{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.368200836820083,
  "eval_steps": 500,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.768266344070435,
      "epoch": 0.0041841004184100415,
      "grad_norm": 12.875,
      "learning_rate": 2e-06,
      "loss": 14.3834,
      "mean_token_accuracy": 0.0,
      "num_tokens": 9848.0,
      "step": 5
    },
    {
      "entropy": 4.7789147853851315,
      "epoch": 0.008368200836820083,
      "grad_norm": 14.3125,
      "learning_rate": 4.5e-06,
      "loss": 14.2092,
      "mean_token_accuracy": 0.00013605442363768817,
      "num_tokens": 18196.0,
      "step": 10
    },
    {
      "entropy": 4.821842193603516,
      "epoch": 0.012552301255230125,
      "grad_norm": 17.25,
      "learning_rate": 7e-06,
      "loss": 14.0541,
      "mean_token_accuracy": 0.0002630572533234954,
      "num_tokens": 26927.0,
      "step": 15
    },
    {
      "entropy": 5.096258974075317,
      "epoch": 0.016736401673640166,
      "grad_norm": 27.125,
      "learning_rate": 9.5e-06,
      "loss": 13.3612,
      "mean_token_accuracy": 0.00012658227933570743,
      "num_tokens": 34987.0,
      "step": 20
    },
    {
      "entropy": 7.207125568389893,
      "epoch": 0.02092050209205021,
      "grad_norm": 19.5,
      "learning_rate": 1.2e-05,
      "loss": 11.9388,
      "mean_token_accuracy": 0.0,
      "num_tokens": 44936.0,
      "step": 25
    },
    {
      "entropy": 10.419103336334228,
      "epoch": 0.02510460251046025,
      "grad_norm": 4.25,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.8146,
      "mean_token_accuracy": 0.0020074134168680756,
      "num_tokens": 54562.0,
      "step": 30
    },
    {
      "entropy": 10.69917049407959,
      "epoch": 0.029288702928870293,
      "grad_norm": 3.296875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.5537,
      "mean_token_accuracy": 0.0153768889605999,
      "num_tokens": 63844.0,
      "step": 35
    },
    {
      "entropy": 10.692293643951416,
      "epoch": 0.03347280334728033,
      "grad_norm": 2.71875,
      "learning_rate": 1.95e-05,
      "loss": 10.2761,
      "mean_token_accuracy": 0.017915738467127084,
      "num_tokens": 74331.0,
      "step": 40
    },
    {
      "entropy": 10.646587371826172,
      "epoch": 0.03765690376569038,
      "grad_norm": 2.6875,
      "learning_rate": 2.2e-05,
      "loss": 9.904,
      "mean_token_accuracy": 0.028478659689426422,
      "num_tokens": 83028.0,
      "step": 45
    },
    {
      "entropy": 10.585913181304932,
      "epoch": 0.04184100418410042,
      "grad_norm": 2.203125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6977,
      "mean_token_accuracy": 0.04541164524853229,
      "num_tokens": 91395.0,
      "step": 50
    },
    {
      "entropy": 10.563683032989502,
      "epoch": 0.04602510460251046,
      "grad_norm": 2.421875,
      "learning_rate": 2.7e-05,
      "loss": 9.5786,
      "mean_token_accuracy": 0.048107235319912435,
      "num_tokens": 101868.0,
      "step": 55
    },
    {
      "entropy": 10.548775959014893,
      "epoch": 0.0502092050209205,
      "grad_norm": 2.03125,
      "learning_rate": 2.95e-05,
      "loss": 9.4624,
      "mean_token_accuracy": 0.04703106284141541,
      "num_tokens": 110403.0,
      "step": 60
    },
    {
      "entropy": 10.547953414916993,
      "epoch": 0.05439330543933055,
      "grad_norm": 1.9609375,
      "learning_rate": 3.2e-05,
      "loss": 9.5071,
      "mean_token_accuracy": 0.041691455990076065,
      "num_tokens": 119557.0,
      "step": 65
    },
    {
      "entropy": 10.555264663696288,
      "epoch": 0.058577405857740586,
      "grad_norm": 2.09375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.3084,
      "mean_token_accuracy": 0.04668764472007751,
      "num_tokens": 128903.0,
      "step": 70
    },
    {
      "entropy": 10.545580863952637,
      "epoch": 0.06276150627615062,
      "grad_norm": 1.859375,
      "learning_rate": 3.7e-05,
      "loss": 9.3022,
      "mean_token_accuracy": 0.04386243894696236,
      "num_tokens": 137630.0,
      "step": 75
    },
    {
      "entropy": 10.519606494903565,
      "epoch": 0.06694560669456066,
      "grad_norm": 2.046875,
      "learning_rate": 3.95e-05,
      "loss": 9.12,
      "mean_token_accuracy": 0.05209527462720871,
      "num_tokens": 146335.0,
      "step": 80
    },
    {
      "entropy": 10.482771396636963,
      "epoch": 0.07112970711297072,
      "grad_norm": 2.03125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0283,
      "mean_token_accuracy": 0.05407022833824158,
      "num_tokens": 155766.0,
      "step": 85
    },
    {
      "entropy": 10.450575256347657,
      "epoch": 0.07531380753138076,
      "grad_norm": 1.890625,
      "learning_rate": 4.45e-05,
      "loss": 8.9477,
      "mean_token_accuracy": 0.05165062099695206,
      "num_tokens": 165844.0,
      "step": 90
    },
    {
      "entropy": 10.425101280212402,
      "epoch": 0.0794979079497908,
      "grad_norm": 1.8671875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8842,
      "mean_token_accuracy": 0.05370258502662182,
      "num_tokens": 174858.0,
      "step": 95
    },
    {
      "entropy": 10.32920789718628,
      "epoch": 0.08368200836820083,
      "grad_norm": 1.9140625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7167,
      "mean_token_accuracy": 0.05981660895049572,
      "num_tokens": 184020.0,
      "step": 100
    },
    {
      "entropy": 10.181023502349854,
      "epoch": 0.08786610878661087,
      "grad_norm": 1.859375,
      "learning_rate": 5.2e-05,
      "loss": 8.5856,
      "mean_token_accuracy": 0.05290363691747189,
      "num_tokens": 193714.0,
      "step": 105
    },
    {
      "entropy": 10.198339366912842,
      "epoch": 0.09205020920502092,
      "grad_norm": 1.796875,
      "learning_rate": 5.45e-05,
      "loss": 8.5126,
      "mean_token_accuracy": 0.0547657772898674,
      "num_tokens": 203139.0,
      "step": 110
    },
    {
      "entropy": 10.105471420288087,
      "epoch": 0.09623430962343096,
      "grad_norm": 1.6171875,
      "learning_rate": 5.7e-05,
      "loss": 8.3484,
      "mean_token_accuracy": 0.05634176060557365,
      "num_tokens": 212446.0,
      "step": 115
    },
    {
      "entropy": 9.886044597625732,
      "epoch": 0.100418410041841,
      "grad_norm": 1.4765625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1725,
      "mean_token_accuracy": 0.06103258766233921,
      "num_tokens": 220910.0,
      "step": 120
    },
    {
      "entropy": 9.764387893676759,
      "epoch": 0.10460251046025104,
      "grad_norm": 1.3984375,
      "learning_rate": 6.2e-05,
      "loss": 8.096,
      "mean_token_accuracy": 0.053500480577349664,
      "num_tokens": 230522.0,
      "step": 125
    },
    {
      "entropy": 9.598654460906982,
      "epoch": 0.1087866108786611,
      "grad_norm": 1.3125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.918,
      "mean_token_accuracy": 0.05621385425329208,
      "num_tokens": 240092.0,
      "step": 130
    },
    {
      "entropy": 9.302083587646484,
      "epoch": 0.11297071129707113,
      "grad_norm": 1.0,
      "learning_rate": 6.7e-05,
      "loss": 7.8502,
      "mean_token_accuracy": 0.05275089591741562,
      "num_tokens": 249478.0,
      "step": 135
    },
    {
      "entropy": 9.09952507019043,
      "epoch": 0.11715481171548117,
      "grad_norm": 1.2109375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.6323,
      "mean_token_accuracy": 0.06236257366836071,
      "num_tokens": 257714.0,
      "step": 140
    },
    {
      "entropy": 8.845372676849365,
      "epoch": 0.12133891213389121,
      "grad_norm": 1.28125,
      "learning_rate": 7.2e-05,
      "loss": 7.672,
      "mean_token_accuracy": 0.05868846960365772,
      "num_tokens": 266569.0,
      "step": 145
    },
    {
      "entropy": 8.660739707946778,
      "epoch": 0.12552301255230125,
      "grad_norm": 0.89453125,
      "learning_rate": 7.45e-05,
      "loss": 7.5683,
      "mean_token_accuracy": 0.056071584671735765,
      "num_tokens": 276688.0,
      "step": 150
    },
    {
      "entropy": 8.484380149841309,
      "epoch": 0.1297071129707113,
      "grad_norm": 0.9453125,
      "learning_rate": 7.7e-05,
      "loss": 7.4186,
      "mean_token_accuracy": 0.06363016776740552,
      "num_tokens": 285430.0,
      "step": 155
    },
    {
      "entropy": 8.278593254089355,
      "epoch": 0.13389121338912133,
      "grad_norm": 1.1640625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4137,
      "mean_token_accuracy": 0.06225575767457485,
      "num_tokens": 295727.0,
      "step": 160
    },
    {
      "entropy": 8.146096420288085,
      "epoch": 0.13807531380753138,
      "grad_norm": 1.9453125,
      "learning_rate": 8.2e-05,
      "loss": 7.3823,
      "mean_token_accuracy": 0.0689712069928646,
      "num_tokens": 304854.0,
      "step": 165
    },
    {
      "entropy": 7.988676738739014,
      "epoch": 0.14225941422594143,
      "grad_norm": 1.2109375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3536,
      "mean_token_accuracy": 0.07053346484899521,
      "num_tokens": 313579.0,
      "step": 170
    },
    {
      "entropy": 7.9785551071167,
      "epoch": 0.14644351464435146,
      "grad_norm": 1.0703125,
      "learning_rate": 8.7e-05,
      "loss": 7.4132,
      "mean_token_accuracy": 0.07007367871701717,
      "num_tokens": 323310.0,
      "step": 175
    },
    {
      "entropy": 7.922500085830689,
      "epoch": 0.1506276150627615,
      "grad_norm": 1.140625,
      "learning_rate": 8.95e-05,
      "loss": 7.2859,
      "mean_token_accuracy": 0.0698877926915884,
      "num_tokens": 331720.0,
      "step": 180
    },
    {
      "entropy": 7.924635601043701,
      "epoch": 0.15481171548117154,
      "grad_norm": 1.34375,
      "learning_rate": 9.2e-05,
      "loss": 7.24,
      "mean_token_accuracy": 0.07415937893092632,
      "num_tokens": 340331.0,
      "step": 185
    },
    {
      "entropy": 7.836748027801514,
      "epoch": 0.1589958158995816,
      "grad_norm": 0.98828125,
      "learning_rate": 9.45e-05,
      "loss": 7.2493,
      "mean_token_accuracy": 0.07680297121405602,
      "num_tokens": 349677.0,
      "step": 190
    },
    {
      "entropy": 7.846647310256958,
      "epoch": 0.16317991631799164,
      "grad_norm": 1.125,
      "learning_rate": 9.7e-05,
      "loss": 7.2344,
      "mean_token_accuracy": 0.06987173184752464,
      "num_tokens": 358674.0,
      "step": 195
    },
    {
      "entropy": 7.753779888153076,
      "epoch": 0.16736401673640167,
      "grad_norm": 1.2421875,
      "learning_rate": 9.95e-05,
      "loss": 7.2296,
      "mean_token_accuracy": 0.072166907787323,
      "num_tokens": 368258.0,
      "step": 200
    },
    {
      "entropy": 7.816751956939697,
      "epoch": 0.17154811715481172,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000102,
      "loss": 7.1436,
      "mean_token_accuracy": 0.07943947166204453,
      "num_tokens": 377083.0,
      "step": 205
    },
    {
      "entropy": 7.698982191085816,
      "epoch": 0.17573221757322174,
      "grad_norm": 1.578125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.1951,
      "mean_token_accuracy": 0.07452086620032787,
      "num_tokens": 385456.0,
      "step": 210
    },
    {
      "entropy": 7.709808778762818,
      "epoch": 0.1799163179916318,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000107,
      "loss": 7.0971,
      "mean_token_accuracy": 0.0767411794513464,
      "num_tokens": 395295.0,
      "step": 215
    },
    {
      "entropy": 7.692369031906128,
      "epoch": 0.18410041841004185,
      "grad_norm": 1.765625,
      "learning_rate": 0.0001095,
      "loss": 7.1406,
      "mean_token_accuracy": 0.07074209861457348,
      "num_tokens": 405023.0,
      "step": 220
    },
    {
      "entropy": 7.62271146774292,
      "epoch": 0.18828451882845187,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000112,
      "loss": 7.1623,
      "mean_token_accuracy": 0.07522214204072952,
      "num_tokens": 414224.0,
      "step": 225
    },
    {
      "entropy": 7.621326494216919,
      "epoch": 0.19246861924686193,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001145,
      "loss": 7.0582,
      "mean_token_accuracy": 0.0759123720228672,
      "num_tokens": 423951.0,
      "step": 230
    },
    {
      "entropy": 7.696398591995239,
      "epoch": 0.19665271966527198,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.0805,
      "mean_token_accuracy": 0.07548626773059368,
      "num_tokens": 432427.0,
      "step": 235
    },
    {
      "entropy": 7.593403148651123,
      "epoch": 0.200836820083682,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.0032,
      "mean_token_accuracy": 0.0808411881327629,
      "num_tokens": 441560.0,
      "step": 240
    },
    {
      "entropy": 7.589205026626587,
      "epoch": 0.20502092050209206,
      "grad_norm": 1.28125,
      "learning_rate": 0.000122,
      "loss": 7.092,
      "mean_token_accuracy": 0.0739415742456913,
      "num_tokens": 450364.0,
      "step": 245
    },
    {
      "entropy": 7.541353750228882,
      "epoch": 0.20920502092050208,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001245,
      "loss": 7.0399,
      "mean_token_accuracy": 0.0743038758635521,
      "num_tokens": 460076.0,
      "step": 250
    },
    {
      "entropy": 7.461675310134888,
      "epoch": 0.21338912133891214,
      "grad_norm": 1.015625,
      "learning_rate": 0.000127,
      "loss": 6.997,
      "mean_token_accuracy": 0.07904429398477078,
      "num_tokens": 468718.0,
      "step": 255
    },
    {
      "entropy": 7.617227411270141,
      "epoch": 0.2175732217573222,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001295,
      "loss": 7.0585,
      "mean_token_accuracy": 0.07864544317126274,
      "num_tokens": 478436.0,
      "step": 260
    },
    {
      "entropy": 7.475518226623535,
      "epoch": 0.2217573221757322,
      "grad_norm": 1.1875,
      "learning_rate": 0.000132,
      "loss": 7.0628,
      "mean_token_accuracy": 0.07705017924308777,
      "num_tokens": 486833.0,
      "step": 265
    },
    {
      "entropy": 7.514133024215698,
      "epoch": 0.22594142259414227,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.9359,
      "mean_token_accuracy": 0.08766618892550468,
      "num_tokens": 495119.0,
      "step": 270
    },
    {
      "entropy": 7.481584882736206,
      "epoch": 0.2301255230125523,
      "grad_norm": 1.921875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1039,
      "mean_token_accuracy": 0.08408359885215759,
      "num_tokens": 505396.0,
      "step": 275
    },
    {
      "entropy": 7.400933456420899,
      "epoch": 0.23430962343096234,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001395,
      "loss": 6.9713,
      "mean_token_accuracy": 0.08110813125967979,
      "num_tokens": 515259.0,
      "step": 280
    },
    {
      "entropy": 7.482255935668945,
      "epoch": 0.2384937238493724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.9695,
      "mean_token_accuracy": 0.08075925037264824,
      "num_tokens": 523608.0,
      "step": 285
    },
    {
      "entropy": 7.380628252029419,
      "epoch": 0.24267782426778242,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001445,
      "loss": 6.9614,
      "mean_token_accuracy": 0.08079174906015396,
      "num_tokens": 534613.0,
      "step": 290
    },
    {
      "entropy": 7.431663274765015,
      "epoch": 0.24686192468619247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000147,
      "loss": 6.9779,
      "mean_token_accuracy": 0.08681405410170555,
      "num_tokens": 543347.0,
      "step": 295
    },
    {
      "entropy": 7.367568683624268,
      "epoch": 0.2510460251046025,
      "grad_norm": 1.25,
      "learning_rate": 0.0001495,
      "loss": 6.9373,
      "mean_token_accuracy": 0.08151624575257302,
      "num_tokens": 552371.0,
      "step": 300
    },
    {
      "entropy": 7.431672620773315,
      "epoch": 0.25523012552301255,
      "grad_norm": 1.140625,
      "learning_rate": 0.000152,
      "loss": 6.8709,
      "mean_token_accuracy": 0.08783738985657692,
      "num_tokens": 560742.0,
      "step": 305
    },
    {
      "entropy": 7.372213315963745,
      "epoch": 0.2594142259414226,
      "grad_norm": 1.09375,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.9672,
      "mean_token_accuracy": 0.08577736541628837,
      "num_tokens": 569960.0,
      "step": 310
    },
    {
      "entropy": 7.366195917129517,
      "epoch": 0.26359832635983266,
      "grad_norm": 1.078125,
      "learning_rate": 0.000157,
      "loss": 6.935,
      "mean_token_accuracy": 0.08148402348160744,
      "num_tokens": 578859.0,
      "step": 315
    },
    {
      "entropy": 7.361504554748535,
      "epoch": 0.26778242677824265,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001595,
      "loss": 6.9323,
      "mean_token_accuracy": 0.08335322514176369,
      "num_tokens": 588146.0,
      "step": 320
    },
    {
      "entropy": 7.262340354919433,
      "epoch": 0.2719665271966527,
      "grad_norm": 1.671875,
      "learning_rate": 0.000162,
      "loss": 6.8837,
      "mean_token_accuracy": 0.08676013201475144,
      "num_tokens": 597268.0,
      "step": 325
    },
    {
      "entropy": 7.286304664611817,
      "epoch": 0.27615062761506276,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.8065,
      "mean_token_accuracy": 0.09255077987909317,
      "num_tokens": 606199.0,
      "step": 330
    },
    {
      "entropy": 7.354407405853271,
      "epoch": 0.2803347280334728,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.9221,
      "mean_token_accuracy": 0.07822546139359474,
      "num_tokens": 615966.0,
      "step": 335
    },
    {
      "entropy": 7.328773784637451,
      "epoch": 0.28451882845188287,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8936,
      "mean_token_accuracy": 0.08315553963184356,
      "num_tokens": 625689.0,
      "step": 340
    },
    {
      "entropy": 7.292076635360718,
      "epoch": 0.28870292887029286,
      "grad_norm": 1.09375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9258,
      "mean_token_accuracy": 0.08275936022400857,
      "num_tokens": 634877.0,
      "step": 345
    },
    {
      "entropy": 7.311211967468262,
      "epoch": 0.2928870292887029,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.8606,
      "mean_token_accuracy": 0.08924969360232353,
      "num_tokens": 644042.0,
      "step": 350
    },
    {
      "entropy": 7.213941431045532,
      "epoch": 0.29707112970711297,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000177,
      "loss": 6.8394,
      "mean_token_accuracy": 0.08540538996458054,
      "num_tokens": 653072.0,
      "step": 355
    },
    {
      "entropy": 7.225866985321045,
      "epoch": 0.301255230125523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001795,
      "loss": 6.8411,
      "mean_token_accuracy": 0.08791597038507462,
      "num_tokens": 663686.0,
      "step": 360
    },
    {
      "entropy": 7.237736654281616,
      "epoch": 0.3054393305439331,
      "grad_norm": 1.09375,
      "learning_rate": 0.000182,
      "loss": 6.774,
      "mean_token_accuracy": 0.0887495793402195,
      "num_tokens": 672293.0,
      "step": 365
    },
    {
      "entropy": 7.061402034759522,
      "epoch": 0.30962343096234307,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001845,
      "loss": 6.7483,
      "mean_token_accuracy": 0.09100355952978134,
      "num_tokens": 681758.0,
      "step": 370
    },
    {
      "entropy": 7.178809404373169,
      "epoch": 0.3138075313807531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000187,
      "loss": 6.7736,
      "mean_token_accuracy": 0.09110940396785736,
      "num_tokens": 690727.0,
      "step": 375
    },
    {
      "entropy": 7.200173044204712,
      "epoch": 0.3179916317991632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001895,
      "loss": 6.7292,
      "mean_token_accuracy": 0.08735076114535331,
      "num_tokens": 701210.0,
      "step": 380
    },
    {
      "entropy": 7.065630149841309,
      "epoch": 0.32217573221757323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000192,
      "loss": 6.7168,
      "mean_token_accuracy": 0.09401161968708038,
      "num_tokens": 710869.0,
      "step": 385
    },
    {
      "entropy": 7.200049877166748,
      "epoch": 0.3263598326359833,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001945,
      "loss": 6.8296,
      "mean_token_accuracy": 0.08618490770459175,
      "num_tokens": 721455.0,
      "step": 390
    },
    {
      "entropy": 7.104738330841064,
      "epoch": 0.3305439330543933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7259,
      "mean_token_accuracy": 0.09676406383514405,
      "num_tokens": 730917.0,
      "step": 395
    },
    {
      "entropy": 7.051578664779663,
      "epoch": 0.33472803347280333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7446,
      "mean_token_accuracy": 0.09235388711094857,
      "num_tokens": 739302.0,
      "step": 400
    },
    {
      "entropy": 7.140300941467285,
      "epoch": 0.3389121338912134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000202,
      "loss": 6.7141,
      "mean_token_accuracy": 0.08636429682374,
      "num_tokens": 748785.0,
      "step": 405
    },
    {
      "entropy": 7.019905233383179,
      "epoch": 0.34309623430962344,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6677,
      "mean_token_accuracy": 0.09201755002140999,
      "num_tokens": 757215.0,
      "step": 410
    },
    {
      "entropy": 7.04399356842041,
      "epoch": 0.3472803347280335,
      "grad_norm": 1.3125,
      "learning_rate": 0.000207,
      "loss": 6.5945,
      "mean_token_accuracy": 0.09794526174664497,
      "num_tokens": 765682.0,
      "step": 415
    },
    {
      "entropy": 7.094897365570068,
      "epoch": 0.3514644351464435,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002095,
      "loss": 6.7956,
      "mean_token_accuracy": 0.08788713663816453,
      "num_tokens": 774814.0,
      "step": 420
    },
    {
      "entropy": 7.0791693210601805,
      "epoch": 0.35564853556485354,
      "grad_norm": 1.5625,
      "learning_rate": 0.000212,
      "loss": 6.7096,
      "mean_token_accuracy": 0.09370236843824387,
      "num_tokens": 784019.0,
      "step": 425
    },
    {
      "entropy": 6.943140363693237,
      "epoch": 0.3598326359832636,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002145,
      "loss": 6.5796,
      "mean_token_accuracy": 0.09640125781297684,
      "num_tokens": 793237.0,
      "step": 430
    },
    {
      "entropy": 7.025385618209839,
      "epoch": 0.36401673640167365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6309,
      "mean_token_accuracy": 0.09518588036298752,
      "num_tokens": 801548.0,
      "step": 435
    },
    {
      "entropy": 7.028780126571656,
      "epoch": 0.3682008368200837,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002195,
      "loss": 6.6643,
      "mean_token_accuracy": 0.09595847800374031,
      "num_tokens": 810974.0,
      "step": 440
    },
    {
      "entropy": 6.958902835845947,
      "epoch": 0.3723849372384937,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000222,
      "loss": 6.5681,
      "mean_token_accuracy": 0.09765126407146454,
      "num_tokens": 820023.0,
      "step": 445
    },
    {
      "entropy": 6.949744319915771,
      "epoch": 0.37656903765690375,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002245,
      "loss": 6.6524,
      "mean_token_accuracy": 0.09723440259695053,
      "num_tokens": 828297.0,
      "step": 450
    },
    {
      "entropy": 7.0269598960876465,
      "epoch": 0.3807531380753138,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7028,
      "mean_token_accuracy": 0.08575191274285317,
      "num_tokens": 838268.0,
      "step": 455
    },
    {
      "entropy": 7.016048622131348,
      "epoch": 0.38493723849372385,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.623,
      "mean_token_accuracy": 0.09345634952187538,
      "num_tokens": 846491.0,
      "step": 460
    },
    {
      "entropy": 6.9051624774932865,
      "epoch": 0.3891213389121339,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5585,
      "mean_token_accuracy": 0.10123546048998833,
      "num_tokens": 856501.0,
      "step": 465
    },
    {
      "entropy": 6.930023145675659,
      "epoch": 0.39330543933054396,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.6253,
      "mean_token_accuracy": 0.09574304819107056,
      "num_tokens": 867054.0,
      "step": 470
    },
    {
      "entropy": 6.9083311557769775,
