoch": 0.022102229807234985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.5703,
      "mean_token_accuracy": 0.10873782187700272,
      "num_tokens": 728376.0,
      "step": 395
    },
    {
      "entropy": 7.04664626121521,
      "epoch": 0.02238200486808606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6328,
      "mean_token_accuracy": 0.09875192418694496,
      "num_tokens": 737110.0,
      "step": 400
    },
    {
      "entropy": 7.108943176269531,
      "epoch": 0.022661779928937135,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000202,
      "loss": 6.6957,
      "mean_token_accuracy": 0.09161964356899262,
      "num_tokens": 746133.0,
      "step": 405
    },
    {
      "entropy": 6.9601562976837155,
      "epoch": 0.02294155498978821,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6511,
      "mean_token_accuracy": 0.10639781132340431,
      "num_tokens": 755994.0,
      "step": 410
    },
    {
      "entropy": 6.884982681274414,
      "epoch": 0.023221330050639286,
      "grad_norm": 1.296875,
      "learning_rate": 0.000207,
      "loss": 6.5945,
      "mean_token_accuracy": 0.10252679362893105,
      "num_tokens": 764630.0,
      "step": 415
    },
    {
      "entropy": 7.020522737503052,
      "epoch": 0.023501105111490363,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002095,
      "loss": 6.6171,
      "mean_token_accuracy": 0.10485479906201363,
      "num_tokens": 773577.0,
      "step": 420
    },
    {
      "entropy": 6.976370239257813,
      "epoch": 0.023780880172341437,
      "grad_norm": 1.234375,
      "learning_rate": 0.000212,
      "loss": 6.5175,
      "mean_token_accuracy": 0.10637958571314812,
      "num_tokens": 782405.0,
      "step": 425
    },
    {
      "entropy": 6.990795373916626,
      "epoch": 0.024060655233192514,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002145,
      "loss": 6.679,
      "mean_token_accuracy": 0.1029113069176674,
      "num_tokens": 790940.0,
      "step": 430
    },
    {
      "entropy": 6.904673862457275,
      "epoch": 0.024340430294043588,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6626,
      "mean_token_accuracy": 0.09815835878252983,
      "num_tokens": 800023.0,
      "step": 435
    },
    {
      "entropy": 7.0055601596832275,
      "epoch": 0.024620205354894665,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002195,
      "loss": 6.5512,
      "mean_token_accuracy": 0.11006631255149842,
      "num_tokens": 809146.0,
      "step": 440
    },
    {
      "entropy": 6.848315715789795,
      "epoch": 0.024899980415745742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000222,
      "loss": 6.5253,
      "mean_token_accuracy": 0.10747256800532341,
      "num_tokens": 818300.0,
      "step": 445
    },
    {
      "entropy": 6.93147349357605,
      "epoch": 0.025179755476596816,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002245,
      "loss": 6.5748,
      "mean_token_accuracy": 0.10701234936714173,
      "num_tokens": 826991.0,
      "step": 450
    },
    {
      "entropy": 6.828905248641968,
      "epoch": 0.025459530537447893,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.5695,
      "mean_token_accuracy": 0.10557045862078666,
      "num_tokens": 836530.0,
      "step": 455
    },
    {
      "entropy": 6.996236515045166,
      "epoch": 0.025739305598298966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6996,
      "mean_token_accuracy": 0.10405678376555443,
      "num_tokens": 845754.0,
      "step": 460
    },
    {
      "entropy": 6.909709072113037,
      "epoch": 0.026019080659150044,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5825,
      "mean_token_accuracy": 0.10672992318868638,
      "num_tokens": 854778.0,
      "step": 465
    },
    {
      "entropy": 6.931979942321777,
      "epoch": 0.02629885572000112,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5878,
      "mean_token_accuracy": 0.10348478704690933,
      "num_tokens": 863057.0,
      "step": 470
    },
    {
      "entropy": 6.977036380767823,
      "epoch": 0.026578630780852194,
      "grad_norm": 1.3125,
      "learning_rate": 0.000237,
      "loss": 6.6828,
      "mean_token_accuracy": 0.10146789252758026,
      "num_tokens": 872781.0,
      "step": 475
    },
    {
      "entropy": 6.804804468154908,
      "epoch": 0.02685840584170327,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002395,
      "loss": 6.4354,
      "mean_token_accuracy": 0.10844895020127296,
      "num_tokens": 881815.0,
      "step": 480