      "epoch": 0.39748953974895396,
      "grad_norm": 1.28125,
      "learning_rate": 0.000237,
      "loss": 6.6316,
      "mean_token_accuracy": 0.09689460024237632,
      "num_tokens": 875956.0,
      "step": 475
    },
    {
      "entropy": 6.80507345199585,
      "epoch": 0.401673640167364,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002395,
      "loss": 6.5549,
      "mean_token_accuracy": 0.09844572842121124,
      "num_tokens": 884427.0,
      "step": 480
    },
    {
      "entropy": 6.988656139373779,
      "epoch": 0.40585774058577406,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000242,
      "loss": 6.6402,
      "mean_token_accuracy": 0.10031970292329788,
      "num_tokens": 894181.0,
      "step": 485
    },
    {
      "entropy": 6.854498291015625,
      "epoch": 0.4100418410041841,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002445,
      "loss": 6.4914,
      "mean_token_accuracy": 0.10208117067813874,
      "num_tokens": 904076.0,
      "step": 490
    },
    {
      "entropy": 6.82157187461853,
      "epoch": 0.41422594142259417,
      "grad_norm": 1.34375,
      "learning_rate": 0.000247,
      "loss": 6.6406,
      "mean_token_accuracy": 0.10029161274433136,
      "num_tokens": 913915.0,
      "step": 495
    },
    {
      "entropy": 6.97306637763977,
      "epoch": 0.41841004184100417,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002495,
      "loss": 6.6935,
      "mean_token_accuracy": 0.09564930200576782,
      "num_tokens": 922876.0,
      "step": 500
    },
    {
      "epoch": 0.41841004184100417,
      "eval_entropy": 6.967948366527908,
      "eval_loss": 6.639475345611572,
      "eval_mean_token_accuracy": 0.09896371554743293,
      "eval_num_tokens": 922876.0,
      "eval_runtime": 2.8273,
      "eval_samples_per_second": 1381.864,
      "eval_steps_per_second": 172.954,
      "step": 500
    },
    {
      "entropy": 6.84868803024292,
      "epoch": 0.4225941422594142,
      "grad_norm": 1.46875,
      "learning_rate": 0.000252,
      "loss": 6.4933,
      "mean_token_accuracy": 0.10384752005338668,
      "num_tokens": 931998.0,
      "step": 505
    },
    {
      "entropy": 6.951689100265503,
      "epoch": 0.42677824267782427,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002545,
      "loss": 6.5497,
      "mean_token_accuracy": 0.09913533478975296,
      "num_tokens": 941387.0,
      "step": 510
    },
    {
      "entropy": 6.808173084259034,
      "epoch": 0.4309623430962343,
      "grad_norm": 1.140625,
      "learning_rate": 0.000257,
      "loss": 6.5857,
      "mean_token_accuracy": 0.09512314647436142,
      "num_tokens": 951866.0,
      "step": 515
    },
    {
      "entropy": 6.7702148914337155,
      "epoch": 0.4351464435146444,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002595,
      "loss": 6.5134,
      "mean_token_accuracy": 0.1027239017188549,
      "num_tokens": 961252.0,
      "step": 520
    },
    {
      "entropy": 6.861008834838867,
      "epoch": 0.4393305439330544,
      "grad_norm": 1.71875,
      "learning_rate": 0.000262,
      "loss": 6.5456,
      "mean_token_accuracy": 0.10307287946343421,
      "num_tokens": 969857.0,
      "step": 525
    },
    {
      "entropy": 6.827121829986572,
      "epoch": 0.4435146443514644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5208,
      "mean_token_accuracy": 0.10130328312516212,
      "num_tokens": 978754.0,
      "step": 530
    },
    {
      "entropy": 6.812819004058838,
      "epoch": 0.4476987447698745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5024,
      "mean_token_accuracy": 0.10306508988142013,
      "num_tokens": 987935.0,
      "step": 535
    },
    {
      "entropy": 6.823488569259643,
      "epoch": 0.45188284518828453,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.4771,
      "mean_token_accuracy": 0.1015734501183033,
      "num_tokens": 996873.0,
      "step": 540
    },
    {
      "entropy": 6.88533844947815,
      "epoch": 0.4560669456066946,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.572,
      "mean_token_accuracy": 0.097062898427248,
      "num_tokens": 1006487.0,
      "step": 545
    },
    {
      "entropy": 6.632162094116211,
      "epoch": 0.4602510460251046,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002745,
      "loss": 6.4142,
      "mean_token_accuracy": 0.10585015416145324,
      "num_tokens": 1015715.0,
      "step": 550
    },
    {
      "entropy": 6.738191556930542,
      "epoch": 0.46443514644351463,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000277,
      "loss": 6.53,
      "mean_token_accuracy": 0.10212068408727645,
      "num_tokens": 1025091.0,
      "step": 555
    },
    {
      "entropy": 6.789957094192505,
      "epoch": 0.4686192468619247,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002795,
      "loss": 6.4542,
      "mean_token_accuracy": 0.10204173475503922,
      "num_tokens": 1034174.0,
      "step": 560
    },
    {
      "entropy": 6.847775793075561,
      "epoch": 0.47280334728033474,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6568,
      "mean_token_accuracy": 0.08913957327604294,
      "num_tokens": 1043777.0,
      "step": 565
    },
    {
      "entropy": 6.742847347259522,
      "epoch": 0.4769874476987448,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002845,
      "loss": 6.495,
      "mean_token_accuracy": 0.09928980246186256,
      "num_tokens": 1052749.0,
      "step": 570
    },
    {
      "entropy": 6.8090441703796385,
      "epoch": 0.4811715481171548,
      "grad_norm": 1.125,
      "learning_rate": 0.000287,
      "loss": 6.5077,
      "mean_token_accuracy": 0.10130611211061477,
      "num_tokens": 1061564.0,
      "step": 575
    },
    {
      "entropy": 6.75211615562439,
      "epoch": 0.48535564853556484,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002895,
      "loss": 6.447,
      "mean_token_accuracy": 0.10412005931138993,
      "num_tokens": 1070006.0,
      "step": 580
    },
    {
      "entropy": 6.732285690307617,
      "epoch": 0.4895397489539749,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000292,
      "loss": 6.4958,
      "mean_token_accuracy": 0.10480489581823349,
      "num_tokens": 1079799.0,
      "step": 585
    },
    {
      "entropy": 6.671957015991211,
      "epoch": 0.49372384937238495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002945,
      "loss": 6.3577,
      "mean_token_accuracy": 0.10448934435844422,
      "num_tokens": 1088842.0,
      "step": 590
    },
    {
      "entropy": 6.6517304420471195,
      "epoch": 0.497907949790795,
      "grad_norm": 1.265625,
      "learning_rate": 0.000297,
      "loss": 6.4098,
      "mean_token_accuracy": 0.11377439573407173,
      "num_tokens": 1097960.0,
      "step": 595
    },
    {
      "entropy": 6.644801044464112,
      "epoch": 0.502092050209205,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002995,
      "loss": 6.3789,
      "mean_token_accuracy": 0.10490157902240753,
      "num_tokens": 1106797.0,
      "step": 600
    },
    {
      "entropy": 6.7212605476379395,
      "epoch": 0.5062761506276151,
      "grad_norm": 1.375,
      "learning_rate": 0.000302,
      "loss": 6.5517,
      "mean_token_accuracy": 0.10722804665565491,
      "num_tokens": 1116384.0,
      "step": 605
    },
    {
      "entropy": 6.467281103134155,
      "epoch": 0.5104602510460251,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003045,
      "loss": 6.313,
      "mean_token_accuracy": 0.11083850488066674,
      "num_tokens": 1125206.0,
      "step": 610
    },
    {
      "entropy": 6.815744638442993,
      "epoch": 0.5146443514644351,
      "grad_norm": 1.203125,
      "learning_rate": 0.000307,
      "loss": 6.538,
      "mean_token_accuracy": 0.10858194977045059,
      "num_tokens": 1134110.0,
      "step": 615
    },
    {
      "entropy": 6.790111780166626,
      "epoch": 0.5188284518828452,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003095,
      "loss": 6.529,
      "mean_token_accuracy": 0.10404339656233788,
      "num_tokens": 1142833.0,
      "step": 620
    },
    {
      "entropy": 6.61668643951416,
      "epoch": 0.5230125523012552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000312,
      "loss": 6.4423,
      "mean_token_accuracy": 0.10611013397574424,
      "num_tokens": 1152006.0,
      "step": 625
    },
    {
      "entropy": 6.617235422134399,
      "epoch": 0.5271966527196653,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003145,
      "loss": 6.3947,
      "mean_token_accuracy": 0.10517996922135353,
      "num_tokens": 1161061.0,
      "step": 630
    },
    {
      "entropy": 6.674160814285278,
      "epoch": 0.5313807531380753,
      "grad_norm": 1.09375,
      "learning_rate": 0.000317,
      "loss": 6.412,
      "mean_token_accuracy": 0.10514549165964127,
      "num_tokens": 1170754.0,
      "step": 635
    },
    {
      "entropy": 6.587571859359741,
      "epoch": 0.5355648535564853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003195,
      "loss": 6.3534,
      "mean_token_accuracy": 0.11051966845989228,
      "num_tokens": 1180854.0,
      "step": 640
    },
    {
      "entropy": 6.644711542129516,
      "epoch": 0.5397489539748954,
      "grad_norm": 1.171875,
      "learning_rate": 0.000322,
      "loss": 6.4216,
      "mean_token_accuracy": 0.10841243639588356,
      "num_tokens": 1190567.0,
      "step": 645
    },
    {
      "entropy": 6.525138902664184,
      "epoch": 0.5439330543933054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.2638,
      "mean_token_accuracy": 0.11364024132490158,
      "num_tokens": 1200247.0,
      "step": 650
    },
    {
      "entropy": 6.562865495681763,
      "epoch": 0.5481171548117155,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.3777,
      "mean_token_accuracy": 0.11130299195647239,
      "num_tokens": 1209358.0,
      "step": 655
    },
    {
      "entropy": 6.570280265808106,
      "epoch": 0.5523012552301255,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.3962,
      "mean_token_accuracy": 0.11054785400629044,
      "num_tokens": 1217746.0,
      "step": 660
    },
    {
      "entropy": 6.588881635665894,
      "epoch": 0.5564853556485355,
      "grad_norm": 1.25,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.3589,
      "mean_token_accuracy": 0.11272291392087937,
      "num_tokens": 1226829.0,
      "step": 665
    },
    {
      "entropy": 6.451903820037842,
      "epoch": 0.5606694560669456,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.2812,
      "mean_token_accuracy": 0.11265203654766083,
      "num_tokens": 1236642.0,
      "step": 670
    },
    {
      "entropy": 6.679233074188232,
      "epoch": 0.5648535564853556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000337,
      "loss": 6.4083,
      "mean_token_accuracy": 0.11789099648594856,
      "num_tokens": 1244934.0,
      "step": 675
    },
    {
      "entropy": 6.4409881114959715,
      "epoch": 0.5690376569037657,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003395,
      "loss": 6.2919,
      "mean_token_accuracy": 0.10770596191287041,
      "num_tokens": 1254786.0,
      "step": 680
    },
    {
      "entropy": 6.535488128662109,
      "epoch": 0.5732217573221757,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000342,
      "loss": 6.2508,
      "mean_token_accuracy": 0.11365505158901215,
      "num_tokens": 1263745.0,
      "step": 685
    },
    {
      "entropy": 6.47140793800354,
      "epoch": 0.5774058577405857,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3904,
      "mean_token_accuracy": 0.11604329645633697,
      "num_tokens": 1273101.0,
      "step": 690
    },
    {
      "entropy": 6.5898118019104,
      "epoch": 0.5815899581589958,
      "grad_norm": 1.203125,
      "learning_rate": 0.000347,
      "loss": 6.3736,
      "mean_token_accuracy": 0.11040713787078857,
      "num_tokens": 1282547.0,
      "step": 695
    },
    {
      "entropy": 6.532029438018799,
      "epoch": 0.5857740585774058,
      "grad_norm": 1.125,
      "learning_rate": 0.0003495,
      "loss": 6.4008,
      "mean_token_accuracy": 0.10778395533561706,
      "num_tokens": 1291594.0,
      "step": 700
    },
    {
      "entropy": 6.550088500976562,
      "epoch": 0.5899581589958159,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000352,
      "loss": 6.316,
      "mean_token_accuracy": 0.12029297426342964,
      "num_tokens": 1300282.0,
      "step": 705
    },
    {
      "entropy": 6.516840314865112,
      "epoch": 0.5941422594142259,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003545,
      "loss": 6.3885,
      "mean_token_accuracy": 0.11583921313285828,
      "num_tokens": 1309304.0,
      "step": 710
    },
    {
      "entropy": 6.46544771194458,
      "epoch": 0.5983263598326359,
      "grad_norm": 1.390625,
      "learning_rate": 0.000357,
      "loss": 6.2321,
      "mean_token_accuracy": 0.11730972602963448,
      "num_tokens": 1317638.0,
      "step": 715
    },
    {
      "entropy": 6.597035646438599,
      "epoch": 0.602510460251046,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003595,
      "loss": 6.4257,
      "mean_token_accuracy": 0.11083399206399917,
      "num_tokens": 1327242.0,
      "step": 720
    },
    {
      "entropy": 6.4532746315002445,
      "epoch": 0.606694560669456,
      "grad_norm": 1.03125,
      "learning_rate": 0.000362,
      "loss": 6.2428,
      "mean_token_accuracy": 0.1156072735786438,
      "num_tokens": 1335880.0,
      "step": 725
    },
    {
      "entropy": 6.484941577911377,
      "epoch": 0.6108786610878661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003645,
      "loss": 6.3266,
      "mean_token_accuracy": 0.11275639981031418,
      "num_tokens": 1344909.0,
      "step": 730
    },
    {
      "entropy": 6.437925052642822,
      "epoch": 0.6150627615062761,
      "grad_norm": 1.078125,
      "learning_rate": 0.000367,
      "loss": 6.1876,
      "mean_token_accuracy": 0.11960602849721909,
      "num_tokens": 1353997.0,
      "step": 735
    },
    {
      "entropy": 6.4058027267456055,
      "epoch": 0.6192468619246861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003695,
      "loss": 6.3435,
      "mean_token_accuracy": 0.11241159439086915,
      "num_tokens": 1363293.0,
      "step": 740
    },
    {
      "entropy": 6.570994853973389,
      "epoch": 0.6234309623430963,
      "grad_norm": 1.234375,
      "learning_rate": 0.000372,
      "loss": 6.3573,
      "mean_token_accuracy": 0.11127854585647583,
      "num_tokens": 1372174.0,
      "step": 745
    },
    {
      "entropy": 6.500179719924927,
      "epoch": 0.6276150627615062,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003745,
      "loss": 6.2551,
      "mean_token_accuracy": 0.11865946277976036,
      "num_tokens": 1381067.0,
      "step": 750
    },
    {
      "entropy": 6.397150754928589,
      "epoch": 0.6317991631799164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000377,
      "loss": 6.2352,
      "mean_token_accuracy": 0.12134608998894691,
      "num_tokens": 1390072.0,
      "step": 755
    },
    {
      "entropy": 6.405498886108399,
      "epoch": 0.6359832635983264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003795,
      "loss": 6.2241,
      "mean_token_accuracy": 0.12313714846968651,
      "num_tokens": 1399391.0,
      "step": 760
    },
    {
      "entropy": 6.417546701431275,
      "epoch": 0.6401673640167364,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000382,
      "loss": 6.232,
      "mean_token_accuracy": 0.11985773444175721,
      "num_tokens": 1408054.0,
      "step": 765
    },
    {
      "entropy": 6.511100482940674,
      "epoch": 0.6443514644351465,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003845,
      "loss": 6.2528,
      "mean_token_accuracy": 0.11063949167728424,
      "num_tokens": 1417406.0,
      "step": 770
    },
    {
      "entropy": 6.335081768035889,
      "epoch": 0.6485355648535565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3041,
      "mean_token_accuracy": 0.11019203439354897,
      "num_tokens": 1426907.0,
      "step": 775
    },
    {
      "entropy": 6.429039239883423,
      "epoch": 0.6527196652719666,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3593,
      "mean_token_accuracy": 0.11048192381858826,
      "num_tokens": 1436085.0,
      "step": 780
    },
    {
      "entropy": 6.4894092082977295,
      "epoch": 0.6569037656903766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.2572,
      "mean_token_accuracy": 0.11703520268201828,
      "num_tokens": 1444823.0,
      "step": 785
    },
    {
      "entropy": 6.4137073993682865,
      "epoch": 0.6610878661087866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3072,
      "mean_token_accuracy": 0.1172407552599907,
      "num_tokens": 1454621.0,
      "step": 790
    },
    {
      "entropy": 6.3564558029174805,
      "epoch": 0.6652719665271967,
      "grad_norm": 1.125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.1842,
      "mean_token_accuracy": 0.11996964290738106,
      "num_tokens": 1463784.0,
      "step": 795
    },
    {
      "entropy": 6.409063911437988,
      "epoch": 0.6694560669456067,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003995,
      "loss": 6.3005,
      "mean_token_accuracy": 0.11776863262057305,
      "num_tokens": 1473257.0,
      "step": 800
    },
    {
      "entropy": 6.424042558670044,
      "epoch": 0.6736401673640168,
      "grad_norm": 1.140625,
      "learning_rate": 0.000402,
      "loss": 6.2477,
      "mean_token_accuracy": 0.11706730797886848,
      "num_tokens": 1482195.0,
      "step": 805
    },
    {
      "entropy": 6.349983596801758,
      "epoch": 0.6778242677824268,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004045,
      "loss": 6.286,
      "mean_token_accuracy": 0.11897709146142006,
      "num_tokens": 1491590.0,
      "step": 810
    },
    {
      "entropy": 6.305568790435791,
      "epoch": 0.6820083682008368,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.161,
      "mean_token_accuracy": 0.12807343378663064,
      "num_tokens": 1500208.0,
      "step": 815
    },
    {
      "entropy": 6.405850791931153,
      "epoch": 0.6861924686192469,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004095,
      "loss": 6.2892,
      "mean_token_accuracy": 0.11845047548413276,
      "num_tokens": 1509412.0,
      "step": 820
    },
    {
      "entropy": 6.440489625930786,
      "epoch": 0.6903765690376569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000412,
      "loss": 6.2768,
      "mean_token_accuracy": 0.11962562501430511,
      "num_tokens": 1518426.0,
      "step": 825
    },
    {
      "entropy": 6.419729852676392,
      "epoch": 0.694560669456067,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004145,
      "loss": 6.2377,
      "mean_token_accuracy": 0.11658703610301017,
      "num_tokens": 1528605.0,
      "step": 830
    },
    {
      "entropy": 6.269686937332153,
      "epoch": 0.698744769874477,
      "grad_norm": 1.28125,
      "learning_rate": 0.000417,
      "loss": 6.1758,
      "mean_token_accuracy": 0.12392851561307908,
      "num_tokens": 1537449.0,
      "step": 835
    },
    {
      "entropy": 6.348258590698242,
      "epoch": 0.702928870292887,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195,
      "loss": 6.215,
      "mean_token_accuracy": 0.12626431211829187,
      "num_tokens": 1546743.0,
      "step": 840
    },
    {
      "entropy": 6.275034379959107,
      "epoch": 0.7071129707112971,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422,
      "loss": 6.213,
      "mean_token_accuracy": 0.11712396070361138,
      "num_tokens": 1555691.0,
      "step": 845
    },
    {
      "entropy": 6.30648365020752,
      "epoch": 0.7112970711297071,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004245,
      "loss": 6.1749,
      "mean_token_accuracy": 0.12647855207324027,
      "num_tokens": 1564957.0,
      "step": 850
    },
    {
      "entropy": 6.4026590347290036,
      "epoch": 0.7154811715481172,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000427,
      "loss": 6.2381,
      "mean_token_accuracy": 0.1252881944179535,
      "num_tokens": 1573901.0,
      "step": 855
    },
    {
      "entropy": 6.268061876296997,
      "epoch": 0.7196652719665272,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004295,
      "loss": 6.2031,
      "mean_token_accuracy": 0.12226112931966782,
      "num_tokens": 1582644.0,
      "step": 860
    },
    {
      "entropy": 6.437113666534424,
      "epoch": 0.7238493723849372,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000432,
      "loss": 6.3427,
      "mean_token_accuracy": 0.11745756193995475,
      "num_tokens": 1592363.0,
      "step": 865
    },
    {
      "entropy": 6.349365139007569,
      "epoch": 0.7280334728033473,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004345,
      "loss": 6.2791,
      "mean_token_accuracy": 0.12225733995437622,
      "num_tokens": 1602012.0,
      "step": 870
    },
    {
      "entropy": 6.268002462387085,
      "epoch": 0.7322175732217573,
      "grad_norm": 1.046875,
      "learning_rate": 0.000437,
      "loss": 6.1856,
      "mean_token_accuracy": 0.12651218548417092,
      "num_tokens": 1611625.0,
      "step": 875
    },
    {
      "entropy": 6.227978944778442,
      "epoch": 0.7364016736401674,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004395,
      "loss": 6.1431,
      "mean_token_accuracy": 0.11568873450160026,
      "num_tokens": 1620941.0,
      "step": 880
    },
    {
      "entropy": 6.319770097732544,
      "epoch": 0.7405857740585774,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000442,
      "loss": 6.2285,
      "mean_token_accuracy": 0.11750153079628944,
      "num_tokens": 1629225.0,
      "step": 885
    },
    {
      "entropy": 6.305209541320801,
      "epoch": 0.7447698744769874,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004445,
      "loss": 6.2069,
      "mean_token_accuracy": 0.11938620209693909,
      "num_tokens": 1639551.0,
      "step": 890
    },
    {
      "entropy": 6.2366736888885494,
      "epoch": 0.7489539748953975,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447,
      "loss": 6.1918,
      "mean_token_accuracy": 0.11552652940154076,
      "num_tokens": 1648963.0,
      "step": 895
    },
    {
      "entropy": 6.2946977615356445,
      "epoch": 0.7531380753138075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.1081,
      "mean_token_accuracy": 0.12913521379232407,
      "num_tokens": 1657735.0,
      "step": 900
    },
    {
      "entropy": 6.3238130569458,
      "epoch": 0.7573221757322176,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.155,
      "mean_token_accuracy": 0.12631697207689285,
      "num_tokens": 1666349.0,
      "step": 905
    },
    {
      "entropy": 6.150327157974243,
      "epoch": 0.7615062761506276,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.0795,
      "mean_token_accuracy": 0.126066704839468,
      "num_tokens": 1675920.0,
      "step": 910
    },
    {
      "entropy": 6.217988586425781,
      "epoch": 0.7656903765690377,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.113,
      "mean_token_accuracy": 0.11986384242773056,
      "num_tokens": 1685626.0,
      "step": 915
    },
    {
      "entropy": 6.319421195983887,
      "epoch": 0.7698744769874477,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.2357,
      "mean_token_accuracy": 0.13009796813130378,
      "num_tokens": 1694972.0,
      "step": 920
    },
    {
      "entropy": 6.216734409332275,
      "epoch": 0.7740585774058577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000462,
      "loss": 6.1308,
      "mean_token_accuracy": 0.13374294117093086,
      "num_tokens": 1703806.0,
      "step": 925
    },
    {
      "entropy": 6.2389215469360355,
      "epoch": 0.7782426778242678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004645,
      "loss": 6.2127,
      "mean_token_accuracy": 0.12826964035630226,
      "num_tokens": 1712949.0,
      "step": 930
    },
    {
      "entropy": 6.404123306274414,
      "epoch": 0.7824267782426778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000467,
      "loss": 6.3299,
      "mean_token_accuracy": 0.125592914968729,
      "num_tokens": 1722328.0,
      "step": 935
    },
    {
      "entropy": 6.350027418136596,
      "epoch": 0.7866108786610879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004695,
      "loss": 6.3068,
      "mean_token_accuracy": 0.12012889161705971,
      "num_tokens": 1731880.0,
      "step": 940
    },
    {
      "entropy": 6.140142631530762,
      "epoch": 0.7907949790794979,
      "grad_norm": 1.171875,
      "learning_rate": 0.000472,
      "loss": 6.1248,
      "mean_token_accuracy": 0.12643283903598784,
      "num_tokens": 1740719.0,
      "step": 945
    },
    {
      "entropy": 6.273271703720093,
      "epoch": 0.7949790794979079,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004745,
      "loss": 6.1513,
      "mean_token_accuracy": 0.12553820610046387,
      "num_tokens": 1749753.0,
      "step": 950
    },
    {
      "entropy": 6.292899465560913,
      "epoch": 0.799163179916318,
      "grad_norm": 1.15625,
      "learning_rate": 0.000477,
      "loss": 6.1807,
      "mean_token_accuracy": 0.12436754852533341,
      "num_tokens": 1758913.0,
      "step": 955
    },
    {
      "entropy": 6.273108959197998,
      "epoch": 0.803347280334728,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004795,
      "loss": 6.1882,
      "mean_token_accuracy": 0.12275903448462486,
      "num_tokens": 1768536.0,
      "step": 960
    },
    {
      "entropy": 6.233821058273316,
      "epoch": 0.8075313807531381,
      "grad_norm": 1.140625,
      "learning_rate": 0.000482,
      "loss": 6.1241,
      "mean_token_accuracy": 0.12568357735872268,
      "num_tokens": 1777291.0,
      "step": 965
    },
    {
      "entropy": 6.211701774597168,
      "epoch": 0.8117154811715481,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004845,
      "loss": 6.2163,
      "mean_token_accuracy": 0.12532426416873932,
      "num_tokens": 1787186.0,
      "step": 970
    },
    {
      "entropy": 6.19328761100769,
      "epoch": 0.8158995815899581,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000487,
      "loss": 6.0235,
      "mean_token_accuracy": 0.1305873580276966,
      "num_tokens": 1795581.0,
      "step": 975
    },
    {
      "entropy": 6.08319125175476,
      "epoch": 0.8200836820083682,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004895,
      "loss": 6.0153,
      "mean_token_accuracy": 0.13362710550427437,
      "num_tokens": 1804627.0,
      "step": 980
    },
    {
      "entropy": 6.261023092269897,
      "epoch": 0.8242677824267782,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000492,
      "loss": 6.1549,
      "mean_token_accuracy": 0.1278303623199463,
      "num_tokens": 1813808.0,
      "step": 985
    },
    {
      "entropy": 6.165553712844849,
      "epoch": 0.8284518828451883,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004945,
      "loss": 6.1557,
      "mean_token_accuracy": 0.13012527599930762,
      "num_tokens": 1823422.0,
      "step": 990
    },
    {
      "entropy": 6.19823989868164,
      "epoch": 0.8326359832635983,
      "grad_norm": 1.125,
      "learning_rate": 0.000497,
      "loss": 6.1752,
      "mean_token_accuracy": 0.13083416000008583,
      "num_tokens": 1832437.0,
      "step": 995
    },
    {
      "entropy": 6.2002424716949465,
      "epoch": 0.8368200836820083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995,
      "loss": 6.1044,
      "mean_token_accuracy": 0.12464204803109169,
      "num_tokens": 1842063.0,
      "step": 1000
    },
    {
      "epoch": 0.8368200836820083,
      "eval_entropy": 6.097238668391066,
      "eval_loss": 6.15334939956665,
      "eval_mean_token_accuracy": 0.1348609975815307,
      "eval_num_tokens": 1842063.0,
      "eval_runtime": 2.7384,
      "eval_samples_per_second": 1426.743,
      "eval_steps_per_second": 178.571,
      "step": 1000
    },
    {
      "entropy": 6.241647672653198,
      "epoch": 0.8410041841004184,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999998518355666,
      "loss": 6.2485,
      "mean_token_accuracy": 0.12185878977179528,
      "num_tokens": 1851043.0,
      "step": 1005
    },
    {
      "entropy": 6.126685523986817,
      "epoch": 0.8451882845188284,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999992499178905,
      "loss": 6.0931,
      "mean_token_accuracy": 0.13160960227251053,
      "num_tokens": 1860268.0,
      "step": 1010
    },
    {
      "entropy": 6.303120422363281,
      "epoch": 0.8493723849372385,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999981849879322,
      "loss": 6.2042,
      "mean_token_accuracy": 0.1309683345258236,
      "num_tokens": 1870096.0,
      "step": 1015
    },
    {
      "entropy": 6.130255317687988,
      "epoch": 0.8535564853556485,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999966570478833,
      "loss": 6.0849,
      "mean_token_accuracy": 0.1364796683192253,
      "num_tokens": 1879164.0,
      "step": 1020
    },
    {
      "entropy": 6.1024699211120605,
      "epoch": 0.8577405857740585,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999946661008878,
      "loss": 6.0328,
      "mean_token_accuracy": 0.13583800122141837,
      "num_tokens": 1888135.0,
      "step": 1025
    },
    {
      "entropy": 6.158143806457519,
      "epoch": 0.8619246861924686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499992212151043,
      "loss": 6.1092,
      "mean_token_accuracy": 0.1269300736486912,
      "num_tokens": 1898302.0,
      "step": 1030
    },
    {
      "entropy": 6.110213136672973,
      "epoch": 0.8661087866108786,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999892952033987,
      "loss": 6.0712,
      "mean_token_accuracy": 0.129842671751976,
      "num_tokens": 1907753.0,
      "step": 1035
    },
    {
      "entropy": 6.154698228836059,
      "epoch": 0.8702928870292888,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999859152639573,
      "loss": 6.1328,
      "mean_token_accuracy": 0.12858494743704796,
      "num_tokens": 1917543.0,
      "step": 1040
    },
    {
      "entropy": 6.2287651062011715,
      "epoch": 0.8744769874476988,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999820723396745,
      "loss": 6.1718,
      "mean_token_accuracy": 0.12609437778592109,
      "num_tokens": 1927335.0,
      "step": 1045
    },
    {
      "entropy": 6.19956407546997,
      "epoch": 0.8786610878661087,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999777664384581,
      "loss": 6.0957,
      "mean_token_accuracy": 0.13153462409973143,
      "num_tokens": 1936381.0,
      "step": 1050
    },
    {
      "entropy": 6.174496364593506,
      "epoch": 0.8828451882845189,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999729975691692,
      "loss": 6.1341,
      "mean_token_accuracy": 0.12876357659697532,
      "num_tokens": 1945559.0,
      "step": 1055
    },
    {
      "entropy": 6.041715145111084,
      "epoch": 0.8870292887029289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999677657416212,
      "loss": 6.0012,
      "mean_token_accuracy": 0.12970590144395827,
      "num_tokens": 1954709.0,
      "step": 1060
    },
    {
      "entropy": 6.168726396560669,
      "epoch": 0.891213389121339,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999620709665804,
      "loss": 6.1249,
      "mean_token_accuracy": 0.1281352959573269,
      "num_tokens": 1963664.0,
      "step": 1065
    },
    {
      "entropy": 6.181927251815796,
      "epoch": 0.895397489539749,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999559132557659,
      "loss": 6.171,
      "mean_token_accuracy": 0.12771156281232834,
      "num_tokens": 1974103.0,
      "step": 1070
    },
    {
      "entropy": 6.23676552772522,
      "epoch": 0.899581589958159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999492926218489,
      "loss": 6.0457,
      "mean_token_accuracy": 0.1328116737306118,
      "num_tokens": 1982924.0,
      "step": 1075
    },
    {
      "entropy": 6.003208255767822,
      "epoch": 0.9037656903765691,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499942209078454,
      "loss": 5.973,
      "mean_token_accuracy": 0.14368802681565285,
      "num_tokens": 1992033.0,
      "step": 1080
    },
    {
      "entropy": 6.048687314987182,
      "epoch": 0.9079497907949791,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999346626401578,
      "loss": 5.9894,
      "mean_token_accuracy": 0.13765854164958,
      "num_tokens": 2001218.0,
      "step": 1085
    },
    {
      "entropy": 6.094419670104981,
      "epoch": 0.9121338912133892,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999266533224897,
      "loss": 6.1441,
      "mean_token_accuracy": 0.13423838838934898,
      "num_tokens": 2010463.0,
      "step": 1090
    },
    {
      "entropy": 6.1051229476928714,
      "epoch": 0.9163179916317992,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004999181811419314,
      "loss": 6.0181,
      "mean_token_accuracy": 0.13275936618447304,
      "num_tokens": 2019246.0,
      "step": 1095
    },
    {
      "entropy": 6.170205354690552,
      "epoch": 0.9205020920502092,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999092461159177,
      "loss": 6.0876,
      "mean_token_accuracy": 0.12886958941817284,
      "num_tokens": 2028083.0,
      "step": 1100
    },
    {
      "entropy": 6.076777601242066,
      "epoch": 0.9246861924686193,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998998482628351,
      "loss": 5.9235,
      "mean_token_accuracy": 0.1374969571828842,
      "num_tokens": 2036631.0,
      "step": 1105
    },
    {
      "entropy": 6.035492610931397,
      "epoch": 0.9288702928870293,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004998899876020229,
      "loss": 6.0327,
      "mean_token_accuracy": 0.133451097458601,
      "num_tokens": 2045910.0,
      "step": 1110
    },
    {
      "entropy": 6.021814250946045,
      "epoch": 0.9330543933054394,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499879664153773,
      "loss": 6.0868,
      "mean_token_accuracy": 0.14059085845947267,
      "num_tokens": 2055959.0,
      "step": 1115
    },
    {
      "entropy": 6.29487099647522,
      "epoch": 0.9372384937238494,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998688779393291,
      "loss": 6.1577,
      "mean_token_accuracy": 0.12911083772778512,
      "num_tokens": 2065475.0,
      "step": 1120
    },
    {
      "entropy": 5.898234081268311,
      "epoch": 0.9414225941422594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998576289808877,
      "loss": 5.9211,
      "mean_token_accuracy": 0.1388360470533371,
      "num_tokens": 2074469.0,
      "step": 1125
    },
    {
      "entropy": 6.0572669982910154,
      "epoch": 0.9456066945606695,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998459173015972,
      "loss": 6.0961,
      "mean_token_accuracy": 0.13587369546294212,
      "num_tokens": 2084604.0,
      "step": 1130
    },
    {
      "entropy": 6.207867860794067,
      "epoch": 0.9497907949790795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998337429255584,
      "loss": 6.0656,
      "mean_token_accuracy": 0.1341245450079441,
      "num_tokens": 2093377.0,
      "step": 1135
    },
    {
      "entropy": 6.09445858001709,
      "epoch": 0.9539748953974896,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998211058778243,
      "loss": 6.0838,
      "mean_token_accuracy": 0.13606315553188325,
      "num_tokens": 2102798.0,
      "step": 1140
    },
    {
      "entropy": 6.12096939086914,
      "epoch": 0.9581589958158996,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998080061843999,
      "loss": 6.0464,
      "mean_token_accuracy": 0.1335694245994091,
      "num_tokens": 2112516.0,
      "step": 1145
    },
    {
      "entropy": 6.146814584732056,
      "epoch": 0.9623430962343096,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499794443872242,
      "loss": 6.1417,
      "mean_token_accuracy": 0.13098908141255378,
      "num_tokens": 2121677.0,
      "step": 1150
    },
    {
      "entropy": 6.00798225402832,
      "epoch": 0.9665271966527197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997804189692601,
      "loss": 5.971,
      "mean_token_accuracy": 0.1396520145237446,
      "num_tokens": 2131220.0,
      "step": 1155
    },
    {
      "entropy": 6.118230247497559,
      "epoch": 0.9707112970711297,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997659315043149,
      "loss": 6.1345,
      "mean_token_accuracy": 0.13085383027791977,
      "num_tokens": 2142104.0,
      "step": 1160
    },
    {
      "entropy": 6.051174783706665,
      "epoch": 0.9748953974895398,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997509815072191,
      "loss": 6.0492,
      "mean_token_accuracy": 0.13568617030978203,
      "num_tokens": 2151970.0,
      "step": 1165
    },
    {
      "entropy": 6.027523326873779,
      "epoch": 0.9790794979079498,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997355690087379,
      "loss": 6.0072,
      "mean_token_accuracy": 0.1357894979417324,
      "num_tokens": 2161371.0,
      "step": 1170
    },
    {
      "entropy": 6.045231628417969,
      "epoch": 0.9832635983263598,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997196940405874,
      "loss": 5.9485,
      "mean_token_accuracy": 0.13663906380534172,
      "num_tokens": 2170650.0,
      "step": 1175
    },
    {
      "entropy": 5.984046506881714,
      "epoch": 0.9874476987447699,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997033566354358,
      "loss": 6.028,
      "mean_token_accuracy": 0.1392426609992981,
      "num_tokens": 2180934.0,
      "step": 1180
    },
    {
      "entropy": 6.154821681976318,
      "epoch": 0.9916317991631799,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996865568269028,
      "loss": 5.9669,
      "mean_token_accuracy": 0.14529061168432236,
      "num_tokens": 2189894.0,
      "step": 1185
    },
    {
      "entropy": 5.945505809783936,
      "epoch": 0.99581589958159,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996692946495597,
      "loss": 5.9377,
      "mean_token_accuracy": 0.14892658293247224,
      "num_tokens": 2200208.0,
      "step": 1190
    },
    {
      "entropy": 6.064693832397461,
      "epoch": 1.0,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996515701389295,
      "loss": 5.9726,
      "mean_token_accuracy": 0.13724684566259385,
      "num_tokens": 2209322.0,
      "step": 1195
    },
    {
      "entropy": 5.993635749816894,
      "epoch": 1.00418410041841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996333833314862,
      "loss": 5.7905,
      "mean_token_accuracy": 0.15274174511432648,
      "num_tokens": 2217663.0,
      "step": 1200
    },
    {
      "entropy": 5.90445466041565,
      "epoch": 1.00836820083682,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996147342646554,
      "loss": 5.8433,
      "mean_token_accuracy": 0.14284583255648614,
      "num_tokens": 2226559.0,
      "step": 1205
    },
    {
      "entropy": 6.067541551589966,
      "epoch": 1.0125523012552302,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995956229768137,
      "loss": 5.8828,
      "mean_token_accuracy": 0.14996479004621505,
      "num_tokens": 2235468.0,
      "step": 1210
    },
    {
      "entropy": 5.908719444274903,
      "epoch": 1.0167364016736402,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995760495072893,
      "loss": 5.6674,
      "mean_token_accuracy": 0.15197395384311677,
      "num_tokens": 2245176.0,
      "step": 1215
    },
    {
      "entropy": 5.926908683776856,
      "epoch": 1.0209205020920502,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995560138963611,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1469671368598938,
      "num_tokens": 2255123.0,
      "step": 1220
    },
    {
      "entropy": 5.958147192001343,
      "epoch": 1.0251046025104602,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995355161852593,
      "loss": 5.7525,
      "mean_token_accuracy": 0.1474505454301834,
      "num_tokens": 2263596.0,
      "step": 1225
    },
    {
      "entropy": 5.980195045471191,
      "epoch": 1.0292887029288702,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995145564161648,
      "loss": 5.8006,
      "mean_token_accuracy": 0.13994376361370087,
      "num_tokens": 2272228.0,
      "step": 1230
    },
    {
      "entropy": 5.907618856430053,
      "epoch": 1.0334728033472804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994931346322094,
      "loss": 5.8431,
      "mean_token_accuracy": 0.1445181243121624,
      "num_tokens": 2281248.0,
      "step": 1235
    },
    {
      "entropy": 6.041008806228637,
      "epoch": 1.0376569037656904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994712508774757,
      "loss": 5.8114,
      "mean_token_accuracy": 0.1452713206410408,
      "num_tokens": 2290958.0,
      "step": 1240
    },
    {
      "entropy": 5.83619909286499,
      "epoch": 1.0418410041841004,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499448905196997,
      "loss": 5.7584,
      "mean_token_accuracy": 0.15415944904088974,
      "num_tokens": 2299763.0,
      "step": 1245
    },
    {
      "entropy": 5.935490417480469,
      "epoch": 1.0460251046025104,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994260976367573,
      "loss": 5.7788,
      "mean_token_accuracy": 0.15245575234293937,
      "num_tokens": 2309847.0,
      "step": 1250
    },
    {
      "entropy": 5.983281230926513,
      "epoch": 1.0502092050209204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994028282436905,
      "loss": 5.8069,
      "mean_token_accuracy": 0.14806725829839706,
      "num_tokens": 2317556.0,
      "step": 1255
    },
    {
      "entropy": 5.781259489059448,
      "epoch": 1.0543933054393306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993790970656814,
      "loss": 5.7374,
      "mean_token_accuracy": 0.1501047745347023,
      "num_tokens": 2327151.0,
      "step": 1260
    },
    {
      "entropy": 5.952362155914306,
      "epoch": 1.0585774058577406,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993549041515652,
      "loss": 5.7418,
      "mean_token_accuracy": 0.1499510034918785,
      "num_tokens": 2336119.0,
      "step": 1265
    },
    {
      "entropy": 5.803351783752442,
      "epoch": 1.0627615062761506,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993302495511268,
      "loss": 5.7755,
      "mean_token_accuracy": 0.14605644345283508,
      "num_tokens": 2345262.0,
      "step": 1270
    },
    {
      "entropy": 6.077183580398559,
      "epoch": 1.0669456066945606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993051333151014,
      "loss": 5.8956,
      "mean_token_accuracy": 0.14342452213168144,
      "num_tokens": 2355210.0,
      "step": 1275
    },
    {
      "entropy": 5.791180896759033,
      "epoch": 1.0711297071129706,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992795554951742,
      "loss": 5.6401,
      "mean_token_accuracy": 0.15315188989043235,
      "num_tokens": 2364599.0,
      "step": 1280
    },
    {
      "entropy": 5.763968133926392,
      "epoch": 1.0753138075313808,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992535161439803,
      "loss": 5.8054,
      "mean_token_accuracy": 0.1445068120956421,
      "num_tokens": 2374143.0,
      "step": 1285
    },
    {
      "entropy": 6.005795478820801,
      "epoch": 1.0794979079497908,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992270153151045,
      "loss": 5.8188,
      "mean_token_accuracy": 0.14890599846839905,
      "num_tokens": 2383155.0,
      "step": 1290
    },
    {
      "entropy": 6.011566638946533,
      "epoch": 1.0836820083682008,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992000530630812,
      "loss": 5.8663,
      "mean_token_accuracy": 0.14473781883716583,
      "num_tokens": 2392542.0,
      "step": 1295
    },
    {
      "entropy": 5.806177711486816,
      "epoch": 1.0878661087866108,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991726294433943,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1449222221970558,
      "num_tokens": 2401851.0,
      "step": 1300
    },
    {
      "entropy": 5.932667779922485,
      "epoch": 1.0920502092050208,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991447445124774,
      "loss": 5.7593,
      "mean_token_accuracy": 0.149945068359375,
      "num_tokens": 2410726.0,
      "step": 1305
    },
    {
      "entropy": 5.812711238861084,
      "epoch": 1.096234309623431,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499116398327713,
      "loss": 5.8305,
      "mean_token_accuracy": 0.14005152210593225,
      "num_tokens": 2420702.0,
      "step": 1310
    },
    {
      "entropy": 5.843758773803711,
      "epoch": 1.100418410041841,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990875909474331,
      "loss": 5.7499,
      "mean_token_accuracy": 0.14834965020418167,
      "num_tokens": 2431335.0,
      "step": 1315
    },
    {
      "entropy": 5.966901779174805,
      "epoch": 1.104602510460251,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990583224309187,
      "loss": 5.7261,
      "mean_token_accuracy": 0.15352486222982406,
      "num_tokens": 2440722.0,
      "step": 1320
    },
    {
      "entropy": 5.845820617675781,
      "epoch": 1.108786610878661,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990285928383997,
      "loss": 5.8035,
      "mean_token_accuracy": 0.14780879467725755,
      "num_tokens": 2451230.0,
      "step": 1325
    },
    {
      "entropy": 5.904784488677978,
      "epoch": 1.112970711297071,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989984022310545,
      "loss": 5.8394,
      "mean_token_accuracy": 0.14220331013202667,
      "num_tokens": 2461442.0,
      "step": 1330
    },
    {
      "entropy": 5.87468147277832,
      "epoch": 1.1171548117154813,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989677506710109,
      "loss": 5.6527,
      "mean_token_accuracy": 0.1563212588429451,
      "num_tokens": 2471381.0,
      "step": 1335
    },
    {
      "entropy": 5.808247089385986,
      "epoch": 1.1213389121338913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989366382213449,
      "loss": 5.7156,
      "mean_token_accuracy": 0.15480270683765412,
      "num_tokens": 2479937.0,
      "step": 1340
    },
    {
      "entropy": 5.837823104858399,
      "epoch": 1.1255230125523012,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989050649460806,
      "loss": 5.6731,
      "mean_token_accuracy": 0.1569208934903145,
      "num_tokens": 2488974.0,
      "step": 1345
    },
    {
      "entropy": 5.856475782394409,
      "epoch": 1.1297071129707112,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498873030910191,
      "loss": 5.8444,
      "mean_token_accuracy": 0.1466200530529022,
      "num_tokens": 2498800.0,
      "step": 1350
    },
    {
      "entropy": 5.822557353973389,
      "epoch": 1.1338912133891212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988405361795969,
      "loss": 5.7363,
      "mean_token_accuracy": 0.14498689994215966,
      "num_tokens": 2506580.0,
      "step": 1355
    },
    {
      "entropy": 5.84963550567627,
      "epoch": 1.1380753138075315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988075808211672,
      "loss": 5.7934,
      "mean_token_accuracy": 0.15228895395994185,
      "num_tokens": 2516342.0,
      "step": 1360
    },
    {
      "entropy": 5.809200716018677,
      "epoch": 1.1422594142259415,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004987741649027189,
      "loss": 5.8174,
      "mean_token_accuracy": 0.14777714088559152,
      "num_tokens": 2526054.0,
      "step": 1365
    },
    {
      "entropy": 5.951186895370483,
      "epoch": 1.1464435146443515,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987402884930163,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1517642021179199,
      "num_tokens": 2535556.0,
      "step": 1370
    },
    {
      "entropy": 5.828822660446167,
      "epoch": 1.1506276150627615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987059516617717,
      "loss": 5.7459,
      "mean_token_accuracy": 0.1545324891805649,
      "num_tokens": 2544930.0,
      "step": 1375
    },
    {
      "entropy": 5.7893500328063965,
      "epoch": 1.1548117154811715,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986711544796449,
      "loss": 5.6609,
      "mean_token_accuracy": 0.1535905510187149,
      "num_tokens": 2554362.0,
      "step": 1380
    },
    {
      "entropy": 5.884280490875244,
      "epoch": 1.1589958158995817,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498635897018243,
      "loss": 5.777,
      "mean_token_accuracy": 0.14912986755371094,
      "num_tokens": 2563499.0,
      "step": 1385
    },
    {
      "entropy": 5.8048481941223145,
      "epoch": 1.1631799163179917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986001793501198,