    },
    {
      "entropy": 6.778733253479004,
      "epoch": 0.027138180902554345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000242,
      "loss": 6.5231,
      "mean_token_accuracy": 0.10757277086377144,
      "num_tokens": 890745.0,
      "step": 485
    },
    {
      "entropy": 6.819875049591064,
      "epoch": 0.027417955963405422,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002445,
      "loss": 6.5657,
      "mean_token_accuracy": 0.10660690292716027,
      "num_tokens": 900315.0,
      "step": 490
    },
    {
      "entropy": 6.791854763031006,
      "epoch": 0.0276977310242565,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000247,
      "loss": 6.5405,
      "mean_token_accuracy": 0.10886701792478562,
      "num_tokens": 909367.0,
      "step": 495
    },
    {
      "entropy": 6.921594429016113,
      "epoch": 0.027977506085107573,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002495,
      "loss": 6.6413,
      "mean_token_accuracy": 0.10079527497291565,
      "num_tokens": 919180.0,
      "step": 500
    },
    {
      "entropy": 6.830562448501587,
      "epoch": 0.02825728114595865,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000252,
      "loss": 6.5006,
      "mean_token_accuracy": 0.11153338849544525,
      "num_tokens": 928619.0,
      "step": 505
    },
    {
      "entropy": 6.770093488693237,
      "epoch": 0.028537056206809724,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002545,
      "loss": 6.5084,
      "mean_token_accuracy": 0.10852807909250259,
      "num_tokens": 938419.0,
      "step": 510
    },
    {
      "entropy": 6.805925559997559,
      "epoch": 0.0288168312676608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000257,
      "loss": 6.5576,
      "mean_token_accuracy": 0.10293798074126244,
      "num_tokens": 947852.0,
      "step": 515
    },
    {
      "entropy": 6.79404993057251,
      "epoch": 0.029096606328511878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002595,
      "loss": 6.453,
      "mean_token_accuracy": 0.11094750836491585,
      "num_tokens": 956202.0,
      "step": 520
    },
    {
      "entropy": 6.7460289478302,
      "epoch": 0.02937638138936295,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000262,
      "loss": 6.5059,
      "mean_token_accuracy": 0.11226325705647469,
      "num_tokens": 965045.0,
      "step": 525
    },
    {
      "entropy": 6.691509819030761,
      "epoch": 0.02965615645021403,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.4299,
      "mean_token_accuracy": 0.11053601801395416,
      "num_tokens": 974100.0,
      "step": 530
    },
    {
      "entropy": 6.774121332168579,
      "epoch": 0.029935931511065102,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6318,
      "mean_token_accuracy": 0.10233537256717681,
      "num_tokens": 983177.0,
      "step": 535
    },
    {
      "entropy": 6.8688215732574465,
      "epoch": 0.03021570657191618,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.5507,
      "mean_token_accuracy": 0.10545486286282539,
      "num_tokens": 991945.0,
      "step": 540
    },
    {
      "entropy": 6.710129165649414,
      "epoch": 0.030495481632767257,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.4721,
      "mean_token_accuracy": 0.1087389424443245,
      "num_tokens": 1000987.0,
      "step": 545
    },
    {
      "entropy": 6.800864505767822,
      "epoch": 0.03077525669361833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002745,
      "loss": 6.4643,
      "mean_token_accuracy": 0.10743949860334397,
      "num_tokens": 1010531.0,
      "step": 550
    },
    {
      "entropy": 6.696709108352661,
      "epoch": 0.031055031754469407,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000277,
      "loss": 6.5063,
      "mean_token_accuracy": 0.1075245812535286,
      "num_tokens": 1018854.0,
      "step": 555
    },
    {
      "entropy": 6.655027151107788,
      "epoch": 0.031334806815320485,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002795,
      "loss": 6.548,
      "mean_token_accuracy": 0.1032408431172371,
      "num_tokens": 1028972.0,
      "step": 560
    },
    {
      "entropy": 6.847776794433594,
      "epoch": 0.031614581876171555,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5906,
      "mean_token_accuracy": 0.10243177711963654,
      "num_tokens": 1038730.0,
      "step": 565
    },
    {
      "entropy": 6.837841749191284,
      "epoch": 0.03189435693702263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002845,