      "loss": 5.7883,
      "mean_token_accuracy": 0.15240272358059884,
      "num_tokens": 2573346.0,
      "step": 1390
    },
    {
      "entropy": 5.857279062271118,
      "epoch": 1.1673640167364017,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985640015487768,
      "loss": 5.665,
      "mean_token_accuracy": 0.15898251682519912,
      "num_tokens": 2582541.0,
      "step": 1395
    },
    {
      "entropy": 5.832502317428589,
      "epoch": 1.1715481171548117,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985273636886621,
      "loss": 5.7536,
      "mean_token_accuracy": 0.14916445165872574,
      "num_tokens": 2592015.0,
      "step": 1400
    },
    {
      "entropy": 5.822238969802856,
      "epoch": 1.1757322175732217,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004984902658451704,
      "loss": 5.7545,
      "mean_token_accuracy": 0.15448953062295914,
      "num_tokens": 2601998.0,
      "step": 1405
    },
    {
      "entropy": 5.775497055053711,
      "epoch": 1.1799163179916319,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984527080946432,
      "loss": 5.6909,
      "mean_token_accuracy": 0.15859137922525407,
      "num_tokens": 2611435.0,
      "step": 1410
    },
    {
      "entropy": 5.902001571655274,
      "epoch": 1.1841004184100419,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984146905143683,
      "loss": 5.7934,
      "mean_token_accuracy": 0.14567320495843888,
      "num_tokens": 2621639.0,
      "step": 1415
    },
    {
      "entropy": 5.837963247299195,
      "epoch": 1.1882845188284519,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983762131825797,
      "loss": 5.7495,
      "mean_token_accuracy": 0.1509477600455284,
      "num_tokens": 2631390.0,
      "step": 1420
    },
    {
      "entropy": 5.777963781356812,
      "epoch": 1.1924686192468619,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983372761784576,
      "loss": 5.7968,
      "mean_token_accuracy": 0.14694754034280777,
      "num_tokens": 2640769.0,
      "step": 1425
    },
    {
      "entropy": 5.9535966396331785,
      "epoch": 1.196652719665272,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004982978795821281,
      "loss": 5.7937,
      "mean_token_accuracy": 0.1512495271861553,
      "num_tokens": 2651137.0,
      "step": 1430
    },
    {
      "entropy": 5.794545030593872,
      "epoch": 1.200836820083682,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982580234746632,
      "loss": 5.7258,
      "mean_token_accuracy": 0.1513078287243843,
      "num_tokens": 2660530.0,
      "step": 1435
    },
    {
      "entropy": 5.867940950393677,
      "epoch": 1.205020920502092,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049821770793808,
      "loss": 5.784,
      "mean_token_accuracy": 0.1505849204957485,
      "num_tokens": 2669541.0,
      "step": 1440
    },
    {
      "entropy": 5.860204744338989,
      "epoch": 1.209205020920502,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981769330553418,
      "loss": 5.7822,
      "mean_token_accuracy": 0.1534332662820816,
      "num_tokens": 2678542.0,
      "step": 1445
    },
    {
      "entropy": 5.87125940322876,
      "epoch": 1.213389121338912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004981356989103566,
      "loss": 5.8384,
      "mean_token_accuracy": 0.15294507145881653,
      "num_tokens": 2687720.0,
      "step": 1450
    },
    {
      "entropy": 5.862550973892212,
      "epoch": 1.217573221757322,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980940055879774,
      "loss": 5.698,
      "mean_token_accuracy": 0.15199038982391358,
      "num_tokens": 2697641.0,
      "step": 1455
    },
    {
      "entropy": 5.753113889694214,
      "epoch": 1.2217573221757323,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980518531740027,
      "loss": 5.7504,
      "mean_token_accuracy": 0.14872822910547256,
      "num_tokens": 2706600.0,
      "step": 1460
    },
    {
      "entropy": 5.950168657302856,
      "epoch": 1.2259414225941423,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980092417551752,
      "loss": 5.8755,
      "mean_token_accuracy": 0.14688518047332763,
      "num_tokens": 2716633.0,
      "step": 1465
    },
    {
      "entropy": 5.817790365219116,
      "epoch": 1.2301255230125523,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004979661714191825,
      "loss": 5.6416,
      "mean_token_accuracy": 0.1666289746761322,
      "num_tokens": 2724829.0,
      "step": 1470
    },
    {
      "entropy": 5.805331134796143,
      "epoch": 1.2343096234309623,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979226422546563,
      "loss": 5.7465,
      "mean_token_accuracy": 0.15683723539113997,
      "num_tokens": 2734276.0,
      "step": 1475
    },
    {
      "entropy": 5.913634300231934,
      "epoch": 1.2384937238493725,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978786543511726,
      "loss": 5.8622,
      "mean_token_accuracy": 0.15280352383852006,
      "num_tokens": 2744413.0,
      "step": 1480
    },
    {
      "entropy": 5.847559309005737,
      "epoch": 1.2426778242677825,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978342077992515,
      "loss": 5.7394,
      "mean_token_accuracy": 0.1537095919251442,
      "num_tokens": 2753242.0,
      "step": 1485
    },
    {
      "entropy": 5.650951862335205,
      "epoch": 1.2468619246861925,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004977893026903568,
      "loss": 5.6605,
      "mean_token_accuracy": 0.16106596887111663,
      "num_tokens": 2761811.0,
      "step": 1490
    },
    {
      "entropy": 5.895610046386719,
      "epoch": 1.2510460251046025,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977439391168963,
      "loss": 5.7547,
      "mean_token_accuracy": 0.1545099973678589,
      "num_tokens": 2770560.0,
      "step": 1495
    },
    {
      "entropy": 5.8653045177459715,
      "epoch": 1.2552301255230125,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004976981171722205,
      "loss": 5.7531,
      "mean_token_accuracy": 0.15552559345960618,
      "num_tokens": 2780131.0,
      "step": 1500
    },
    {
      "epoch": 1.2552301255230125,
      "eval_entropy": 5.61207090565032,
      "eval_loss": 5.8904619216918945,
      "eval_mean_token_accuracy": 0.15481723342205117,
      "eval_num_tokens": 2780131.0,
      "eval_runtime": 2.9514,
      "eval_samples_per_second": 1323.781,
      "eval_steps_per_second": 165.684,
      "step": 1500
    },
    {
      "entropy": 5.848963212966919,
      "epoch": 1.2594142259414225,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976518369506237,
      "loss": 5.8545,
      "mean_token_accuracy": 0.1463423416018486,
      "num_tokens": 2789319.0,
      "step": 1505
    },
    {
      "entropy": 5.7695282936096195,
      "epoch": 1.2635983263598327,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976050985473433,
      "loss": 5.7321,
      "mean_token_accuracy": 0.14833602607250213,
      "num_tokens": 2799253.0,
      "step": 1510
    },
    {
      "entropy": 5.928028345108032,
      "epoch": 1.2677824267782427,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975579020585593,
      "loss": 5.8401,
      "mean_token_accuracy": 0.14651586264371871,
      "num_tokens": 2808968.0,
      "step": 1515
    },
    {
      "entropy": 5.814293909072876,
      "epoch": 1.2719665271966527,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004975102475813947,
      "loss": 5.7177,
      "mean_token_accuracy": 0.15377511978149414,
      "num_tokens": 2817282.0,
      "step": 1520
    },
    {
      "entropy": 5.87192554473877,
      "epoch": 1.2761506276150627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974621352139144,
      "loss": 5.7843,
      "mean_token_accuracy": 0.14812919795513152,
      "num_tokens": 2826395.0,
      "step": 1525
    },
    {
      "entropy": 5.8476917266845705,
      "epoch": 1.280334728033473,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974135650551262,
      "loss": 5.865,
      "mean_token_accuracy": 0.14753438383340836,
      "num_tokens": 2836195.0,
      "step": 1530
    },
    {
      "entropy": 5.853739261627197,
      "epoch": 1.284518828451883,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973645372049796,
      "loss": 5.693,
      "mean_token_accuracy": 0.15756383240222932,
      "num_tokens": 2846020.0,
      "step": 1535
    },
    {
      "entropy": 5.688941192626953,
      "epoch": 1.288702928870293,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973150517643659,
      "loss": 5.6256,
      "mean_token_accuracy": 0.16124313473701476,
      "num_tokens": 2854895.0,
      "step": 1540
    },
    {
      "entropy": 5.844737005233765,
      "epoch": 1.292887029288703,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972651088351185,
      "loss": 5.6901,
      "mean_token_accuracy": 0.15637971609830856,
      "num_tokens": 2864491.0,
      "step": 1545
    },
    {
      "entropy": 5.811354351043701,
      "epoch": 1.297071129707113,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004972147085200116,
      "loss": 5.6969,
      "mean_token_accuracy": 0.1557008296251297,
      "num_tokens": 2873962.0,
      "step": 1550
    },
    {
      "entropy": 5.800867891311645,
      "epoch": 1.301255230125523,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971638509227611,
      "loss": 5.7301,
      "mean_token_accuracy": 0.15785083174705505,
      "num_tokens": 2883530.0,
      "step": 1555
    },
    {
      "entropy": 5.771515417098999,
      "epoch": 1.3054393305439331,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004971125361480236,
      "loss": 5.7266,
      "mean_token_accuracy": 0.151897095143795,
      "num_tokens": 2892867.0,
      "step": 1560
    },
    {
      "entropy": 5.756432056427002,
      "epoch": 1.3096234309623431,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970607643013969,
      "loss": 5.6874,
      "mean_token_accuracy": 0.15923002511262893,
      "num_tokens": 2902320.0,
      "step": 1565
    },
    {
      "entropy": 5.777465343475342,
      "epoch": 1.3138075313807531,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497008535489419,
      "loss": 5.6413,
      "mean_token_accuracy": 0.15751131474971772,
      "num_tokens": 2911390.0,
      "step": 1570
    },
    {
      "entropy": 5.701142072677612,
      "epoch": 1.3179916317991631,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004969558498195684,
      "loss": 5.673,
      "mean_token_accuracy": 0.16534920781850815,
      "num_tokens": 2921295.0,
      "step": 1575
    },
    {
      "entropy": 5.815394020080566,
      "epoch": 1.3221757322175733,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004969027074002638,
      "loss": 5.6874,
      "mean_token_accuracy": 0.15951785892248155,
      "num_tokens": 2931328.0,
      "step": 1580
    },
    {
      "entropy": 5.713910436630249,
      "epoch": 1.3263598326359833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004968491083408636,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1613614335656166,
      "num_tokens": 2940193.0,
      "step": 1585
    },
    {
      "entropy": 5.683044862747193,
      "epoch": 1.3305439330543933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004967950527516663,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15876813232898712,
      "num_tokens": 2949759.0,
      "step": 1590
    },
    {
      "entropy": 5.708407974243164,
      "epoch": 1.3347280334728033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004967405407439092,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1553261622786522,
      "num_tokens": 2959185.0,
      "step": 1595
    },
    {
      "entropy": 5.833631801605224,
      "epoch": 1.3389121338912133,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004966855724297698,
      "loss": 5.7455,
      "mean_token_accuracy": 0.15250789225101472,
      "num_tokens": 2968945.0,
      "step": 1600
    },
    {
      "entropy": 5.736705303192139,
      "epoch": 1.3430962343096233,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004966301479223637,
      "loss": 5.6967,
      "mean_token_accuracy": 0.1556132599711418,
      "num_tokens": 2978395.0,
      "step": 1605
    },
    {
      "entropy": 5.739750719070434,
      "epoch": 1.3472803347280335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965742673357456,
      "loss": 5.6131,
      "mean_token_accuracy": 0.16823979020118712,
      "num_tokens": 2987030.0,
      "step": 1610
    },
    {
      "entropy": 5.7740395069122314,
      "epoch": 1.3514644351464435,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004965179307849088,
      "loss": 5.6609,
      "mean_token_accuracy": 0.15987352132797242,
      "num_tokens": 2995736.0,
      "step": 1615
    },
    {
      "entropy": 5.8161561489105225,
      "epoch": 1.3556485355648535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000496461138385785,
      "loss": 5.7754,
      "mean_token_accuracy": 0.156988662481308,
      "num_tokens": 3005105.0,
      "step": 1620
    },
    {
      "entropy": 5.745142889022827,
      "epoch": 1.3598326359832635,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004964038902552438,
      "loss": 5.6924,
      "mean_token_accuracy": 0.15291614979505538,
      "num_tokens": 3013857.0,
      "step": 1625
    },
    {
      "entropy": 5.775795555114746,
      "epoch": 1.3640167364016738,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004963461865110926,
      "loss": 5.7171,
      "mean_token_accuracy": 0.15699103474617004,
      "num_tokens": 3023972.0,
      "step": 1630
    },
    {
      "entropy": 5.664225387573242,
      "epoch": 1.3682008368200838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004962880272720764,
      "loss": 5.6885,
      "mean_token_accuracy": 0.16199921816587448,
      "num_tokens": 3032705.0,
      "step": 1635
    },
    {
      "entropy": 5.7547321796417235,
      "epoch": 1.3723849372384938,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004962294126578775,
      "loss": 5.62,
      "mean_token_accuracy": 0.16171107590198516,
      "num_tokens": 3040900.0,
      "step": 1640
    },
    {
      "entropy": 5.74275279045105,
      "epoch": 1.3765690376569037,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961703427891154,
      "loss": 5.6599,
      "mean_token_accuracy": 0.16257089227437974,
      "num_tokens": 3050477.0,
      "step": 1645
    },
    {
      "entropy": 5.7742781162261965,
      "epoch": 1.3807531380753137,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961108177873466,
      "loss": 5.7237,
      "mean_token_accuracy": 0.154131980240345,
      "num_tokens": 3059743.0,
      "step": 1650
    },
    {
      "entropy": 5.735147619247437,
      "epoch": 1.3849372384937237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004960508377750638,
      "loss": 5.5306,
      "mean_token_accuracy": 0.16080360263586044,
      "num_tokens": 3067801.0,
      "step": 1655
    },
    {
      "entropy": 5.600426530838012,
      "epoch": 1.389121338912134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959904028756961,
      "loss": 5.7173,
      "mean_token_accuracy": 0.15346099585294723,
      "num_tokens": 3077404.0,
      "step": 1660
    },
    {
      "entropy": 5.864295959472656,
      "epoch": 1.393305439330544,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004959295132136091,
      "loss": 5.7488,
      "mean_token_accuracy": 0.1578315958380699,
      "num_tokens": 3087130.0,
      "step": 1665
    },
    {
      "entropy": 5.866291618347168,
      "epoch": 1.397489539748954,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958681689141036,
      "loss": 5.7664,
      "mean_token_accuracy": 0.15363284721970558,
      "num_tokens": 3096038.0,
      "step": 1670
    },
    {
      "entropy": 5.603582286834717,
      "epoch": 1.401673640167364,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004958063701034167,
      "loss": 5.5974,
      "mean_token_accuracy": 0.1607895866036415,
      "num_tokens": 3105668.0,
      "step": 1675
    },
    {
      "entropy": 5.711364984512329,
      "epoch": 1.4058577405857742,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049574411690872,
      "loss": 5.5904,
      "mean_token_accuracy": 0.15727734118700026,
      "num_tokens": 3114020.0,
      "step": 1680
    },
    {
      "entropy": 5.642215585708618,
      "epoch": 1.4100418410041842,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004956814094581208,
      "loss": 5.6182,
      "mean_token_accuracy": 0.16320676058530809,
      "num_tokens": 3124096.0,
      "step": 1685
    },
    {
      "entropy": 5.794946718215942,
      "epoch": 1.4142259414225942,
      "grad_norm": 1.015625,
      "learning_rate": 0.000495618247880661,
      "loss": 5.6859,
      "mean_token_accuracy": 0.16073822230100632,
      "num_tokens": 3132806.0,
      "step": 1690
    },
    {
      "entropy": 5.671970987319947,
      "epoch": 1.4184100418410042,
      "grad_norm": 1.0,
      "learning_rate": 0.0004955546323063168,
      "loss": 5.6278,
      "mean_token_accuracy": 0.16975785344839095,
      "num_tokens": 3142001.0,
      "step": 1695
    },
    {
      "entropy": 5.647975635528565,
      "epoch": 1.4225941422594142,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954905628659989,
      "loss": 5.5844,
      "mean_token_accuracy": 0.1559284970164299,
      "num_tokens": 3151658.0,
      "step": 1700
    },
    {
      "entropy": 5.718328285217285,
      "epoch": 1.4267782426778242,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004954260396915521,
      "loss": 5.6258,
      "mean_token_accuracy": 0.16052106022834778,
      "num_tokens": 3159573.0,
      "step": 1705
    },
    {
      "entropy": 5.544895315170288,
      "epoch": 1.4309623430962344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004953610629157544,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16683861315250398,
      "num_tokens": 3168312.0,
      "step": 1710
    },
    {
      "entropy": 5.742662715911865,
      "epoch": 1.4351464435146444,
      "grad_norm": 1.125,
      "learning_rate": 0.0004952956326723177,
      "loss": 5.5968,
      "mean_token_accuracy": 0.16999766528606414,
      "num_tokens": 3176811.0,
      "step": 1715
    },
    {
      "entropy": 5.708902883529663,
      "epoch": 1.4393305439330544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000495229749095887,
      "loss": 5.685,
      "mean_token_accuracy": 0.15340494439005853,
      "num_tokens": 3186599.0,
      "step": 1720
    },
    {
      "entropy": 5.844608926773072,
      "epoch": 1.4435146443514644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049516341232204,
      "loss": 5.7892,
      "mean_token_accuracy": 0.15978376120328902,
      "num_tokens": 3195905.0,
      "step": 1725
    },
    {
      "entropy": 5.724213314056397,
      "epoch": 1.4476987447698746,
      "grad_norm": 1.078125,
      "learning_rate": 0.000495096622487287,
      "loss": 5.75,
      "mean_token_accuracy": 0.1553723022341728,
      "num_tokens": 3205132.0,
      "step": 1730
    },
    {
      "entropy": 5.88250732421875,
      "epoch": 1.4518828451882846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495029379729071,
      "loss": 5.7631,
      "mean_token_accuracy": 0.14813076555728913,
      "num_tokens": 3214362.0,
      "step": 1735
    },
    {
      "entropy": 5.781597042083741,
      "epoch": 1.4560669456066946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004949616841857666,
      "loss": 5.7411,
      "mean_token_accuracy": 0.15504500269889832,
      "num_tokens": 3224248.0,
      "step": 1740
    },
    {
      "entropy": 5.69829797744751,
      "epoch": 1.4602510460251046,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004948935359966806,
      "loss": 5.61,
      "mean_token_accuracy": 0.1609645113348961,
      "num_tokens": 3233416.0,
      "step": 1745
    },
    {
      "entropy": 5.607308721542358,
      "epoch": 1.4644351464435146,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004948249353020506,
      "loss": 5.5365,
      "mean_token_accuracy": 0.16924193501472473,
      "num_tokens": 3242354.0,
      "step": 1750
    },
    {
      "entropy": 5.732311677932739,
      "epoch": 1.4686192468619246,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947558822430461,
      "loss": 5.6397,
      "mean_token_accuracy": 0.16035355031490325,
      "num_tokens": 3251026.0,
      "step": 1755
    },
    {
      "entropy": 5.692071390151978,
      "epoch": 1.4728033472803348,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004946863769617672,
      "loss": 5.5944,
      "mean_token_accuracy": 0.16337772607803344,
      "num_tokens": 3259429.0,
      "step": 1760
    },
    {
      "entropy": 5.646625089645386,
      "epoch": 1.4769874476987448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004946164196012445,
      "loss": 5.6404,
      "mean_token_accuracy": 0.15840144455432892,
      "num_tokens": 3269305.0,
      "step": 1765
    },
    {
      "entropy": 5.740614318847657,
      "epoch": 1.4811715481171548,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945460103054389,
      "loss": 5.7426,
      "mean_token_accuracy": 0.14881804287433625,
      "num_tokens": 3278846.0,
      "step": 1770
    },
    {
      "entropy": 5.732423925399781,
      "epoch": 1.4853556485355648,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004944751492192416,
      "loss": 5.65,
      "mean_token_accuracy": 0.16647016108036042,
      "num_tokens": 3287200.0,
      "step": 1775
    },
    {
      "entropy": 5.745698308944702,
      "epoch": 1.489539748953975,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004944038364884734,
      "loss": 5.6791,
      "mean_token_accuracy": 0.15847390592098237,
      "num_tokens": 3296651.0,
      "step": 1780
    },
    {
      "entropy": 5.708291864395141,
      "epoch": 1.493723849372385,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004943320722598842,
      "loss": 5.541,
      "mean_token_accuracy": 0.16429150104522705,
      "num_tokens": 3304887.0,
      "step": 1785
    },
    {
      "entropy": 5.684462928771973,
      "epoch": 1.497907949790795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004942598566811532,
      "loss": 5.5646,
      "mean_token_accuracy": 0.16510989218950273,
      "num_tokens": 3313925.0,
      "step": 1790
    },
    {
      "entropy": 5.6262242794036865,
      "epoch": 1.502092050209205,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004941871899008887,
      "loss": 5.5645,
      "mean_token_accuracy": 0.16308156549930572,
      "num_tokens": 3322943.0,
      "step": 1795
    },
    {
      "entropy": 5.638538360595703,
      "epoch": 1.506276150627615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000494114072068627,
      "loss": 5.5779,
      "mean_token_accuracy": 0.16292459666728973,
      "num_tokens": 3332742.0,
      "step": 1800
    },
    {
      "entropy": 5.693442869186401,
      "epoch": 1.510460251046025,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004940405033348332,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15975773185491562,
      "num_tokens": 3342228.0,
      "step": 1805
    },
    {
      "entropy": 5.666954040527344,
      "epoch": 1.514644351464435,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939664838508996,
      "loss": 5.6349,
      "mean_token_accuracy": 0.15982257425785065,
      "num_tokens": 3351854.0,
      "step": 1810
    },
    {
      "entropy": 5.664078140258789,
      "epoch": 1.5188284518828452,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938920137691465,
      "loss": 5.6191,
      "mean_token_accuracy": 0.16421182453632355,
      "num_tokens": 3360809.0,
      "step": 1815
    },
    {
      "entropy": 5.664334726333618,
      "epoch": 1.5230125523012552,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938170932428214,
      "loss": 5.6452,
      "mean_token_accuracy": 0.15385862588882446,
      "num_tokens": 3371033.0,
      "step": 1820
    },
    {
      "entropy": 5.738883924484253,
      "epoch": 1.5271966527196654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004937417224260984,
      "loss": 5.5209,
      "mean_token_accuracy": 0.16514809876680375,
      "num_tokens": 3379943.0,
      "step": 1825
    },
    {
      "entropy": 5.57040958404541,
      "epoch": 1.5313807531380754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004936659014740791,
      "loss": 5.6499,
      "mean_token_accuracy": 0.1651262253522873,