      "loss": 6.5264,
      "mean_token_accuracy": 0.10695758983492851,
      "num_tokens": 1048084.0,
      "step": 570
    },
    {
      "entropy": 6.699676752090454,
      "epoch": 0.03217413199787371,
      "grad_norm": 1.234375,
      "learning_rate": 0.000287,
      "loss": 6.4611,
      "mean_token_accuracy": 0.11191356033086777,
      "num_tokens": 1056977.0,
      "step": 575
    },
    {
      "entropy": 6.561819696426392,
      "epoch": 0.032453907058724786,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002895,
      "loss": 6.3217,
      "mean_token_accuracy": 0.11810181066393852,
      "num_tokens": 1065806.0,
      "step": 580
    },
    {
      "entropy": 6.624036598205566,
      "epoch": 0.03273368211957586,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292,
      "loss": 6.3256,
      "mean_token_accuracy": 0.11459934562444687,
      "num_tokens": 1075131.0,
      "step": 585
    },
    {
      "entropy": 6.6630463123321535,
      "epoch": 0.03301345718042693,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002945,
      "loss": 6.4097,
      "mean_token_accuracy": 0.11841127201914788,
      "num_tokens": 1083924.0,
      "step": 590
    },
    {
      "entropy": 6.601504135131836,
      "epoch": 0.03329323224127801,
      "grad_norm": 1.234375,
      "learning_rate": 0.000297,
      "loss": 6.3862,
      "mean_token_accuracy": 0.11470304578542709,
      "num_tokens": 1093113.0,
      "step": 595
    },
    {
      "entropy": 6.64865255355835,
      "epoch": 0.03357300730212909,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002995,
      "loss": 6.5657,
      "mean_token_accuracy": 0.10640456900000572,
      "num_tokens": 1102300.0,
      "step": 600
    },
    {
      "entropy": 6.638816690444946,
      "epoch": 0.033852782362980165,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000302,
      "loss": 6.428,
      "mean_token_accuracy": 0.10878462493419647,
      "num_tokens": 1111202.0,
      "step": 605
    },
    {
      "entropy": 6.702850008010865,
      "epoch": 0.03413255742383124,
      "grad_norm": 1.25,
      "learning_rate": 0.0003045,
      "loss": 6.4169,
      "mean_token_accuracy": 0.11105761155486107,
      "num_tokens": 1120478.0,
      "step": 610
    },
    {
      "entropy": 6.571131038665771,
      "epoch": 0.03441233248468231,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000307,
      "loss": 6.4164,
      "mean_token_accuracy": 0.10687699913978577,
      "num_tokens": 1130469.0,
      "step": 615
    },
    {
      "entropy": 6.731548213958741,
      "epoch": 0.03469210754553339,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003095,
      "loss": 6.4974,
      "mean_token_accuracy": 0.1069604717195034,
      "num_tokens": 1140612.0,
      "step": 620
    },
    {
      "entropy": 6.594935464859009,
      "epoch": 0.034971882606384466,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000312,
      "loss": 6.4248,
      "mean_token_accuracy": 0.12048774287104606,
      "num_tokens": 1148798.0,
      "step": 625
    },
    {
      "entropy": 6.648222637176514,
      "epoch": 0.03525165766723554,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003145,
      "loss": 6.4101,
      "mean_token_accuracy": 0.11359511390328407,
      "num_tokens": 1157445.0,
      "step": 630
    },
    {
      "entropy": 6.539268827438354,
      "epoch": 0.03553143272808662,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000317,
      "loss": 6.4647,
      "mean_token_accuracy": 0.11291300356388093,
      "num_tokens": 1166407.0,
      "step": 635
    },
    {
      "entropy": 6.767864990234375,
      "epoch": 0.03581120778893769,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003195,
      "loss": 6.5337,
      "mean_token_accuracy": 0.10441585406661033,
      "num_tokens": 1175127.0,
      "step": 640
    },
    {
      "entropy": 6.594341659545899,
      "epoch": 0.03609098284978877,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000322,
      "loss": 6.442,
      "mean_token_accuracy": 0.11554119363427162,
      "num_tokens": 1184385.0,
      "step": 645
    },
    {
      "entropy": 6.585985898971558,
      "epoch": 0.036370757910639845,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.358,
      "mean_token_accuracy": 0.11490959599614144,
      "num_tokens": 1193463.0,
      "step": 650
    },
    {
      "entropy": 6.626502561569214,
      "epoch": 0.03665053297149092,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4127,
      "mean_token_accuracy": 0.11384888887405395,
      "num_tokens": 1201996.0,
      "step": 655
    },
    {
      "entropy": 6.5213230609893795,
      "epoch": 0.036930308032342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.3467,
      "mean_token_accuracy": 0.11206159815192222,
      "num_tokens": 1211489.0,
      "step": 660
    },
    {
      "entropy": 6.517415475845337,
      "epoch": 0.037210083093193076,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.3471,
      "mean_token_accuracy": 0.12239813059568405,
      "num_tokens": 1221201.0,
      "step": 665
    },
    {
      "entropy": 6.4704029083251955,
      "epoch": 0.037489858154044146,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.3499,
      "mean_token_accuracy": 0.1122575543820858,
      "num_tokens": 1229578.0,
      "step": 670
    },
    {
      "entropy": 6.515661096572876,
      "epoch": 0.037769633214895224,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000337,
      "loss": 6.3019,
      "mean_token_accuracy": 0.11475893035531044,
      "num_tokens": 1238017.0,
      "step": 675
    },
    {
      "entropy": 6.505735158920288,
      "epoch": 0.0380494082757463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003395,
      "loss": 6.3619,
      "mean_token_accuracy": 0.11612248569726943,
      "num_tokens": 1246817.0,
      "step": 680
    },
    {
      "entropy": 6.607122898101807,
      "epoch": 0.03832918333659738,
      "grad_norm": 1.25,
      "learning_rate": 0.000342,
      "loss": 6.4622,
      "mean_token_accuracy": 0.11365066468715668,
      "num_tokens": 1255598.0,
      "step": 685
    },
    {
      "entropy": 6.572498512268067,
      "epoch": 0.038608958397448455,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3592,
      "mean_token_accuracy": 0.119878139346838,
      "num_tokens": 1264197.0,
      "step": 690
    },
    {
      "entropy": 6.447942590713501,
      "epoch": 0.038888733458299525,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000347,
      "loss": 6.3071,
      "mean_token_accuracy": 0.11660419553518295,
      "num_tokens": 1273711.0,
      "step": 695
    },
    {
      "entropy": 6.46368556022644,
      "epoch": 0.0391685085191506,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003495,
      "loss": 6.3266,
      "mean_token_accuracy": 0.1126562386751175,
      "num_tokens": 1283293.0,
      "step": 700
    },
    {
      "entropy": 6.579007863998413,
      "epoch": 0.03944828358000168,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000352,
      "loss": 6.3279,
      "mean_token_accuracy": 0.11288972720503807,
      "num_tokens": 1292425.0,
      "step": 705
    },
    {
      "entropy": 6.375118112564087,
      "epoch": 0.039728058640852756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003545,
      "loss": 6.3529,
      "mean_token_accuracy": 0.1110106885433197,
      "num_tokens": 1302168.0,
      "step": 710
    },
    {
      "entropy": 6.491702222824097,
      "epoch": 0.040007833701703834,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000357,
      "loss": 6.319,
      "mean_token_accuracy": 0.11627506762742996,
      "num_tokens": 1310463.0,
      "step": 715
    },
    {
      "entropy": 6.47831015586853,
      "epoch": 0.040287608762554904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003595,
      "loss": 6.2763,
      "mean_token_accuracy": 0.12363684177398682,
      "num_tokens": 1319161.0,
      "step": 720
    },
    {
      "entropy": 6.397577905654908,
      "epoch": 0.04056738382340598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000362,
      "loss": 6.2375,
      "mean_token_accuracy": 0.11746156364679336,
      "num_tokens": 1328504.0,
      "step": 725
    },
    {
      "entropy": 6.565845918655396,
      "epoch": 0.04084715888425706,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003645,
      "loss": 6.4749,
      "mean_token_accuracy": 0.11466470956802369,
      "num_tokens": 1338246.0,
      "step": 730
    },
    {
      "entropy": 6.520300817489624,
      "epoch": 0.041126933945108135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000367,
      "loss": 6.4025,
      "mean_token_accuracy": 0.11708130165934563,
      "num_tokens": 1347338.0,
      "step": 735
    },
    {
      "entropy": 6.482167720794678,
      "epoch": 0.04140670900595921,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003695,
      "loss": 6.4618,
      "mean_token_accuracy": 0.12073256149888038,
      "num_tokens": 1356687.0,
      "step": 740
    },
    {
      "entropy": 6.528582620620727,