      "num_tokens": 3389316.0,
      "step": 1830
    },
    {
      "entropy": 5.77739405632019,
      "epoch": 1.5355648535564854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004935896305427904,
      "loss": 5.6779,
      "mean_token_accuracy": 0.153752800822258,
      "num_tokens": 3398914.0,
      "step": 1835
    },
    {
      "entropy": 5.85550889968872,
      "epoch": 1.5397489539748954,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004935129097891857,
      "loss": 5.6735,
      "mean_token_accuracy": 0.15657827109098435,
      "num_tokens": 3407967.0,
      "step": 1840
    },
    {
      "entropy": 5.7320945262908936,
      "epoch": 1.5439330543933054,
      "grad_norm": 1.15625,
      "learning_rate": 0.000493435739371144,
      "loss": 5.6904,
      "mean_token_accuracy": 0.15912245512008666,
      "num_tokens": 3416530.0,
      "step": 1845
    },
    {
      "entropy": 5.685615587234497,
      "epoch": 1.5481171548117154,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004933581194474695,
      "loss": 5.5558,
      "mean_token_accuracy": 0.16052745580673217,
      "num_tokens": 3425762.0,
      "step": 1850
    },
    {
      "entropy": 5.632355833053589,
      "epoch": 1.5523012552301254,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004932800501778918,
      "loss": 5.6357,
      "mean_token_accuracy": 0.15861250311136246,
      "num_tokens": 3434799.0,
      "step": 1855
    },
    {
      "entropy": 5.721937561035157,
      "epoch": 1.5564853556485354,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004932015317230646,
      "loss": 5.6407,
      "mean_token_accuracy": 0.1599887117743492,
      "num_tokens": 3442986.0,
      "step": 1860
    },
    {
      "entropy": 5.6735435962677006,
      "epoch": 1.5606694560669456,
      "grad_norm": 1.125,
      "learning_rate": 0.0004931225642445665,
      "loss": 5.6696,
      "mean_token_accuracy": 0.16079186052083969,
      "num_tokens": 3452914.0,
      "step": 1865
    },
    {
      "entropy": 5.717952060699463,
      "epoch": 1.5648535564853556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004930431479048998,
      "loss": 5.5173,
      "mean_token_accuracy": 0.17034173756837845,
      "num_tokens": 3462093.0,
      "step": 1870
    },
    {
      "entropy": 5.666427803039551,
      "epoch": 1.5690376569037658,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004929632828674904,
      "loss": 5.7166,
      "mean_token_accuracy": 0.15459873676300048,
      "num_tokens": 3471259.0,
      "step": 1875
    },
    {
      "entropy": 5.682976150512696,
      "epoch": 1.5732217573221758,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004928829692966879,
      "loss": 5.6492,
      "mean_token_accuracy": 0.16574902534484864,
      "num_tokens": 3480071.0,
      "step": 1880
    },
    {
      "entropy": 5.791259860992431,
      "epoch": 1.5774058577405858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928022073577646,
      "loss": 5.6762,
      "mean_token_accuracy": 0.16466853469610215,
      "num_tokens": 3489075.0,
      "step": 1885
    },
    {
      "entropy": 5.68497748374939,
      "epoch": 1.5815899581589958,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004927209972169157,
      "loss": 5.4978,
      "mean_token_accuracy": 0.16591176837682725,
      "num_tokens": 3498495.0,
      "step": 1890
    },
    {
      "entropy": 5.634211444854737,
      "epoch": 1.5857740585774058,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004926393390412586,
      "loss": 5.6434,
      "mean_token_accuracy": 0.15969491004943848,
      "num_tokens": 3508590.0,
      "step": 1895
    },
    {
      "entropy": 5.74862232208252,
      "epoch": 1.5899581589958158,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004925572329988325,
      "loss": 5.6574,
      "mean_token_accuracy": 0.15974607914686204,
      "num_tokens": 3518397.0,
      "step": 1900
    },
    {
      "entropy": 5.652823257446289,
      "epoch": 1.5941422594142258,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004924746792585987,
      "loss": 5.6367,
      "mean_token_accuracy": 0.16384051144123077,
      "num_tokens": 3526620.0,
      "step": 1905
    },
    {
      "entropy": 5.743967056274414,
      "epoch": 1.5983263598326358,
      "grad_norm": 1.125,
      "learning_rate": 0.0004923916779904394,
      "loss": 5.6546,
      "mean_token_accuracy": 0.163092640042305,
      "num_tokens": 3535534.0,
      "step": 1910
    },
    {
      "entropy": 5.571524715423584,
      "epoch": 1.602510460251046,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004923082293651578,
      "loss": 5.6151,
      "mean_token_accuracy": 0.15783242136240005,
      "num_tokens": 3544771.0,
      "step": 1915
    },
    {
      "entropy": 5.698504400253296,
      "epoch": 1.606694560669456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492224333554478,
      "loss": 5.5663,
      "mean_token_accuracy": 0.1609413132071495,
      "num_tokens": 3553970.0,
      "step": 1920
    },
    {
      "entropy": 5.621961355209351,
      "epoch": 1.6108786610878663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004921399907310437,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16530968993902206,
      "num_tokens": 3563116.0,
      "step": 1925
    },
    {
      "entropy": 5.569674491882324,
      "epoch": 1.6150627615062763,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004920552010684191,
      "loss": 5.6301,
      "mean_token_accuracy": 0.16663024872541427,
      "num_tokens": 3571605.0,
      "step": 1930
    },
    {
      "entropy": 5.673672008514404,
      "epoch": 1.6192468619246863,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919699647410878,
      "loss": 5.7027,
      "mean_token_accuracy": 0.1581038311123848,
      "num_tokens": 3581837.0,
      "step": 1935
    },
    {
      "entropy": 5.702061128616333,
      "epoch": 1.6234309623430963,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004918842819244523,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1579617217183113,
      "num_tokens": 3590853.0,
      "step": 1940
    },
    {
      "entropy": 5.6623131275177006,
      "epoch": 1.6276150627615062,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917981527948342,
      "loss": 5.5433,
      "mean_token_accuracy": 0.16482285112142564,
      "num_tokens": 3599708.0,
      "step": 1945
    },
    {
      "entropy": 5.614315223693848,
      "epoch": 1.6317991631799162,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004917115775294731,
      "loss": 5.5778,
      "mean_token_accuracy": 0.1669729769229889,
      "num_tokens": 3608634.0,
      "step": 1950
    },
    {
      "entropy": 5.6976964473724365,
      "epoch": 1.6359832635983262,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004916245563065273,
      "loss": 5.6369,
      "mean_token_accuracy": 0.16016416847705842,
      "num_tokens": 3618776.0,
      "step": 1955
    },
    {
      "entropy": 5.666608476638794,
      "epoch": 1.6401673640167362,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915370893050722,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15682819336652756,
      "num_tokens": 3628595.0,
      "step": 1960
    },
    {
      "entropy": 5.806681823730469,
      "epoch": 1.6443514644351465,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004914491767051012,
      "loss": 5.7436,
      "mean_token_accuracy": 0.14956202432513238,
      "num_tokens": 3638133.0,
      "step": 1965
    },
    {
      "entropy": 5.697069072723389,
      "epoch": 1.6485355648535565,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004913608186875241,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1624540075659752,
      "num_tokens": 3646679.0,
      "step": 1970
    },
    {
      "entropy": 5.659416246414184,
      "epoch": 1.6527196652719667,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004912720154341674,
      "loss": 5.6739,
      "mean_token_accuracy": 0.15715343728661538,
      "num_tokens": 3656131.0,
      "step": 1975
    },
    {
      "entropy": 5.671878480911255,
      "epoch": 1.6569037656903767,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911827671277739,
      "loss": 5.568,
      "mean_token_accuracy": 0.16296357065439224,
      "num_tokens": 3665187.0,
      "step": 1980
    },
    {
      "entropy": 5.695562028884888,
      "epoch": 1.6610878661087867,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910930739520025,
      "loss": 5.6291,
      "mean_token_accuracy": 0.16535532623529434,
      "num_tokens": 3674649.0,
      "step": 1985
    },
    {
      "entropy": 5.7589911937713625,
      "epoch": 1.6652719665271967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004910029360914271,
      "loss": 5.6536,
      "mean_token_accuracy": 0.15396214723587037,
      "num_tokens": 3684680.0,
      "step": 1990
    },
    {
      "entropy": 5.60883321762085,
      "epoch": 1.6694560669456067,
      "grad_norm": 1.125,
      "learning_rate": 0.000490912353731537,
      "loss": 5.5797,
      "mean_token_accuracy": 0.16498577892780303,
      "num_tokens": 3693560.0,
      "step": 1995
    },
    {
      "entropy": 5.607213926315308,
      "epoch": 1.6736401673640167,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004908213270587363,
      "loss": 5.5851,
      "mean_token_accuracy": 0.1629691630601883,
      "num_tokens": 3702632.0,
      "step": 2000
    },
    {
      "epoch": 1.6736401673640167,
      "eval_entropy": 5.544569926271653,
      "eval_loss": 5.715911388397217,
      "eval_mean_token_accuracy": 0.16552528923335494,
      "eval_num_tokens": 3702632.0,
      "eval_runtime": 2.7368,
      "eval_samples_per_second": 1427.569,
      "eval_steps_per_second": 178.674,
      "step": 2000
    },
    {
      "entropy": 5.604301977157593,
      "epoch": 1.6778242677824267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004907298562603431,
      "loss": 5.4632,
      "mean_token_accuracy": 0.1718360811471939,
      "num_tokens": 3711022.0,
      "step": 2005
    },
    {
      "entropy": 5.5926025390625,
      "epoch": 1.6820083682008367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004906379415245896,
      "loss": 5.5684,
      "mean_token_accuracy": 0.16862087249755858,
      "num_tokens": 3719778.0,
      "step": 2010
    },
    {
      "entropy": 5.590682411193848,
      "epoch": 1.6861924686192469,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004905455830406215,
      "loss": 5.5672,
      "mean_token_accuracy": 0.16277657300233841,
      "num_tokens": 3728773.0,
      "step": 2015
    },
    {
      "entropy": 5.677611303329468,
      "epoch": 1.6903765690376569,
      "grad_norm": 1.125,
      "learning_rate": 0.0004904527809984978,
      "loss": 5.6141,
      "mean_token_accuracy": 0.16766730397939683,
      "num_tokens": 3737781.0,
      "step": 2020
    },
    {
      "entropy": 5.7193811416625975,
      "epoch": 1.694560669456067,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903595355891902,
      "loss": 5.6613,
      "mean_token_accuracy": 0.15416572988033295,
      "num_tokens": 3748010.0,
      "step": 2025
    },
    {
      "entropy": 5.650969839096069,
      "epoch": 1.698744769874477,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004902658470045826,
      "loss": 5.642,
      "mean_token_accuracy": 0.16055549085140228,
      "num_tokens": 3757894.0,
      "step": 2030
    },
    {
      "entropy": 5.753380298614502,
      "epoch": 1.702928870292887,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901717154374713,
      "loss": 5.7388,
      "mean_token_accuracy": 0.15718897730112075,
      "num_tokens": 3768760.0,
      "step": 2035
    },
    {
      "entropy": 5.690996360778809,
      "epoch": 1.707112970711297,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004900771410815638,
      "loss": 5.5709,
      "mean_token_accuracy": 0.16574515402317047,
      "num_tokens": 3778273.0,
      "step": 2040
    },
    {
      "entropy": 5.5101611614227295,
      "epoch": 1.711297071129707,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004899821241314788,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1653672218322754,
      "num_tokens": 3787012.0,
      "step": 2045
    },
    {
      "entropy": 5.649552869796753,
      "epoch": 1.715481171548117,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489886664782746,
      "loss": 5.5247,
      "mean_token_accuracy": 0.1680687978863716,
      "num_tokens": 3795144.0,
      "step": 2050
    },
    {
      "entropy": 5.708580875396729,
      "epoch": 1.719665271966527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897907632318055,
      "loss": 5.7534,
      "mean_token_accuracy": 0.15822094976902007,
      "num_tokens": 3804534.0,
      "step": 2055
    },
    {
      "entropy": 5.641533422470093,
      "epoch": 1.723849372384937,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896944196760071,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1683221310377121,
      "num_tokens": 3813757.0,
      "step": 2060
    },
    {
      "entropy": 5.594599962234497,
      "epoch": 1.7280334728033473,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004895976343136106,
      "loss": 5.5548,
      "mean_token_accuracy": 0.1677917420864105,
      "num_tokens": 3823157.0,
      "step": 2065
    },
    {
      "entropy": 5.538334083557129,
      "epoch": 1.7322175732217573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004895004073437844,
      "loss": 5.4542,
      "mean_token_accuracy": 0.16612266898155212,
      "num_tokens": 3831661.0,
      "step": 2070
    },
    {
      "entropy": 5.541912317276001,
      "epoch": 1.7364016736401675,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004894027389666061,
      "loss": 5.4498,
      "mean_token_accuracy": 0.16812364608049393,
      "num_tokens": 3840591.0,
      "step": 2075
    },
    {
      "entropy": 5.639562273025513,
      "epoch": 1.7405857740585775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004893046293830616,
      "loss": 5.5785,
      "mean_token_accuracy": 0.16422712057828903,
      "num_tokens": 3850094.0,
      "step": 2080
    },
    {
      "entropy": 5.621643018722534,
      "epoch": 1.7447698744769875,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892060787950447,
      "loss": 5.626,
      "mean_token_accuracy": 0.16461298018693923,
      "num_tokens": 3858973.0,
      "step": 2085
    },
    {
      "entropy": 5.559000730514526,
      "epoch": 1.7489539748953975,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004891070874053564,
      "loss": 5.4022,
      "mean_token_accuracy": 0.1763455167412758,
      "num_tokens": 3868315.0,
      "step": 2090
    },
    {
      "entropy": 5.538459062576294,
      "epoch": 1.7531380753138075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004890076554177051,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1631178990006447,
      "num_tokens": 3877605.0,
      "step": 2095
    },
    {
      "entropy": 5.700150108337402,
      "epoch": 1.7573221757322175,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004889077830367061,
      "loss": 5.5782,
      "mean_token_accuracy": 0.16976662129163742,
      "num_tokens": 3886998.0,
      "step": 2100
    },
    {
      "entropy": 5.5663464069366455,
      "epoch": 1.7615062761506275,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004888074704678806,
      "loss": 5.5283,
      "mean_token_accuracy": 0.17066608518362045,
      "num_tokens": 3896018.0,
      "step": 2105
    },
    {
      "entropy": 5.621518468856811,
      "epoch": 1.7656903765690377,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004887067179176558,
      "loss": 5.5343,
      "mean_token_accuracy": 0.1670089140534401,
      "num_tokens": 3904957.0,
      "step": 2110
    },
    {
      "entropy": 5.587416505813598,
      "epoch": 1.7698744769874477,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004886055255933639,
      "loss": 5.6542,
      "mean_token_accuracy": 0.16026381626725197,
      "num_tokens": 3913427.0,
      "step": 2115
    },
    {
      "entropy": 5.667502689361572,
      "epoch": 1.7740585774058577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004885038937032429,
      "loss": 5.5657,
      "mean_token_accuracy": 0.16527705043554305,
      "num_tokens": 3923134.0,
      "step": 2120
    },
    {
      "entropy": 5.622828578948974,
      "epoch": 1.778242677824268,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004884018224564345,
      "loss": 5.4785,
      "mean_token_accuracy": 0.16797400861978531,
      "num_tokens": 3931867.0,
      "step": 2125
    },
    {
      "entropy": 5.58387508392334,
      "epoch": 1.782426778242678,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004882993120629852,
      "loss": 5.5638,
      "mean_token_accuracy": 0.16899107545614242,
      "num_tokens": 3941057.0,
      "step": 2130
    },
    {
      "entropy": 5.744916820526123,
      "epoch": 1.786610878661088,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004881963627338449,
      "loss": 5.6846,
      "mean_token_accuracy": 0.15740710943937303,
      "num_tokens": 3950511.0,
      "step": 2135
    },
    {
      "entropy": 5.63847131729126,
      "epoch": 1.790794979079498,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048809297468086654,
      "loss": 5.581,
      "mean_token_accuracy": 0.1745120495557785,
      "num_tokens": 3959769.0,
      "step": 2140
    },
    {
      "entropy": 5.657891464233399,
      "epoch": 1.794979079497908,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004879891481168063,
      "loss": 5.5191,
      "mean_token_accuracy": 0.1700441509485245,
      "num_tokens": 3968462.0,
      "step": 2145
    },
    {
      "entropy": 5.530794906616211,
      "epoch": 1.799163179916318,
      "grad_norm": 1.125,
      "learning_rate": 0.00048788488325532235,
      "loss": 5.5703,
      "mean_token_accuracy": 0.16803695857524872,
      "num_tokens": 3977464.0,
      "step": 2150
    },
    {
      "entropy": 5.5986074924469,
      "epoch": 1.803347280334728,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004877801803109753,
      "loss": 5.4767,
      "mean_token_accuracy": 0.1752568155527115,
      "num_tokens": 3986303.0,
      "step": 2155
    },
    {
      "entropy": 5.683440685272217,
      "epoch": 1.8075313807531381,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048767503949922683,
      "loss": 5.5264,
      "mean_token_accuracy": 0.16948044002056123,
      "num_tokens": 3994970.0,
      "step": 2160
    },
    {
      "entropy": 5.605940961837769,
      "epoch": 1.8117154811715481,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004875694610364398,
      "loss": 5.6154,
      "mean_token_accuracy": 0.1611637741327286,
      "num_tokens": 4004679.0,
      "step": 2165
    },
    {
      "entropy": 5.728769540786743,
      "epoch": 1.8158995815899581,
      "grad_norm": 1.125,
      "learning_rate": 0.00048746344513987775,
      "loss": 5.6381,
      "mean_token_accuracy": 0.1625831186771393,
      "num_tokens": 4013516.0,
      "step": 2170
    },
    {
      "entropy": 5.613612127304077,
      "epoch": 1.8200836820083683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004873569920277043,
      "loss": 5.5096,
      "mean_token_accuracy": 0.17253974229097366,
      "num_tokens": 4022335.0,
      "step": 2175
    },
    {
      "entropy": 5.533032703399658,
      "epoch": 1.8242677824267783,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048725010191898296,
      "loss": 5.4899,
      "mean_token_accuracy": 0.17203625291585922,
      "num_tokens": 4031295.0,
      "step": 2180
    },
    {
      "entropy": 5.687892150878906,
      "epoch": 1.8284518828451883,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048714277503367627,
      "loss": 5.7066,
      "mean_token_accuracy": 0.1616518810391426,
      "num_tokens": 4040378.0,
      "step": 2185
    },
    {
      "entropy": 5.640391111373901,
      "epoch": 1.8326359832635983,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048703501159264576,
      "loss": 5.553,
      "mean_token_accuracy": 0.15976608246564866,
      "num_tokens": 4050217.0,
      "step": 2190
    },
    {
      "entropy": 5.5045853614807125,
      "epoch": 1.8368200836820083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869268118176512,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16942883282899857,
      "num_tokens": 4059088.0,
      "step": 2195
    },
    {
      "entropy": 5.6252336502075195,
      "epoch": 1.8410041841004183,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048681817593135046,
      "loss": 5.5366,
      "mean_token_accuracy": 0.17682300955057145,
      "num_tokens": 4067609.0,
      "step": 2200
    },
    {
      "entropy": 5.567916822433472,
      "epoch": 1.8451882845188283,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048670910415729855,
      "loss": 5.5195,
      "mean_token_accuracy": 0.17112305611371995,
      "num_tokens": 4076763.0,
      "step": 2205
    },
    {
      "entropy": 5.655111455917359,
      "epoch": 1.8493723849372385,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048659959671994786,
      "loss": 5.6785,
      "mean_token_accuracy": 0.16254908740520477,
      "num_tokens": 4086552.0,
      "step": 2210
    },
    {
      "entropy": 5.494708251953125,
      "epoch": 1.8535564853556485,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048648965384464693,
      "loss": 5.4333,
      "mean_token_accuracy": 0.17247406542301177,
      "num_tokens": 4096200.0,
      "step": 2215
    },
    {
      "entropy": 5.529922199249268,
      "epoch": 1.8577405857740585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048637927575764057,
      "loss": 5.4908,
      "mean_token_accuracy": 0.16739632934331894,
      "num_tokens": 4105080.0,
      "step": 2220
    },
    {
      "entropy": 5.598203134536743,
      "epoch": 1.8619246861924688,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048626846268606913,
      "loss": 5.4519,
      "mean_token_accuracy": 0.17144161313772202,
      "num_tokens": 4113972.0,
      "step": 2225
    },
    {
      "entropy": 5.681439542770386,
      "epoch": 1.8661087866108788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048615721485796803,
      "loss": 5.6933,
      "mean_token_accuracy": 0.15622689872980117,
      "num_tokens": 4123615.0,
      "step": 2230
    },
    {
      "entropy": 5.63415675163269,
      "epoch": 1.8702928870292888,
      "grad_norm": 1.125,
      "learning_rate": 0.0004860455325022676,
      "loss": 5.451,
      "mean_token_accuracy": 0.1640198454260826,
      "num_tokens": 4132128.0,
      "step": 2235
    },
    {
      "entropy": 5.658318138122558,
      "epoch": 1.8744769874476988,
      "grad_norm": 1.109375,
      "learning_rate": 0.000485933415848792,
      "loss": 5.6749,
      "mean_token_accuracy": 0.162701378762722,
      "num_tokens": 4141575.0,
      "step": 2240
    },
    {
      "entropy": 5.6866742134094235,
      "epoch": 1.8786610878661087,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004858208651282593,
      "loss": 5.505,
      "mean_token_accuracy": 0.16838331073522567,
      "num_tokens": 4150372.0,
      "step": 2245
    },
    {
      "entropy": 5.5462789058685305,
      "epoch": 1.8828451882845187,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000485707880572281,
      "loss": 5.6084,
      "mean_token_accuracy": 0.16846281588077544,
      "num_tokens": 4159007.0,
      "step": 2250
    },
    {
      "entropy": 5.625886917114258,
      "epoch": 1.8870292887029287,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004855944624133608,
      "loss": 5.6619,
      "mean_token_accuracy": 0.16254033297300338,
      "num_tokens": 4168646.0,
      "step": 2255
    },
    {
      "entropy": 5.565406274795532,
      "epoch": 1.891213389121339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048548061088489523,
      "loss": 5.5563,
      "mean_token_accuracy": 0.1624855801463127,
      "num_tokens": 4177979.0,
      "step": 2260
    },
    {
      "entropy": 5.653033447265625,
      "epoch": 1.895397489539749,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048536632622117246,
      "loss": 5.4694,
      "mean_token_accuracy": 0.17232209146022798,
      "num_tokens": 4186781.0,
      "step": 2265
    },
    {
      "entropy": 5.540466260910034,
      "epoch": 1.899581589958159,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048525160865737186,
      "loss": 5.5396,
      "mean_token_accuracy": 0.1641465112566948,