      "epoch": 0.04168648406681028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000372,
      "loss": 6.3314,
      "mean_token_accuracy": 0.11828168481588364,
      "num_tokens": 1366039.0,
      "step": 745
    },
    {
      "entropy": 6.514997148513794,
      "epoch": 0.04196625912766136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003745,
      "loss": 6.3703,
      "mean_token_accuracy": 0.12007363438606262,
      "num_tokens": 1375022.0,
      "step": 750
    },
    {
      "entropy": 6.558941888809204,
      "epoch": 0.04224603418851244,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.3302,
      "mean_token_accuracy": 0.12099939808249474,
      "num_tokens": 1383362.0,
      "step": 755
    },
    {
      "entropy": 6.420841073989868,
      "epoch": 0.042525809249363514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003795,
      "loss": 6.357,
      "mean_token_accuracy": 0.12097687721252441,
      "num_tokens": 1392759.0,
      "step": 760
    },
    {
      "entropy": 6.417665243148804,
      "epoch": 0.04280558431021459,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000382,
      "loss": 6.354,
      "mean_token_accuracy": 0.1138949602842331,
      "num_tokens": 1402396.0,
      "step": 765
    },
    {
      "entropy": 6.445731592178345,
      "epoch": 0.04308535937106566,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003845,
      "loss": 6.3612,
      "mean_token_accuracy": 0.12402707412838936,
      "num_tokens": 1412376.0,
      "step": 770
    },
    {
      "entropy": 6.5388038635253904,
      "epoch": 0.04336513443191674,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.351,
      "mean_token_accuracy": 0.12133746147155762,
      "num_tokens": 1422101.0,
      "step": 775
    },
    {
      "entropy": 6.427302026748658,
      "epoch": 0.043644909492767815,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3701,
      "mean_token_accuracy": 0.11939814165234566,
      "num_tokens": 1431494.0,
      "step": 780
    },
    {
      "entropy": 6.348874950408936,
      "epoch": 0.04392468455361889,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.2279,
      "mean_token_accuracy": 0.12000525146722793,
      "num_tokens": 1440444.0,
      "step": 785
    },
    {
      "entropy": 6.335124874114991,
      "epoch": 0.04420445961446997,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.1876,
      "mean_token_accuracy": 0.12222216054797172,
      "num_tokens": 1448882.0,
      "step": 790
    },
    {
      "entropy": 6.43249545097351,
      "epoch": 0.04448423467532104,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.2684,
      "mean_token_accuracy": 0.12162192389369011,
      "num_tokens": 1457730.0,
      "step": 795
    },
    {
      "entropy": 6.388329219818115,
      "epoch": 0.04476400973617212,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003995,
      "loss": 6.2851,
      "mean_token_accuracy": 0.12025910690426826,
      "num_tokens": 1466148.0,
      "step": 800
    },
    {
      "entropy": 6.339116954803467,
      "epoch": 0.045043784797023194,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402,
      "loss": 6.2755,
      "mean_token_accuracy": 0.12411149218678474,
      "num_tokens": 1475057.0,
      "step": 805
    },
    {
      "entropy": 6.443486499786377,
      "epoch": 0.04532355985787427,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004045,
      "loss": 6.3132,
      "mean_token_accuracy": 0.1179264523088932,
      "num_tokens": 1484620.0,
      "step": 810
    },
    {
      "entropy": 6.4343949317932125,
      "epoch": 0.04560333491872535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3047,
      "mean_token_accuracy": 0.12113796547055244,
      "num_tokens": 1493122.0,
      "step": 815
    },
    {
      "entropy": 6.388117790222168,
      "epoch": 0.04588310997957642,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004095,
      "loss": 6.295,
      "mean_token_accuracy": 0.11858351081609726,
      "num_tokens": 1502040.0,
      "step": 820
    },
    {
      "entropy": 6.285999011993408,
      "epoch": 0.046162885040427495,
      "grad_norm": 1.109375,
      "learning_rate": 0.000412,
      "loss": 6.2361,
      "mean_token_accuracy": 0.1263887993991375,
      "num_tokens": 1511366.0,
      "step": 825
    },
    {
      "entropy": 6.496093702316284,
      "epoch": 0.04644266010127857,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004145,
      "loss": 6.2927,
      "mean_token_accuracy": 0.128447774797678,
      "num_tokens": 1520323.0,
      "step": 830
    },
    {
      "entropy": 6.367697906494141,
      "epoch": 0.04672243516212965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000417,
      "loss": 6.3751,
      "mean_token_accuracy": 0.11920144334435463,
      "num_tokens": 1529395.0,
      "step": 835
    },
    {
      "entropy": 6.348806762695313,
      "epoch": 0.04700221022298073,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004195,
      "loss": 6.2316,
      "mean_token_accuracy": 0.12247003018856048,
      "num_tokens": 1538683.0,
      "step": 840
    },
    {
      "entropy": 6.338663005828858,
      "epoch": 0.0472819852838318,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000422,
      "loss": 6.154,
      "mean_token_accuracy": 0.13075604513287545,
      "num_tokens": 1547812.0,
      "step": 845
    },
    {
      "entropy": 6.3082603931427,
      "epoch": 0.047561760344682874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 6.273,
      "mean_token_accuracy": 0.12150180563330651,
      "num_tokens": 1557932.0,
      "step": 850
    },
    {
      "entropy": 6.31761679649353,
      "epoch": 0.04784153540553395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000427,
      "loss": 6.3218,
      "mean_token_accuracy": 0.11715059727430344,
      "num_tokens": 1567775.0,
      "step": 855
    },
    {
      "entropy": 6.289759969711303,
      "epoch": 0.04812131046638503,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004295,
      "loss": 6.2692,
      "mean_token_accuracy": 0.11850202679634095,
      "num_tokens": 1577707.0,
      "step": 860
    },
    {
      "entropy": 6.4301290035247805,
      "epoch": 0.048401085527236105,
      "grad_norm": 1.21875,
      "learning_rate": 0.000432,
      "loss": 6.2371,
      "mean_token_accuracy": 0.12808190882205964,
      "num_tokens": 1586114.0,
      "step": 865
    },
    {
      "entropy": 6.2476832389831545,
      "epoch": 0.048680860588087176,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004345,
      "loss": 6.2356,
      "mean_token_accuracy": 0.12776880636811255,
      "num_tokens": 1595172.0,
      "step": 870
    },
    {
      "entropy": 6.332582092285156,
      "epoch": 0.04896063564893825,
      "grad_norm": 1.265625,
      "learning_rate": 0.000437,
      "loss": 6.1941,
      "mean_token_accuracy": 0.12674442008137704,
      "num_tokens": 1604025.0,
      "step": 875
    },
    {
      "entropy": 6.378393983840942,
      "epoch": 0.04924041070978933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004395,
      "loss": 6.3774,
      "mean_token_accuracy": 0.12179499417543412,
      "num_tokens": 1612890.0,
      "step": 880
    },
    {
      "entropy": 6.334475088119507,
      "epoch": 0.04952018577064041,
      "grad_norm": 1.265625,
      "learning_rate": 0.000442,
      "loss": 6.2375,
      "mean_token_accuracy": 0.11860106736421586,
      "num_tokens": 1622153.0,
      "step": 885
    },
    {
      "entropy": 6.3245104312896725,
      "epoch": 0.049799960831491484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004445,
      "loss": 6.2902,
      "mean_token_accuracy": 0.12127727717161178,
      "num_tokens": 1631687.0,
      "step": 890
    },
    {
      "entropy": 6.444584751129151,
      "epoch": 0.050079735892342554,
      "grad_norm": 1.265625,
      "learning_rate": 0.000447,
      "loss": 6.363,
      "mean_token_accuracy": 0.11959558203816414,
      "num_tokens": 1641062.0,
      "step": 895
    },
    {
      "entropy": 6.220570611953735,
      "epoch": 0.05035951095319363,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.268,
      "mean_token_accuracy": 0.11896927058696746,
      "num_tokens": 1650291.0,
      "step": 900
    },
    {
      "entropy": 6.363810300827026,
      "epoch": 0.05063928601404471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.1853,
      "mean_token_accuracy": 0.12382128313183785,
      "num_tokens": 1659102.0,
      "step": 905
    },
    {
      "entropy": 6.265335369110107,
      "epoch": 0.050919061074895786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.1083,
      "mean_token_accuracy": 0.12587042823433875,
      "num_tokens": 1667835.0,
      "step": 910
    },
    {
      "entropy": 6.241576528549194,
      "epoch": 0.05119883613574686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.2574,
      "mean_token_accuracy": 0.12526341378688813,
      "num_tokens": 1676619.0,
      "step": 915
    },
    {
      "entropy": 6.208951044082641,