      "num_tokens": 4195661.0,
      "step": 2270
    },
    {
      "entropy": 5.586657428741455,
      "epoch": 1.9037656903765692,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048513645842956384,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1706296131014824,
      "num_tokens": 4203902.0,
      "step": 2275
    },
    {
      "entropy": 5.555630874633789,
      "epoch": 1.9079497907949792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048502087577470895,
      "loss": 5.5015,
      "mean_token_accuracy": 0.16890660524368287,
      "num_tokens": 4213663.0,
      "step": 2280
    },
    {
      "entropy": 5.649169540405273,
      "epoch": 1.9121338912133892,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048490486093065774,
      "loss": 5.5524,
      "mean_token_accuracy": 0.16829514801502227,
      "num_tokens": 4222152.0,
      "step": 2285
    },
    {
      "entropy": 5.625352954864502,
      "epoch": 1.9163179916317992,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048478841413615017,
      "loss": 5.572,
      "mean_token_accuracy": 0.17131819427013398,
      "num_tokens": 4232347.0,
      "step": 2290
    },
    {
      "entropy": 5.624868249893188,
      "epoch": 1.9205020920502092,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048467153563081487,
      "loss": 5.6332,
      "mean_token_accuracy": 0.16910638511180878,
      "num_tokens": 4240607.0,
      "step": 2295
    },
    {
      "entropy": 5.633149242401123,
      "epoch": 1.9246861924686192,
      "grad_norm": 1.0,
      "learning_rate": 0.0004845542256551692,
      "loss": 5.6394,
      "mean_token_accuracy": 0.16387657672166825,
      "num_tokens": 4251131.0,
      "step": 2300
    },
    {
      "entropy": 5.637137794494629,
      "epoch": 1.9288702928870292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004844364844506181,
      "loss": 5.5719,
      "mean_token_accuracy": 0.16235824227333068,
      "num_tokens": 4259955.0,
      "step": 2305
    },
    {
      "entropy": 5.679235124588013,
      "epoch": 1.9330543933054394,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004843183122594541,
      "loss": 5.5448,
      "mean_token_accuracy": 0.1709941640496254,
      "num_tokens": 4268686.0,
      "step": 2310
    },
    {
      "entropy": 5.612339782714844,
      "epoch": 1.9372384937238494,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004841997093248566,
      "loss": 5.5195,
      "mean_token_accuracy": 0.17368945777416228,
      "num_tokens": 4277377.0,
      "step": 2315
    },
    {
      "entropy": 5.483291482925415,
      "epoch": 1.9414225941422594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004840806758908913,
      "loss": 5.5151,
      "mean_token_accuracy": 0.16756559759378434,
      "num_tokens": 4287037.0,
      "step": 2320
    },
    {
      "entropy": 5.706357431411743,
      "epoch": 1.9456066945606696,
      "grad_norm": 1.0,
      "learning_rate": 0.00048396121220250983,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1578308656811714,
      "num_tokens": 4296923.0,
      "step": 2325
    },
    {
      "entropy": 5.4903162002563475,
      "epoch": 1.9497907949790796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004838413185055494,
      "loss": 5.4163,
      "mean_token_accuracy": 0.17858784198760985,
      "num_tokens": 4305873.0,
      "step": 2330
    },
    {
      "entropy": 5.615921258926392,
      "epoch": 1.9539748953974896,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004837209950467318,
      "loss": 5.5754,
      "mean_token_accuracy": 0.16469462513923644,
      "num_tokens": 4314912.0,
      "step": 2335
    },
    {
      "entropy": 5.605984258651733,
      "epoch": 1.9581589958158996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004836002420736635,
      "loss": 5.5774,
      "mean_token_accuracy": 0.16062707901000978,
      "num_tokens": 4324606.0,
      "step": 2340
    },
    {
      "entropy": 5.546719932556153,
      "epoch": 1.9623430962343096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004834790598348347,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16764551401138306,
      "num_tokens": 4333076.0,
      "step": 2345
    },
    {
      "entropy": 5.71311993598938,
      "epoch": 1.9665271966527196,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004833574485796188,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1685716599225998,
      "num_tokens": 4341893.0,
      "step": 2350
    },
    {
      "entropy": 5.57413420677185,
      "epoch": 1.9707112970711296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048323540855827244,
      "loss": 5.5458,
      "mean_token_accuracy": 0.16568866670131682,
      "num_tokens": 4352303.0,
      "step": 2355
    },
    {
      "entropy": 5.654368257522583,
      "epoch": 1.9748953974895398,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004831129400219341,
      "loss": 5.5109,
      "mean_token_accuracy": 0.16720473915338516,
      "num_tokens": 4361166.0,
      "step": 2360
    },
    {
      "entropy": 5.690392589569091,
      "epoch": 1.9790794979079498,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048299004322262457,
      "loss": 5.5567,
      "mean_token_accuracy": 0.16406455636024475,
      "num_tokens": 4369999.0,
      "step": 2365
    },
    {
      "entropy": 5.539315795898437,
      "epoch": 1.9832635983263598,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048286671841324565,
      "loss": 5.5161,
      "mean_token_accuracy": 0.17316748797893525,
      "num_tokens": 4379373.0,
      "step": 2370
    },
    {
      "entropy": 5.598080635070801,
      "epoch": 1.98744769874477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004827429658475799,
      "loss": 5.5158,
      "mean_token_accuracy": 0.17036739885807037,
      "num_tokens": 4389035.0,
      "step": 2375
    },
    {
      "entropy": 5.6628663539886475,
      "epoch": 1.99163179916318,
      "grad_norm": 1.0,
      "learning_rate": 0.00048261878578029026,
      "loss": 5.6745,
      "mean_token_accuracy": 0.16741672456264495,
      "num_tokens": 4399366.0,
      "step": 2380
    },
    {
      "entropy": 5.643234777450561,
      "epoch": 1.99581589958159,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004824941784669193,
      "loss": 5.5101,
      "mean_token_accuracy": 0.16699415892362596,
      "num_tokens": 4408183.0,
      "step": 2385
    },
    {
      "entropy": 5.595900821685791,
      "epoch": 2.0,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004823691441638889,
      "loss": 5.6154,
      "mean_token_accuracy": 0.15458238124847412,
      "num_tokens": 4418644.0,
      "step": 2390
    },
    {
      "entropy": 5.52336573600769,
      "epoch": 2.00418410041841,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004822436831284995,
      "loss": 5.2173,
      "mean_token_accuracy": 0.18154313266277314,
      "num_tokens": 4427828.0,
      "step": 2395
    },
    {
      "entropy": 5.49915246963501,
      "epoch": 2.00836820083682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004821177956189297,
      "loss": 5.1385,
      "mean_token_accuracy": 0.18106852173805238,
      "num_tokens": 4436194.0,
      "step": 2400
    },
    {
      "entropy": 5.449331378936767,
      "epoch": 2.01255230125523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004819914818942359,
      "loss": 5.2112,
      "mean_token_accuracy": 0.1826639547944069,
      "num_tokens": 4444468.0,
      "step": 2405
    },
    {
      "entropy": 5.534541130065918,
      "epoch": 2.01673640167364,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048186474221435127,
      "loss": 5.1351,
      "mean_token_accuracy": 0.18929099142551423,
      "num_tokens": 4452760.0,
      "step": 2410
    },
    {
      "entropy": 5.447423124313355,
      "epoch": 2.02092050209205,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048173757684008583,
      "loss": 5.2009,
      "mean_token_accuracy": 0.18149912357330322,
      "num_tokens": 4462154.0,
      "step": 2415
    },
    {
      "entropy": 5.4820966720581055,
      "epoch": 2.0251046025104604,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004816099860331255,
      "loss": 5.2825,
      "mean_token_accuracy": 0.17497736364603042,
      "num_tokens": 4471987.0,
      "step": 2420
    },
    {
      "entropy": 5.497165203094482,
      "epoch": 2.0292887029288704,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048148197005603145,
      "loss": 5.2191,
      "mean_token_accuracy": 0.17726924568414687,
      "num_tokens": 4481645.0,
      "step": 2425
    },
    {
      "entropy": 5.368935203552246,
      "epoch": 2.0334728033472804,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048135352917224005,
      "loss": 5.1751,
      "mean_token_accuracy": 0.1868342339992523,
      "num_tokens": 4491166.0,
      "step": 2430
    },
    {
      "entropy": 5.578146934509277,
      "epoch": 2.0376569037656904,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048122466364606214,
      "loss": 5.1973,
      "mean_token_accuracy": 0.18251455277204515,
      "num_tokens": 4500159.0,
      "step": 2435
    },
    {
      "entropy": 5.389235353469848,
      "epoch": 2.0418410041841004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048109537374268216,
      "loss": 5.1464,
      "mean_token_accuracy": 0.18923695534467697,
      "num_tokens": 4509737.0,
      "step": 2440
    },
    {
      "entropy": 5.387840270996094,
      "epoch": 2.0460251046025104,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048096565972815793,
      "loss": 5.0715,
      "mean_token_accuracy": 0.18712350428104402,
      "num_tokens": 4518379.0,
      "step": 2445
    },
    {
      "entropy": 5.4111696720123295,
      "epoch": 2.0502092050209204,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004808355218694199,
      "loss": 5.2007,
      "mean_token_accuracy": 0.1853238508105278,
      "num_tokens": 4527885.0,
      "step": 2450
    },
    {
      "entropy": 5.4629090309143065,
      "epoch": 2.0543933054393304,
      "grad_norm": 1.125,
      "learning_rate": 0.00048070496043427116,
      "loss": 5.2467,
      "mean_token_accuracy": 0.17788249850273133,
      "num_tokens": 4537894.0,
      "step": 2455
    },
    {
      "entropy": 5.447832202911377,
      "epoch": 2.0585774058577404,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048057397569138596,
      "loss": 5.1357,
      "mean_token_accuracy": 0.19053970128297806,
      "num_tokens": 4546462.0,
      "step": 2460
    },
    {
      "entropy": 5.438874244689941,
      "epoch": 2.062761506276151,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004804425679103099,
      "loss": 5.1718,
      "mean_token_accuracy": 0.18075911104679107,
      "num_tokens": 4555970.0,
      "step": 2465
    },
    {
      "entropy": 5.429273414611816,
      "epoch": 2.066945606694561,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004803107373614589,
      "loss": 5.2205,
      "mean_token_accuracy": 0.1768643453717232,
      "num_tokens": 4565799.0,
      "step": 2470
    },
    {
      "entropy": 5.524178123474121,
      "epoch": 2.071129707112971,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004801784843161194,
      "loss": 5.251,
      "mean_token_accuracy": 0.17221496254205704,
      "num_tokens": 4575571.0,
      "step": 2475
    },
    {
      "entropy": 5.416622257232666,
      "epoch": 2.075313807531381,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004800458090464465,
      "loss": 5.1677,
      "mean_token_accuracy": 0.18350304514169694,
      "num_tokens": 4584263.0,
      "step": 2480
    },
    {
      "entropy": 5.466772651672363,
      "epoch": 2.079497907949791,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047991271182546476,
      "loss": 5.1469,
      "mean_token_accuracy": 0.18672786206007003,
      "num_tokens": 4592979.0,
      "step": 2485
    },
    {
      "entropy": 5.362415981292725,
      "epoch": 2.083682008368201,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047977919292706696,
      "loss": 5.1671,
      "mean_token_accuracy": 0.18652344197034837,
      "num_tokens": 4602042.0,
      "step": 2490
    },
    {
      "entropy": 5.459194040298462,
      "epoch": 2.087866108786611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004796452526260134,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18386663645505905,
      "num_tokens": 4611197.0,
      "step": 2495
    },
    {
      "entropy": 5.290938758850098,
      "epoch": 2.092050209205021,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047951089119793164,
      "loss": 5.1656,
      "mean_token_accuracy": 0.18329917788505554,
      "num_tokens": 4621612.0,
      "step": 2500
    },
    {
      "epoch": 2.092050209205021,
      "eval_entropy": 5.208794407561757,
      "eval_loss": 5.624749183654785,
      "eval_mean_token_accuracy": 0.17307274384854518,
      "eval_num_tokens": 4621612.0,
      "eval_runtime": 2.7319,
      "eval_samples_per_second": 1430.146,
      "eval_steps_per_second": 178.997,
      "step": 2500
    },
    {
      "entropy": 5.342732238769531,
      "epoch": 2.096234309623431,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000479376108919316,
      "loss": 5.0527,
      "mean_token_accuracy": 0.19201176315546037,
      "num_tokens": 4630198.0,
      "step": 2505
    },
    {
      "entropy": 5.436562967300415,
      "epoch": 2.100418410041841,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004792409060675267,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1784696400165558,
      "num_tokens": 4638729.0,
      "step": 2510
    },
    {
      "entropy": 5.353607606887818,
      "epoch": 2.104602510460251,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004791052829207895,
      "loss": 5.2745,
      "mean_token_accuracy": 0.18427398651838303,
      "num_tokens": 4648759.0,
      "step": 2515
    },
    {
      "entropy": 5.442644739151001,
      "epoch": 2.1087866108786613,
      "grad_norm": 1.140625,
      "learning_rate": 0.000478969239758195,
      "loss": 5.2189,
      "mean_token_accuracy": 0.18320283889770508,
      "num_tokens": 4657744.0,
      "step": 2520
    },
    {
      "entropy": 5.3526036739349365,
      "epoch": 2.1129707112970713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047883277685969815,
      "loss": 5.1865,
      "mean_token_accuracy": 0.18448097854852677,
      "num_tokens": 4666154.0,
      "step": 2525
    },
    {
      "entropy": 5.471697568893433,
      "epoch": 2.1171548117154813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047869589450611773,
      "loss": 5.2199,
      "mean_token_accuracy": 0.17987029403448104,
      "num_tokens": 4675677.0,
      "step": 2530
    },
    {
      "entropy": 5.375281524658203,
      "epoch": 2.1213389121338913,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004785585929791355,
      "loss": 5.2017,
      "mean_token_accuracy": 0.18070452511310578,
      "num_tokens": 4684733.0,
      "step": 2535
    },
    {
      "entropy": 5.405807685852051,
      "epoch": 2.1255230125523012,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000478420872561296,
      "loss": 5.1302,
      "mean_token_accuracy": 0.18370863795280457,
      "num_tokens": 4694155.0,
      "step": 2540
    },
    {
      "entropy": 5.34947624206543,
      "epoch": 2.1297071129707112,
      "grad_norm": 1.125,
      "learning_rate": 0.00047828273353600577,
      "loss": 5.1689,
      "mean_token_accuracy": 0.19383873641490937,
      "num_tokens": 4703286.0,
      "step": 2545
    },
    {
      "entropy": 5.456436491012573,
      "epoch": 2.1338912133891212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047814417618753256,
      "loss": 5.1856,
      "mean_token_accuracy": 0.17915833443403245,
      "num_tokens": 4711666.0,
      "step": 2550
    },
    {
      "entropy": 5.34681134223938,
      "epoch": 2.1380753138075312,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047800520080100526,
      "loss": 5.1984,
      "mean_token_accuracy": 0.17571722120046615,
      "num_tokens": 4721644.0,
      "step": 2555
    },
    {
      "entropy": 5.424099111557007,
      "epoch": 2.1422594142259412,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047786580766241265,
      "loss": 5.1626,
      "mean_token_accuracy": 0.19055590331554412,
      "num_tokens": 4730885.0,
      "step": 2560
    },
    {
      "entropy": 5.332646608352661,
      "epoch": 2.1464435146443517,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004777259970586036,
      "loss": 5.1425,
      "mean_token_accuracy": 0.18602774739265443,
      "num_tokens": 4740871.0,
      "step": 2565
    },
    {
      "entropy": 5.349590539932251,
      "epoch": 2.1506276150627617,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004775857692772859,
      "loss": 5.1157,
      "mean_token_accuracy": 0.18249449878931046,
      "num_tokens": 4750603.0,
      "step": 2570
    },
    {
      "entropy": 5.368098211288452,
      "epoch": 2.1548117154811717,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004774451246070257,
      "loss": 5.2329,
      "mean_token_accuracy": 0.18403066843748092,
      "num_tokens": 4760681.0,
      "step": 2575
    },
    {
      "entropy": 5.442989015579224,
      "epoch": 2.1589958158995817,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004773040633372471,
      "loss": 5.2284,
      "mean_token_accuracy": 0.18556591123342514,
      "num_tokens": 4770237.0,
      "step": 2580
    },
    {
      "entropy": 5.362786769866943,
      "epoch": 2.1631799163179917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047716258575823147,
      "loss": 5.2329,
      "mean_token_accuracy": 0.19020240604877472,
      "num_tokens": 4779022.0,
      "step": 2585
    },
    {
      "entropy": 5.4511971950531,
      "epoch": 2.1673640167364017,
      "grad_norm": 1.03125,
      "learning_rate": 0.000477020692161117,
      "loss": 5.2638,
      "mean_token_accuracy": 0.18160027861595154,
      "num_tokens": 4788644.0,
      "step": 2590
    },
    {
      "entropy": 5.375961828231811,
      "epoch": 2.1715481171548117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004768783828378978,
      "loss": 5.2574,
      "mean_token_accuracy": 0.17766260355710983,
      "num_tokens": 4798628.0,
      "step": 2595
    },
    {
      "entropy": 5.573528003692627,
      "epoch": 2.1757322175732217,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004767356580814237,
      "loss": 5.3659,
      "mean_token_accuracy": 0.17849716544151306,
      "num_tokens": 4807727.0,
      "step": 2600
    },
    {
      "entropy": 5.4480596542358395,
      "epoch": 2.1799163179916317,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004765925181853993,
      "loss": 5.2467,
      "mean_token_accuracy": 0.17864772230386733,
      "num_tokens": 4817896.0,
      "step": 2605
    },
    {
      "entropy": 5.428419685363769,
      "epoch": 2.1841004184100417,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004764489634443834,
      "loss": 5.2058,
      "mean_token_accuracy": 0.18008076697587966,
      "num_tokens": 4827135.0,
      "step": 2610
    },
    {
      "entropy": 5.491842937469483,
      "epoch": 2.1882845188284517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047630499415378863,
      "loss": 5.2869,
      "mean_token_accuracy": 0.17407437562942504,
      "num_tokens": 4837554.0,
      "step": 2615
    },
    {
      "entropy": 5.330071115493775,
      "epoch": 2.192468619246862,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004761606106098806,
      "loss": 5.1608,
      "mean_token_accuracy": 0.1822814255952835,
      "num_tokens": 4846739.0,
      "step": 2620
    },
    {
      "entropy": 5.404991960525512,
      "epoch": 2.196652719665272,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047601581310977747,
      "loss": 5.1916,
      "mean_token_accuracy": 0.1816651403903961,
      "num_tokens": 4855707.0,
      "step": 2625
    },
    {
      "entropy": 5.412823104858399,
      "epoch": 2.200836820083682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004758706019514493,
      "loss": 5.207,
      "mean_token_accuracy": 0.18123696148395538,
      "num_tokens": 4865098.0,
      "step": 2630
    },
    {
      "entropy": 5.448468255996704,
      "epoch": 2.205020920502092,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004757249774337172,
      "loss": 5.2198,
      "mean_token_accuracy": 0.1823917880654335,
      "num_tokens": 4874426.0,
      "step": 2635
    },
    {
      "entropy": 5.36552324295044,
      "epoch": 2.209205020920502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047557893985625293,
      "loss": 5.1819,
      "mean_token_accuracy": 0.18760903626680375,
      "num_tokens": 4883240.0,
      "step": 2640
    },
    {
      "entropy": 5.359978675842285,
      "epoch": 2.213389121338912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004754324895195786,
      "loss": 5.182,
      "mean_token_accuracy": 0.1838080033659935,
      "num_tokens": 4891989.0,
      "step": 2645
    },
    {
      "entropy": 5.422989845275879,
      "epoch": 2.217573221757322,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004752856267250652,
      "loss": 5.2809,
      "mean_token_accuracy": 0.1831986978650093,
      "num_tokens": 4901546.0,
      "step": 2650
    },
    {
      "entropy": 5.334002828598022,
      "epoch": 2.221757322175732,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047513835177493286,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19226252734661103,
      "num_tokens": 4910222.0,
      "step": 2655
    },
    {
      "entropy": 5.4006264209747314,
      "epoch": 2.225941422594142,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004749906649722498,
      "loss": 5.3129,
      "mean_token_accuracy": 0.17310767769813537,
      "num_tokens": 4920184.0,
      "step": 2660
    },
    {
      "entropy": 5.370958471298218,
      "epoch": 2.2301255230125525,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004748425666209316,
      "loss": 5.2172,
      "mean_token_accuracy": 0.18402200639247895,
      "num_tokens": 4930014.0,
      "step": 2665
    },
    {
      "entropy": 5.380784320831299,
      "epoch": 2.2343096234309625,
      "grad_norm": 1.0,
      "learning_rate": 0.0004746940570257409,
      "loss": 5.146,
      "mean_token_accuracy": 0.18692640513181685,
      "num_tokens": 4939556.0,
      "step": 2670
    },
    {
      "entropy": 5.322055530548096,
      "epoch": 2.2384937238493725,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004745451364922867,
      "loss": 5.2095,
      "mean_token_accuracy": 0.181628580391407,
      "num_tokens": 4949066.0,
      "step": 2675
    },
    {
      "entropy": 5.443222427368164,
      "epoch": 2.2426778242677825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004743958053270233,
      "loss": 5.2076,
      "mean_token_accuracy": 0.18355339765548706,
      "num_tokens": 4957881.0,
      "step": 2680
    },
    {
      "entropy": 5.419846630096435,
      "epoch": 2.2468619246861925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047424606383725037,
      "loss": 5.245,
      "mean_token_accuracy": 0.18528629839420319,
      "num_tokens": 4967105.0,
      "step": 2685
    },
    {
      "entropy": 5.280267333984375,
      "epoch": 2.2510460251046025,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004740959123311118,
      "loss": 5.1591,
      "mean_token_accuracy": 0.1876959204673767,
      "num_tokens": 4977676.0,
      "step": 2690
    },
    {
      "entropy": 5.382284307479859,
      "epoch": 2.2552301255230125,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047394535111759526,
      "loss": 5.173,
      "mean_token_accuracy": 0.1855098471045494,
      "num_tokens": 4986163.0,
      "step": 2695
    },
    {
      "entropy": 5.391265153884888,
      "epoch": 2.2594142259414225,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004737943805065314,
      "loss": 5.1826,
      "mean_token_accuracy": 0.17979367673397065,
      "num_tokens": 4995179.0,
      "step": 2700
    },
    {
      "entropy": 5.460883665084839,
      "epoch": 2.2635983263598325,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047364300080859363,
      "loss": 5.2715,
      "mean_token_accuracy": 0.17535117119550706,
      "num_tokens": 5003856.0,
      "step": 2705
    },
    {
      "entropy": 5.416038179397583,
      "epoch": 2.2677824267782425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047349121233529695,
      "loss": 5.2948,
      "mean_token_accuracy": 0.1777150511741638,