      "epoch": 0.05147861119659793,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2091,
      "mean_token_accuracy": 0.12771726846694947,
      "num_tokens": 1686145.0,
      "step": 920
    },
    {
      "entropy": 6.403143787384034,
      "epoch": 0.05175838625744901,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000462,
      "loss": 6.3385,
      "mean_token_accuracy": 0.1172799788415432,
      "num_tokens": 1695667.0,
      "step": 925
    },
    {
      "entropy": 6.366543626785278,
      "epoch": 0.05203816131830009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004645,
      "loss": 6.2425,
      "mean_token_accuracy": 0.11943399086594582,
      "num_tokens": 1704269.0,
      "step": 930
    },
    {
      "entropy": 6.236631679534912,
      "epoch": 0.052317936379151164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000467,
      "loss": 6.2598,
      "mean_token_accuracy": 0.12105512395501136,
      "num_tokens": 1713185.0,
      "step": 935
    },
    {
      "entropy": 6.354249143600464,
      "epoch": 0.05259771144000224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004695,
      "loss": 6.2887,
      "mean_token_accuracy": 0.12139227539300919,
      "num_tokens": 1722433.0,
      "step": 940
    },
    {
      "entropy": 6.341638278961182,
      "epoch": 0.05287748650085331,
      "grad_norm": 1.15625,
      "learning_rate": 0.000472,
      "loss": 6.2594,
      "mean_token_accuracy": 0.12757667824625968,
      "num_tokens": 1731763.0,
      "step": 945
    },
    {
      "entropy": 6.205323266983032,
      "epoch": 0.05315726156170439,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004745,
      "loss": 6.2029,
      "mean_token_accuracy": 0.12266376763582229,
      "num_tokens": 1741470.0,
      "step": 950
    },
    {
      "entropy": 6.175953912734985,
      "epoch": 0.053437036622555466,
      "grad_norm": 1.140625,
      "learning_rate": 0.000477,
      "loss": 6.0959,
      "mean_token_accuracy": 0.12836336940526963,
      "num_tokens": 1751099.0,
      "step": 955
    },
    {
      "entropy": 6.2856032371521,
      "epoch": 0.05371681168340654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004795,
      "loss": 6.2235,
      "mean_token_accuracy": 0.12410176768898964,
      "num_tokens": 1760296.0,
      "step": 960
    },
    {
      "entropy": 6.204359102249145,
      "epoch": 0.05399658674425762,
      "grad_norm": 1.1875,
      "learning_rate": 0.000482,
      "loss": 6.2335,
      "mean_token_accuracy": 0.11641522720456124,
      "num_tokens": 1768728.0,
      "step": 965
    },
    {
      "entropy": 6.363514423370361,
      "epoch": 0.05427636180510869,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004845,
      "loss": 6.1788,
      "mean_token_accuracy": 0.12237933203577996,
      "num_tokens": 1778413.0,
      "step": 970
    },
    {
      "entropy": 6.113896322250366,
      "epoch": 0.05455613686595977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000487,
      "loss": 6.1779,
      "mean_token_accuracy": 0.12829331532120705,
      "num_tokens": 1787604.0,
      "step": 975
    },
    {
      "entropy": 6.2251220703125,
      "epoch": 0.054835911926810844,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004895,
      "loss": 6.1367,
      "mean_token_accuracy": 0.13013790622353555,
      "num_tokens": 1796633.0,
      "step": 980
    },
    {
      "entropy": 6.2790031909942625,
      "epoch": 0.05511568698766192,
      "grad_norm": 1.234375,
      "learning_rate": 0.000492,
      "loss": 6.2145,
      "mean_token_accuracy": 0.1300550229847431,
      "num_tokens": 1805275.0,
      "step": 985
    },
    {
      "entropy": 6.169535732269287,
      "epoch": 0.055395462048513,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004945,
      "loss": 6.1657,
      "mean_token_accuracy": 0.12985724806785584,
      "num_tokens": 1813640.0,
      "step": 990
    },
    {
      "entropy": 6.210285711288452,
      "epoch": 0.05567523710936407,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497,
      "loss": 6.1682,
      "mean_token_accuracy": 0.12482047080993652,
      "num_tokens": 1822524.0,
      "step": 995
    },
    {
      "entropy": 6.18486385345459,
      "epoch": 0.055955012170215146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995,
      "loss": 6.1837,
      "mean_token_accuracy": 0.12522482872009277,
      "num_tokens": 1832799.0,
      "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": 2689797335040000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}