      "num_tokens": 5013506.0,
      "step": 2710
    },
    {
      "entropy": 5.383950185775757,
      "epoch": 2.2719665271966525,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004733390153989976,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1848796382546425,
      "num_tokens": 5022159.0,
      "step": 2715
    },
    {
      "entropy": 5.3696952819824215,
      "epoch": 2.276150627615063,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047318641031289245,
      "loss": 5.1792,
      "mean_token_accuracy": 0.18405805081129073,
      "num_tokens": 5031869.0,
      "step": 2720
    },
    {
      "entropy": 5.280503368377685,
      "epoch": 2.280334728033473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004730333973910183,
      "loss": 5.0345,
      "mean_token_accuracy": 0.19599688649177552,
      "num_tokens": 5039904.0,
      "step": 2725
    },
    {
      "entropy": 5.422358751296997,
      "epoch": 2.284518828451883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004728799769482511,
      "loss": 5.3412,
      "mean_token_accuracy": 0.1808583602309227,
      "num_tokens": 5049144.0,
      "step": 2730
    },
    {
      "entropy": 5.349178791046143,
      "epoch": 2.288702928870293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047272614930030533,
      "loss": 5.1023,
      "mean_token_accuracy": 0.18753841668367385,
      "num_tokens": 5058302.0,
      "step": 2735
    },
    {
      "entropy": 5.3154381275177,
      "epoch": 2.292887029288703,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047257191476373375,
      "loss": 5.1363,
      "mean_token_accuracy": 0.18648675978183746,
      "num_tokens": 5067443.0,
      "step": 2740
    },
    {
      "entropy": 5.345571088790893,
      "epoch": 2.297071129707113,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047241727365592603,
      "loss": 5.2723,
      "mean_token_accuracy": 0.17538231164216994,
      "num_tokens": 5076415.0,
      "step": 2745
    },
    {
      "entropy": 5.421261310577393,
      "epoch": 2.301255230125523,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004722622262951089,
      "loss": 5.2224,
      "mean_token_accuracy": 0.1793004974722862,
      "num_tokens": 5085882.0,
      "step": 2750
    },
    {
      "entropy": 5.438392972946167,
      "epoch": 2.305439330543933,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004721067730003448,
      "loss": 5.1986,
      "mean_token_accuracy": 0.1918323591351509,
      "num_tokens": 5094685.0,
      "step": 2755
    },
    {
      "entropy": 5.354588603973388,
      "epoch": 2.309623430962343,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004719509140915317,
      "loss": 5.2114,
      "mean_token_accuracy": 0.18272558599710464,
      "num_tokens": 5103885.0,
      "step": 2760
    },
    {
      "entropy": 5.385832977294922,
      "epoch": 2.3138075313807533,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047179464988940205,
      "loss": 5.1342,
      "mean_token_accuracy": 0.19514969885349273,
      "num_tokens": 5112164.0,
      "step": 2765
    },
    {
      "entropy": 5.255523633956909,
      "epoch": 2.3179916317991633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047163798071552253,
      "loss": 5.1038,
      "mean_token_accuracy": 0.19346790313720702,
      "num_tokens": 5121913.0,
      "step": 2770
    },
    {
      "entropy": 5.389960050582886,
      "epoch": 2.3221757322175733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004714809068922931,
      "loss": 5.1109,
      "mean_token_accuracy": 0.19163762032985687,
      "num_tokens": 5130447.0,
      "step": 2775
    },
    {
      "entropy": 5.290537691116333,
      "epoch": 2.3263598326359833,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004713234287429465,
      "loss": 5.237,
      "mean_token_accuracy": 0.18625620007514954,
      "num_tokens": 5139863.0,
      "step": 2780
    },
    {
      "entropy": 5.301087760925293,
      "epoch": 2.3305439330543933,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004711655465915473,
      "loss": 5.2041,
      "mean_token_accuracy": 0.1895967036485672,
      "num_tokens": 5149945.0,
      "step": 2785
    },
    {
      "entropy": 5.448004913330078,
      "epoch": 2.3347280334728033,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004710072607629917,
      "loss": 5.1469,
      "mean_token_accuracy": 0.18175724148750305,
      "num_tokens": 5160293.0,
      "step": 2790
    },
    {
      "entropy": 5.332290697097778,
      "epoch": 2.3389121338912133,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047084857158300646,
      "loss": 5.1618,
      "mean_token_accuracy": 0.18061374127864838,
      "num_tokens": 5169896.0,
      "step": 2795
    },
    {
      "entropy": 5.3752429485321045,
      "epoch": 2.3430962343096233,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047068947937814844,
      "loss": 5.2038,
      "mean_token_accuracy": 0.18880946189165115,
      "num_tokens": 5178898.0,
      "step": 2800
    },
    {
      "entropy": 5.346495723724365,
      "epoch": 2.3472803347280333,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047052998447580374,
      "loss": 5.278,
      "mean_token_accuracy": 0.1804632231593132,
      "num_tokens": 5188733.0,
      "step": 2805
    },
    {
      "entropy": 5.341215991973877,
      "epoch": 2.3514644351464433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047037008720418724,
      "loss": 5.1752,
      "mean_token_accuracy": 0.18327055722475052,
      "num_tokens": 5198117.0,
      "step": 2810
    },
    {
      "entropy": 5.441304492950439,
      "epoch": 2.3556485355648533,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004702097878923419,
      "loss": 5.2315,
      "mean_token_accuracy": 0.18148796558380126,
      "num_tokens": 5206791.0,
      "step": 2815
    },
    {
      "entropy": 5.28036470413208,
      "epoch": 2.3598326359832638,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004700490868701377,
      "loss": 5.1199,
      "mean_token_accuracy": 0.19287294000387192,
      "num_tokens": 5216427.0,
      "step": 2820
    },
    {
      "entropy": 5.296500301361084,
      "epoch": 2.3640167364016738,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004698879844682718,
      "loss": 5.1009,
      "mean_token_accuracy": 0.19283896833658218,
      "num_tokens": 5225983.0,
      "step": 2825
    },
    {
      "entropy": 5.286601161956787,
      "epoch": 2.3682008368200838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046972648101826695,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18582138568162918,
      "num_tokens": 5234547.0,
      "step": 2830
    },
    {
      "entropy": 5.49201192855835,
      "epoch": 2.3723849372384938,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004695645768524712,
      "loss": 5.2535,
      "mean_token_accuracy": 0.1892945185303688,
      "num_tokens": 5243281.0,
      "step": 2835
    },
    {
      "entropy": 5.31707010269165,
      "epoch": 2.3765690376569037,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004694022723040573,
      "loss": 5.1356,
      "mean_token_accuracy": 0.18640855848789215,
      "num_tokens": 5252318.0,
      "step": 2840
    },
    {
      "entropy": 5.3913795948028564,
      "epoch": 2.3807531380753137,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046923956770702206,
      "loss": 5.2385,
      "mean_token_accuracy": 0.19023721665143967,
      "num_tokens": 5261781.0,
      "step": 2845
    },
    {
      "entropy": 5.43264856338501,
      "epoch": 2.3849372384937237,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004690764633961852,
      "loss": 5.1422,
      "mean_token_accuracy": 0.1864036113023758,
      "num_tokens": 5270409.0,
      "step": 2850
    },
    {
      "entropy": 5.2199077129364015,
      "epoch": 2.3891213389121337,
      "grad_norm": 1.125,
      "learning_rate": 0.0004689129597071893,
      "loss": 5.1356,
      "mean_token_accuracy": 0.1981827899813652,
      "num_tokens": 5278766.0,
      "step": 2855
    },
    {
      "entropy": 5.398251724243164,
      "epoch": 2.393305439330544,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046874905697649857,
      "loss": 5.3132,
      "mean_token_accuracy": 0.175306336581707,
      "num_tokens": 5287572.0,
      "step": 2860
    },
    {
      "entropy": 5.394223594665528,
      "epoch": 2.397489539748954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046858475554139857,
      "loss": 5.2127,
      "mean_token_accuracy": 0.18782407343387603,
      "num_tokens": 5297348.0,
      "step": 2865
    },
    {
      "entropy": 5.372149515151977,
      "epoch": 2.401673640167364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004684200557399952,
      "loss": 5.2302,
      "mean_token_accuracy": 0.18365682363510133,
      "num_tokens": 5305748.0,
      "step": 2870
    },
    {
      "entropy": 5.339284038543701,
      "epoch": 2.405857740585774,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046825495791121414,
      "loss": 5.1709,
      "mean_token_accuracy": 0.19215793162584305,
      "num_tokens": 5314482.0,
      "step": 2875
    },
    {
      "entropy": 5.379046964645386,
      "epoch": 2.410041841004184,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046808946239480017,
      "loss": 5.2464,
      "mean_token_accuracy": 0.17876814305782318,
      "num_tokens": 5325161.0,
      "step": 2880
    },
    {
      "entropy": 5.321932125091553,
      "epoch": 2.414225941422594,
      "grad_norm": 1.125,
      "learning_rate": 0.0004679235695313165,
      "loss": 5.2372,
      "mean_token_accuracy": 0.17987630516290665,
      "num_tokens": 5335262.0,
      "step": 2885
    },
    {
      "entropy": 5.33101749420166,
      "epoch": 2.418410041841004,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046775727966214393,
      "loss": 5.1546,
      "mean_token_accuracy": 0.18333236128091812,
      "num_tokens": 5345145.0,
      "step": 2890
    },
    {
      "entropy": 5.274477243423462,
      "epoch": 2.422594142259414,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004675905931294803,
      "loss": 5.1414,
      "mean_token_accuracy": 0.18669739812612535,
      "num_tokens": 5354715.0,
      "step": 2895
    },
    {
      "entropy": 5.275536823272705,
      "epoch": 2.426778242677824,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004674235102763398,
      "loss": 5.2024,
      "mean_token_accuracy": 0.18598475754261018,
      "num_tokens": 5364180.0,
      "step": 2900
    },
    {
      "entropy": 5.340838670730591,
      "epoch": 2.430962343096234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046725603144655195,
      "loss": 5.1738,
      "mean_token_accuracy": 0.18643196821212768,
      "num_tokens": 5372919.0,
      "step": 2905
    },
    {
      "entropy": 5.340966749191284,
      "epoch": 2.435146443514644,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004670881569847612,
      "loss": 5.1475,
      "mean_token_accuracy": 0.19049450010061264,
      "num_tokens": 5383002.0,
      "step": 2910
    },
    {
      "entropy": 5.390149021148682,
      "epoch": 2.439330543933054,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046691988723642636,
      "loss": 5.3378,
      "mean_token_accuracy": 0.17228108942508696,
      "num_tokens": 5393622.0,
      "step": 2915
    },
    {
      "entropy": 5.419896459579467,
      "epoch": 2.4435146443514646,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004667512225478195,
      "loss": 5.2151,
      "mean_token_accuracy": 0.18529651761054994,
      "num_tokens": 5403106.0,
      "step": 2920
    },
    {
      "entropy": 5.36248083114624,
      "epoch": 2.4476987447698746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046658216326602526,
      "loss": 5.2015,
      "mean_token_accuracy": 0.18339800983667373,
      "num_tokens": 5412596.0,
      "step": 2925
    },
    {
      "entropy": 5.41774263381958,
      "epoch": 2.4518828451882846,
      "grad_norm": 1.125,
      "learning_rate": 0.0004664127097389406,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1855345368385315,
      "num_tokens": 5422301.0,
      "step": 2930
    },
    {
      "entropy": 5.411597537994385,
      "epoch": 2.4560669456066946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046624286231527367,
      "loss": 5.1668,
      "mean_token_accuracy": 0.18422241806983947,
      "num_tokens": 5432178.0,
      "step": 2935
    },
    {
      "entropy": 5.282891702651978,
      "epoch": 2.4602510460251046,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004660726213445431,
      "loss": 5.209,
      "mean_token_accuracy": 0.18362156450748443,
      "num_tokens": 5441162.0,
      "step": 2940
    },
    {
      "entropy": 5.330004358291626,
      "epoch": 2.4644351464435146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004659019871770774,
      "loss": 5.1474,
      "mean_token_accuracy": 0.19479316771030425,
      "num_tokens": 5450566.0,
      "step": 2945
    },
    {
      "entropy": 5.3660375595092775,
      "epoch": 2.4686192468619246,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046573096016401435,
      "loss": 5.1836,
      "mean_token_accuracy": 0.19586996287107467,
      "num_tokens": 5460289.0,
      "step": 2950
    },
    {
      "entropy": 5.243880128860473,
      "epoch": 2.4728033472803346,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004655595406572999,
      "loss": 5.1583,
      "mean_token_accuracy": 0.19155662804841994,
      "num_tokens": 5469581.0,
      "step": 2955
    },
    {
      "entropy": 5.40550274848938,
      "epoch": 2.476987447698745,
      "grad_norm": 1.0625,
      "learning_rate": 0.000465387729009688,
      "loss": 5.2454,
      "mean_token_accuracy": 0.189969938993454,
      "num_tokens": 5478005.0,
      "step": 2960
    },
    {
      "entropy": 5.331111335754395,
      "epoch": 2.481171548117155,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046521552557473927,
      "loss": 5.2461,
      "mean_token_accuracy": 0.180669966340065,
      "num_tokens": 5486963.0,
      "step": 2965
    },
    {
      "entropy": 5.326622200012207,
      "epoch": 2.485355648535565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046504293070682073,
      "loss": 5.1202,
      "mean_token_accuracy": 0.18449053466320037,
      "num_tokens": 5496368.0,
      "step": 2970
    },
    {
      "entropy": 5.282700920104981,
      "epoch": 2.489539748953975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046486994476110493,
      "loss": 5.1265,
      "mean_token_accuracy": 0.19323625713586806,
      "num_tokens": 5505706.0,
      "step": 2975
    },
    {
      "entropy": 5.259168577194214,
      "epoch": 2.493723849372385,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046469656809356916,
      "loss": 5.1096,
      "mean_token_accuracy": 0.19017785638570786,
      "num_tokens": 5513945.0,
      "step": 2980
    },
    {
      "entropy": 5.2196814060211185,
      "epoch": 2.497907949790795,
      "grad_norm": 1.125,
      "learning_rate": 0.00046452280106099454,
      "loss": 5.1553,
      "mean_token_accuracy": 0.18404532372951507,
      "num_tokens": 5523831.0,
      "step": 2985
    },
    {
      "entropy": 5.420633029937744,
      "epoch": 2.502092050209205,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004643486440209659,
      "loss": 5.2771,
      "mean_token_accuracy": 0.1873683363199234,
      "num_tokens": 5532705.0,
      "step": 2990
    },
    {
      "entropy": 5.390104961395264,
      "epoch": 2.506276150627615,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004641740973318702,
      "loss": 5.1953,
      "mean_token_accuracy": 0.19106512367725373,
      "num_tokens": 5542793.0,
      "step": 2995
    },
    {
      "entropy": 5.359776258468628,
      "epoch": 2.510460251046025,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004639991613528967,
      "loss": 5.1935,
      "mean_token_accuracy": 0.1863290324807167,
      "num_tokens": 5551938.0,
      "step": 3000
    },
    {
      "epoch": 2.510460251046025,
      "eval_entropy": 5.169321855884388,
      "eval_loss": 5.5464887619018555,
      "eval_mean_token_accuracy": 0.1789830338918358,
      "eval_num_tokens": 5551938.0,
      "eval_runtime": 3.0168,
      "eval_samples_per_second": 1295.075,
      "eval_steps_per_second": 162.091,
      "step": 3000
    },
    {
      "entropy": 5.357166910171509,
      "epoch": 2.514644351464435,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046382383644403543,
      "loss": 5.2321,
      "mean_token_accuracy": 0.18058869987726212,
      "num_tokens": 5561374.0,
      "step": 3005
    },
    {
      "entropy": 5.256979990005493,
      "epoch": 2.518828451882845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046364812296607676,
      "loss": 5.0951,
      "mean_token_accuracy": 0.1917743131518364,
      "num_tokens": 5570459.0,
      "step": 3010
    },
    {
      "entropy": 5.255487632751465,
      "epoch": 2.523012552301255,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046347202128061094,
      "loss": 5.2105,
      "mean_token_accuracy": 0.17877186685800553,
      "num_tokens": 5579692.0,
      "step": 3015
    },
    {
      "entropy": 5.4036478996276855,
      "epoch": 2.5271966527196654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004632955317500268,
      "loss": 5.1706,
      "mean_token_accuracy": 0.1878630980849266,
      "num_tokens": 5588338.0,
      "step": 3020
    },
    {
      "entropy": 5.335281419754028,
      "epoch": 2.5313807531380754,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046311865473751154,
      "loss": 5.2403,
      "mean_token_accuracy": 0.1856330528855324,
      "num_tokens": 5598589.0,
      "step": 3025
    },
    {
      "entropy": 5.380405378341675,
      "epoch": 2.5355648535564854,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004629413906070495,
      "loss": 5.2293,
      "mean_token_accuracy": 0.18747957050800323,
      "num_tokens": 5608127.0,
      "step": 3030
    },
    {
      "entropy": 5.384419059753418,
      "epoch": 2.5397489539748954,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046276373972342177,
      "loss": 5.2687,
      "mean_token_accuracy": 0.1905320852994919,
      "num_tokens": 5617728.0,
      "step": 3035
    },
    {
      "entropy": 5.359371852874756,
      "epoch": 2.5439330543933054,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004625857024522054,
      "loss": 5.217,
      "mean_token_accuracy": 0.18671149164438247,
      "num_tokens": 5626823.0,
      "step": 3040
    },
    {
      "entropy": 5.266793966293335,
      "epoch": 2.5481171548117154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046240727915977236,
      "loss": 5.1969,
      "mean_token_accuracy": 0.19211239665746688,
      "num_tokens": 5636284.0,
      "step": 3045
    },
    {
      "entropy": 5.451792049407959,
      "epoch": 2.5523012552301254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004622284702132893,
      "loss": 5.3029,
      "mean_token_accuracy": 0.18001856356859208,
      "num_tokens": 5645721.0,
      "step": 3050
    },
    {
      "entropy": 5.336426019668579,
      "epoch": 2.5564853556485354,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046204927598071595,
      "loss": 5.2377,
      "mean_token_accuracy": 0.182869254052639,
      "num_tokens": 5654734.0,
      "step": 3055
    },
    {
      "entropy": 5.366493511199951,
      "epoch": 2.560669456066946,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046186969683080547,
      "loss": 5.2121,
      "mean_token_accuracy": 0.19295602440834045,
      "num_tokens": 5664010.0,
      "step": 3060
    },
    {
      "entropy": 5.277511835098267,
      "epoch": 2.564853556485356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046168973313310294,
      "loss": 5.2143,
      "mean_token_accuracy": 0.18723931014537812,
      "num_tokens": 5673028.0,
      "step": 3065
    },
    {
      "entropy": 5.325067138671875,
      "epoch": 2.569037656903766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004615093852579445,
      "loss": 5.1813,
      "mean_token_accuracy": 0.19013588428497313,
      "num_tokens": 5682517.0,
      "step": 3070
    },
    {
      "entropy": 5.233159875869751,
      "epoch": 2.573221757322176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046132865357645717,
      "loss": 5.1082,
      "mean_token_accuracy": 0.19707662165164946,
      "num_tokens": 5690920.0,
      "step": 3075
    },
    {
      "entropy": 5.275312185287476,
      "epoch": 2.577405857740586,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046114753846055774,
      "loss": 5.1656,
      "mean_token_accuracy": 0.18295206874608994,
      "num_tokens": 5700336.0,
      "step": 3080
    },
    {
      "entropy": 5.367909049987793,
      "epoch": 2.581589958158996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000460966040282952,
      "loss": 5.3017,
      "mean_token_accuracy": 0.18178110867738723,
      "num_tokens": 5711085.0,
      "step": 3085
    },
    {
      "entropy": 5.30069785118103,
      "epoch": 2.585774058577406,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046078415941713384,
      "loss": 5.1157,
      "mean_token_accuracy": 0.19286441653966904,
      "num_tokens": 5719560.0,
      "step": 3090
    },
    {
      "entropy": 5.313541460037231,
      "epoch": 2.589958158995816,
      "grad_norm": 1.125,
      "learning_rate": 0.0004606018962373852,
      "loss": 5.0783,
      "mean_token_accuracy": 0.1986253798007965,
      "num_tokens": 5727865.0,
      "step": 3095
    },
    {
      "entropy": 5.304821634292603,
      "epoch": 2.594142259414226,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004604192511187741,
      "loss": 5.2132,
      "mean_token_accuracy": 0.18388793617486954,
      "num_tokens": 5737705.0,
      "step": 3100
    },
    {
      "entropy": 5.3230067729949955,
      "epoch": 2.598326359832636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046023622443715483,
      "loss": 5.2124,
      "mean_token_accuracy": 0.1778946965932846,
      "num_tokens": 5747057.0,
      "step": 3105
    },
    {
      "entropy": 5.3239351272583,
      "epoch": 2.602510460251046,
      "grad_norm": 1.125,
      "learning_rate": 0.00046005281656916707,
      "loss": 5.1831,
      "mean_token_accuracy": 0.18459225594997405,
      "num_tokens": 5755793.0,
      "step": 3110
    },
    {
      "entropy": 5.348266983032227,
      "epoch": 2.606694560669456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004598690278922346,
      "loss": 5.204,
      "mean_token_accuracy": 0.18266982585191727,
      "num_tokens": 5765385.0,
      "step": 3115
    },
    {
      "entropy": 5.192241477966308,
      "epoch": 2.6108786610878663,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045968485878456497,
      "loss": 5.1055,
      "mean_token_accuracy": 0.19464822113513947,
      "num_tokens": 5773947.0,
      "step": 3120
    },
    {
      "entropy": 5.307315492630005,
      "epoch": 2.6150627615062763,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004595003096251486,
      "loss": 5.1045,
      "mean_token_accuracy": 0.1931895688176155,
      "num_tokens": 5782700.0,
      "step": 3125
    },
    {
      "entropy": 5.348392486572266,
      "epoch": 2.6192468619246863,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045931538079375803,
      "loss": 5.2651,
      "mean_token_accuracy": 0.1823725715279579,
      "num_tokens": 5792452.0,
      "step": 3130
    },
    {
      "entropy": 5.342971086502075,
      "epoch": 2.6234309623430963,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045913007267094704,
      "loss": 5.1905,
      "mean_token_accuracy": 0.18951852321624757,
      "num_tokens": 5801270.0,
      "step": 3135
    },
    {
      "entropy": 5.358579015731811,
      "epoch": 2.6276150627615062,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004589443856380499,
      "loss": 5.237,
      "mean_token_accuracy": 0.18819057792425156,
      "num_tokens": 5810704.0,
      "step": 3140
    },
    {
      "entropy": 5.340524053573608,
      "epoch": 2.6317991631799162,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004587583200771807,
      "loss": 5.1769,
      "mean_token_accuracy": 0.18378162533044815,
      "num_tokens": 5819690.0,
      "step": 3145
    },
    {
      "entropy": 5.2421385765075685,
      "epoch": 2.6359832635983262,
      "grad_norm": 1.125,
      "learning_rate": 0.0004585718763712325,
      "loss": 5.1039,
      "mean_token_accuracy": 0.18966793715953828,
      "num_tokens": 5828912.0,
      "step": 3150
    },
    {
      "entropy": 5.150143051147461,
      "epoch": 2.6401673640167362,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004583850549038764,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19083824753761292,
      "num_tokens": 5836914.0,
      "step": 3155
    },
    {
      "entropy": 5.313408517837525,
      "epoch": 2.6443514644351467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045819785605956105,
      "loss": 5.2069,
      "mean_token_accuracy": 0.1899588793516159,
      "num_tokens": 5846268.0,
      "step": 3160
    },
    {
      "entropy": 5.314242172241211,
      "epoch": 2.6485355648535567,
      "grad_norm": 1.125,
      "learning_rate": 0.0004580102802235115,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18737068474292756,
      "num_tokens": 5855029.0,
      "step": 3165
    },
    {
      "entropy": 5.374032211303711,
      "epoch": 2.6527196652719667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004578223277817286,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1793490469455719,
      "num_tokens": 5865049.0,
      "step": 3170
    },
    {
      "entropy": 5.3609302043914795,
      "epoch": 2.6569037656903767,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045763399912098835,
      "loss": 5.1902,
      "mean_token_accuracy": 0.1795242965221405,
      "num_tokens": 5873593.0,
      "step": 3175
    },
    {
      "entropy": 5.216292810440064,
      "epoch": 2.6610878661087867,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045744529462884086,
      "loss": 5.1208,
      "mean_token_accuracy": 0.18870629817247392,
      "num_tokens": 5882562.0,
      "step": 3180
    },
    {
      "entropy": 5.355183887481689,
      "epoch": 2.6652719665271967,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045725621469360964,
      "loss": 5.2887,
      "mean_token_accuracy": 0.17769926488399507,
      "num_tokens": 5892516.0,
      "step": 3185
    },
    {
      "entropy": 5.322477912902832,
      "epoch": 2.6694560669456067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004570667597043907,
      "loss": 5.1852,
      "mean_token_accuracy": 0.18842493742704391,
      "num_tokens": 5901494.0,
      "step": 3190
    },
    {
      "entropy": 5.311296653747559,
      "epoch": 2.6736401673640167,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000456876930051052,
      "loss": 5.176,
      "mean_token_accuracy": 0.18503561168909072,
      "num_tokens": 5910750.0,
      "step": 3195
    },
    {
      "entropy": 5.383889198303223,
      "epoch": 2.6778242677824267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004566867261242325,
      "loss": 5.2186,
      "mean_token_accuracy": 0.188629674911499,
      "num_tokens": 5920277.0,
      "step": 3200
    },
    {
      "entropy": 5.310765027999878,
      "epoch": 2.6820083682008367,
      "grad_norm": 1.0,
      "learning_rate": 0.00045649614831534127,
      "loss": 5.1952,
      "mean_token_accuracy": 0.18537470698356628,
      "num_tokens": 5929880.0,
      "step": 3205
    },
    {
      "entropy": 5.428402519226074,
      "epoch": 2.6861924686192467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045630519701655677,
      "loss": 5.3412,
      "mean_token_accuracy": 0.1875407412648201,
      "num_tokens": 5939538.0,
      "step": 3210
    },
    {
      "entropy": 5.308660268783569,
      "epoch": 2.6903765690376567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045611387262082604,
      "loss": 5.1758,
      "mean_token_accuracy": 0.18673402816057205,
      "num_tokens": 5948747.0,
      "step": 3215
    },
    {
      "entropy": 5.280479192733765,
      "epoch": 2.694560669456067,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045592217552186407,
      "loss": 5.1683,
      "mean_token_accuracy": 0.18951367437839509,
      "num_tokens": 5958209.0,
      "step": 3220
    },
    {
      "entropy": 5.37222466468811,
      "epoch": 2.698744769874477,
      "grad_norm": 1.125,
      "learning_rate": 0.0004557301061141525,
      "loss": 5.219,
      "mean_token_accuracy": 0.18609079718589783,
      "num_tokens": 5967184.0,
      "step": 3225
    },
    {
      "entropy": 5.268049621582032,
      "epoch": 2.702928870292887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045553766479293945,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18060262799263,
      "num_tokens": 5976006.0,
      "step": 3230
    },
    {
      "entropy": 5.2162786483764645,
      "epoch": 2.707112970711297,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045534485195423806,
      "loss": 5.0743,
      "mean_token_accuracy": 0.19651037454605103,
      "num_tokens": 5984671.0,
      "step": 3235
    },
    {
      "entropy": 5.418746852874756,
      "epoch": 2.711297071129707,
      "grad_norm": 1.125,
      "learning_rate": 0.00045515166799482633,
      "loss": 5.2046,
      "mean_token_accuracy": 0.18489584624767302,
      "num_tokens": 5993190.0,
      "step": 3240
    },
    {
      "entropy": 5.286226463317871,
      "epoch": 2.715481171548117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045495811331224564,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18630554378032685,
      "num_tokens": 6004080.0,
      "step": 3245
    },
    {
      "entropy": 5.320625925064087,
      "epoch": 2.719665271966527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004547641883048005,
      "loss": 5.2035,
      "mean_token_accuracy": 0.18668723702430726,
      "num_tokens": 6013212.0,
      "step": 3250
    },
    {
      "entropy": 5.352359437942505,
      "epoch": 2.723849372384937,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004545698933715573,
      "loss": 5.1355,
      "mean_token_accuracy": 0.19180063307285308,
      "num_tokens": 6023444.0,
      "step": 3255
    },
    {
      "entropy": 5.286853885650634,
      "epoch": 2.7280334728033475,
      "grad_norm": 1.125,
      "learning_rate": 0.00045437522891234386,
      "loss": 5.1439,
      "mean_token_accuracy": 0.18751030117273332,
      "num_tokens": 6032217.0,
      "step": 3260
    },
    {
      "entropy": 5.258432769775391,
      "epoch": 2.7322175732217575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045418019532774827,
      "loss": 5.1081,
      "mean_token_accuracy": 0.1933275803923607,
      "num_tokens": 6040246.0,
      "step": 3265
    },
    {
      "entropy": 5.236212825775146,
      "epoch": 2.7364016736401675,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045398479301911836,
      "loss": 5.1991,
      "mean_token_accuracy": 0.18418097048997878,
      "num_tokens": 6049447.0,
      "step": 3270
    },
    {
      "entropy": 5.191279935836792,
      "epoch": 2.7405857740585775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045378902238856064,
      "loss": 5.0785,
      "mean_token_accuracy": 0.1938243106007576,
      "num_tokens": 6058294.0,
      "step": 3275
    },
    {
      "entropy": 5.364496278762817,
      "epoch": 2.7447698744769875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004535928838389396,
      "loss": 5.2146,
      "mean_token_accuracy": 0.1871143564581871,
      "num_tokens": 6067237.0,
      "step": 3280
    },
    {
      "entropy": 5.389237117767334,
      "epoch": 2.7489539748953975,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004533963777738768,
      "loss": 5.2133,
      "mean_token_accuracy": 0.18586805909872056,
      "num_tokens": 6075599.0,
      "step": 3285
    },
    {
      "entropy": 5.320382022857666,
      "epoch": 2.7531380753138075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045319950459775024,
      "loss": 5.1923,
      "mean_token_accuracy": 0.18377586752176284,
      "num_tokens": 6085131.0,
      "step": 3290
    },
    {
      "entropy": 5.351099586486816,
      "epoch": 2.7573221757322175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004530022647156931,
      "loss": 5.163,
      "mean_token_accuracy": 0.18796681463718415,
      "num_tokens": 6094745.0,
      "step": 3295
    },
    {
      "entropy": 5.258651208877564,
      "epoch": 2.7615062761506275,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004528046585335935,
      "loss": 5.2053,
      "mean_token_accuracy": 0.18384901583194732,
      "num_tokens": 6104182.0,
      "step": 3300
    },
    {
      "entropy": 5.321671009063721,
      "epoch": 2.7656903765690375,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045260668645809304,
      "loss": 5.0765,
      "mean_token_accuracy": 0.1958131730556488,
      "num_tokens": 6113370.0,
      "step": 3305
    },
    {
      "entropy": 5.219642162322998,
      "epoch": 2.7698744769874475,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004524083488965866,
      "loss": 5.1501,
      "mean_token_accuracy": 0.19209472686052323,
      "num_tokens": 6122202.0,
      "step": 3310
    },
    {
      "entropy": 5.329074144363403,
      "epoch": 2.7740585774058575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004522096462572208,
      "loss": 5.2094,
      "mean_token_accuracy": 0.17987604737281798,
      "num_tokens": 6131081.0,
      "step": 3315
    },
    {
      "entropy": 5.447030544281006,
      "epoch": 2.778242677824268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004520105789488939,
      "loss": 5.2811,
      "mean_token_accuracy": 0.1828023836016655,
      "num_tokens": 6141707.0,
      "step": 3320
    },
    {
      "entropy": 5.246972894668579,
      "epoch": 2.782426778242678,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045181114738125427,
      "loss": 5.1455,
      "mean_token_accuracy": 0.18587521314620972,
      "num_tokens": 6151651.0,
      "step": 3325
    },
    {
      "entropy": 5.329950952529908,
      "epoch": 2.786610878661088,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045161135196470015,
      "loss": 5.0965,
      "mean_token_accuracy": 0.196394182741642,
      "num_tokens": 6159928.0,
      "step": 3330
    },
    {
      "entropy": 5.303078079223633,
      "epoch": 2.790794979079498,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004514111931103783,
      "loss": 5.1992,
      "mean_token_accuracy": 0.18262383192777634,
      "num_tokens": 6169115.0,
      "step": 3335
    },
    {
      "entropy": 5.3082962989807125,
      "epoch": 2.794979079497908,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045121067123018353,
      "loss": 5.2198,
      "mean_token_accuracy": 0.18815610855817794,
      "num_tokens": 6177995.0,
      "step": 3340
    },
    {
      "entropy": 5.365166044235229,
      "epoch": 2.799163179916318,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045100978673675755,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18383896201848984,
      "num_tokens": 6188081.0,
      "step": 3345
    },
    {
      "entropy": 5.383021354675293,
      "epoch": 2.803347280334728,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045080854004348836,
      "loss": 5.161,
      "mean_token_accuracy": 0.18370933681726456,
      "num_tokens": 6198010.0,
      "step": 3350
    },
    {
      "entropy": 5.1978137493133545,
      "epoch": 2.8075313807531384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045060693156450933,
      "loss": 5.1343,
      "mean_token_accuracy": 0.19338563531637193,
      "num_tokens": 6207164.0,
      "step": 3355
    },
    {
      "entropy": 5.36414155960083,
      "epoch": 2.8117154811715483,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004504049617146983,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1865020900964737,
      "num_tokens": 6216512.0,
      "step": 3360
    },
    {
      "entropy": 5.351248121261596,
      "epoch": 2.8158995815899583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045020263090967674,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18984930515289306,
      "num_tokens": 6226196.0,
      "step": 3365
    },
    {
      "entropy": 5.2253562927246096,
      "epoch": 2.8200836820083683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004499999395658088,
      "loss": 5.0961,
      "mean_token_accuracy": 0.18750745207071304,
      "num_tokens": 6234820.0,
      "step": 3370
    },
    {
      "entropy": 5.317824411392212,
      "epoch": 2.8242677824267783,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000449796888100201,
      "loss": 5.155,
      "mean_token_accuracy": 0.1907557189464569,
      "num_tokens": 6243813.0,
      "step": 3375
    },
    {
      "entropy": 5.387785243988037,
      "epoch": 2.8284518828451883,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044959347693070026,
      "loss": 5.2613,
      "mean_token_accuracy": 0.1873804971575737,
      "num_tokens": 6253376.0,
      "step": 3380
    },
    {
      "entropy": 5.392985439300537,
      "epoch": 2.8326359832635983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004493897064758943,
      "loss": 5.2791,
      "mean_token_accuracy": 0.18504671454429628,
      "num_tokens": 6262640.0,
      "step": 3385
    },
    {
      "entropy": 5.400721645355224,
      "epoch": 2.8368200836820083,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004491855771551098,
      "loss": 5.233,
      "mean_token_accuracy": 0.19088169932365417,
      "num_tokens": 6270908.0,
      "step": 3390
    },
    {
      "entropy": 5.2806014060974125,
      "epoch": 2.8410041841004183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004489810893884121,
      "loss": 5.1434,
      "mean_token_accuracy": 0.1863793686032295,
      "num_tokens": 6280361.0,
      "step": 3395
    },
    {
      "entropy": 5.345275402069092,
      "epoch": 2.8451882845188283,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004487762435966042,
      "loss": 5.155,
      "mean_token_accuracy": 0.18769960552453996,
      "num_tokens": 6288832.0,
      "step": 3400
    },
    {
      "entropy": 5.248450183868409,
      "epoch": 2.8493723849372383,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044857104020122585,
      "loss": 5.1292,
      "mean_token_accuracy": 0.19043443650007247,
      "num_tokens": 6298553.0,
      "step": 3405
    },
    {
      "entropy": 5.235322427749634,
      "epoch": 2.8535564853556483,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004483654796245526,
      "loss": 5.096,
      "mean_token_accuracy": 0.1903188243508339,
      "num_tokens": 6307692.0,
      "step": 3410
    },
    {
      "entropy": 5.379913330078125,
      "epoch": 2.8577405857740583,
      "grad_norm": 1.09375,
      "learning_rate": 0.000448159562289595,
      "loss": 5.1818,
      "mean_token_accuracy": 0.18876910507678984,
      "num_tokens": 6316729.0,
      "step": 3415
    },
    {
      "entropy": 5.377156972885132,
      "epoch": 2.8619246861924688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044795328862009794,
      "loss": 5.2586,
      "mean_token_accuracy": 0.17772623151540756,
      "num_tokens": 6326210.0,
      "step": 3420
    },
    {
      "entropy": 5.245429372787475,
      "epoch": 2.8661087866108788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044774665904053943,
      "loss": 5.1695,
      "mean_token_accuracy": 0.18792026191949845,
      "num_tokens": 6335156.0,
      "step": 3425
    },
    {
      "entropy": 5.2129504680633545,
      "epoch": 2.8702928870292888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004475396739761299,
      "loss": 5.0415,
      "mean_token_accuracy": 0.1941840723156929,
      "num_tokens": 6344264.0,
      "step": 3430
    },
    {
      "entropy": 5.278776836395264,
      "epoch": 2.8744769874476988,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044733233385281117,
      "loss": 5.1183,
      "mean_token_accuracy": 0.1926687702536583,
      "num_tokens": 6352997.0,
      "step": 3435
    },
    {
      "entropy": 5.243551445007324,
      "epoch": 2.8786610878661087,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044712463909725626,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19510126113891602,
      "num_tokens": 6362312.0,
      "step": 3440
    },
    {
      "entropy": 5.373373699188233,
      "epoch": 2.8828451882845187,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044691659013686727,
      "loss": 5.2584,
      "mean_token_accuracy": 0.18169642835855485,
      "num_tokens": 6372107.0,
      "step": 3445
    },
    {
      "entropy": 5.264970064163208,
      "epoch": 2.8870292887029287,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004467081873997756,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19638821929693223,
      "num_tokens": 6381306.0,
      "step": 3450
    },
    {
      "entropy": 5.207119464874268,
      "epoch": 2.891213389121339,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044649943131484056,
      "loss": 5.029,
      "mean_token_accuracy": 0.19724439084529877,
      "num_tokens": 6390404.0,
      "step": 3455
    },
    {
      "entropy": 5.392963218688965,
      "epoch": 2.895397489539749,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044629032231164874,
      "loss": 5.2513,
      "mean_token_accuracy": 0.1828698620200157,
      "num_tokens": 6400366.0,
      "step": 3460
    },
    {
      "entropy": 5.281077146530151,
      "epoch": 2.899581589958159,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004460808608205127,
      "loss": 5.0682,
      "mean_token_accuracy": 0.19214740991592408,
      "num_tokens": 6408955.0,
      "step": 3465
    },
    {
      "entropy": 5.19082498550415,
      "epoch": 2.903765690376569,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004458710472724706,
      "loss": 5.101,
      "mean_token_accuracy": 0.1934350922703743,
      "num_tokens": 6417172.0,
      "step": 3470
    },
    {
      "entropy": 5.304494142532349,
      "epoch": 2.907949790794979,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044566088209928485,
      "loss": 5.1402,
      "mean_token_accuracy": 0.19179511964321136,
      "num_tokens": 6425898.0,
      "step": 3475
    },
    {
      "entropy": 5.245939922332764,
      "epoch": 2.912133891213389,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044545036573344175,
      "loss": 5.1741,
      "mean_token_accuracy": 0.18676467686891557,
      "num_tokens": 6435940.0,
      "step": 3480
    },
    {
      "entropy": 5.254410743713379,
      "epoch": 2.916317991631799,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044523949860815015,
      "loss": 5.1028,
      "mean_token_accuracy": 0.1937511071562767,
      "num_tokens": 6445119.0,
      "step": 3485
    },
    {
      "entropy": 5.240779972076416,
      "epoch": 2.920502092050209,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004450282811573405,
      "loss": 5.1481,
      "mean_token_accuracy": 0.19389404505491256,
      "num_tokens": 6453986.0,
      "step": 3490
    },
    {
      "entropy": 5.293921089172363,
      "epoch": 2.924686192468619,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044481671381566446,
      "loss": 5.1189,
      "mean_token_accuracy": 0.18862144500017167,
      "num_tokens": 6463658.0,
      "step": 3495
    },
    {
      "entropy": 5.253119277954101,
      "epoch": 2.928870292887029,
      "grad_norm": 1.125,
      "learning_rate": 0.0004446047970184937,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1902005970478058,
      "num_tokens": 6472370.0,
      "step": 3500
    },
    {
      "epoch": 2.928870292887029,
      "eval_entropy": 5.112072067026712,
      "eval_loss": 5.469865798950195,
      "eval_mean_token_accuracy": 0.18382642931611504,
      "eval_num_tokens": 6472370.0,
      "eval_runtime": 2.7402,
      "eval_samples_per_second": 1425.804,
      "eval_steps_per_second": 178.454,
      "step": 3500
    },
    {
      "entropy": 5.3380297183990475,
      "epoch": 2.933054393305439,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044439253120191877,
      "loss": 5.286,
      "mean_token_accuracy": 0.1812704920768738,
      "num_tokens": 6482642.0,
      "step": 3505
    },
    {
      "entropy": 5.325414514541626,
      "epoch": 2.937238493723849,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044417991680274874,
      "loss": 5.1941,
      "mean_token_accuracy": 0.19350529164075853,
      "num_tokens": 6491818.0,
      "step": 3510
    },
    {
      "entropy": 5.322343444824218,
      "epoch": 2.941422594142259,
      "grad_norm": 1.125,
      "learning_rate": 0.0004439669542585097,
      "loss": 5.1629,
      "mean_token_accuracy": 0.18652729839086532,
      "num_tokens": 6500908.0,
      "step": 3515
    },
    {
      "entropy": 5.245506954193115,
      "epoch": 2.9456066945606696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004437536440074444,
      "loss": 5.1442,
      "mean_token_accuracy": 0.1886870101094246,
      "num_tokens": 6510182.0,
      "step": 3520
    },
    {
      "entropy": 5.287081003189087,
      "epoch": 2.9497907949790796,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004435399864885111,
      "loss": 5.1246,
      "mean_token_accuracy": 0.1900799185037613,
      "num_tokens": 6519051.0,
      "step": 3525
    },
    {
      "entropy": 5.249131345748902,
      "epoch": 2.9539748953974896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004433259821413825,
      "loss": 5.063,
      "mean_token_accuracy": 0.20106301158666612,
      "num_tokens": 6527949.0,
      "step": 3530
    },
    {
      "entropy": 5.367303943634033,
      "epoch": 2.9581589958158996,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044311163140644524,
      "loss": 5.3363,
      "mean_token_accuracy": 0.1785028725862503,
      "num_tokens": 6537227.0,
      "step": 3535
    },
    {
      "entropy": 5.265641212463379,
      "epoch": 2.9623430962343096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044289693472479867,
      "loss": 5.1364,
      "mean_token_accuracy": 0.1846728578209877,
      "num_tokens": 6546938.0,
      "step": 3540
    },
    {
      "entropy": 5.255853223800659,
      "epoch": 2.9665271966527196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044268189253825397,
      "loss": 5.1348,
      "mean_token_accuracy": 0.19034533649683,
      "num_tokens": 6555750.0,
      "step": 3545
    },
    {
      "entropy": 5.248004579544068,
      "epoch": 2.9707112970711296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004424665052893334,
      "loss": 5.0754,
      "mean_token_accuracy": 0.19686341136693955,
      "num_tokens": 6565680.0,
      "step": 3550
    },
    {
      "entropy": 5.215185642242432,
      "epoch": 2.97489539748954,
      "grad_norm": 1.125,
      "learning_rate": 0.00044225077342126935,
      "loss": 5.0488,
      "mean_token_accuracy": 0.19836441725492476,
      "num_tokens": 6574595.0,
      "step": 3555
    },
    {
      "entropy": 5.270293712615967,
      "epoch": 2.97907949790795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004420346973780033,
      "loss": 5.1859,
      "mean_token_accuracy": 0.18911138027906418,
      "num_tokens": 6583333.0,
      "step": 3560
    },
    {
      "entropy": 5.324273633956909,
      "epoch": 2.98326359832636,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000441818277604185,
      "loss": 5.1999,
      "mean_token_accuracy": 0.19488253444433212,
      "num_tokens": 6592252.0,
      "step": 3565
    },
    {
      "entropy": 5.242170810699463,
      "epoch": 2.98744769874477,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004416015145451715,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19601638615131378,
      "num_tokens": 6600675.0,
      "step": 3570
    },
    {
      "entropy": 5.246265029907226,
      "epoch": 2.99163179916318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004413844086470264,
      "loss": 5.1178,
      "mean_token_accuracy": 0.19394250065088273,
      "num_tokens": 6609722.0,
      "step": 3575
    },
    {
      "entropy": 5.239301013946533,
      "epoch": 2.99581589958159,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004411669603565187,
      "loss": 5.116,
      "mean_token_accuracy": 0.19368834048509598,
      "num_tokens": 6618415.0,
      "step": 3580
    },
    {
      "entropy": 5.255900573730469,
      "epoch": 3.0,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000440949170121122,
      "loss": 5.1262,
      "mean_token_accuracy": 0.19740014672279357,
      "num_tokens": 6627966.0,
      "step": 3585
    },
    {
      "entropy": 5.307106065750122,
      "epoch": 3.00418410041841,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044073103838901373,
      "loss": 4.7674,
      "mean_token_accuracy": 0.21737998127937316,
      "num_tokens": 6636820.0,
      "step": 3590
    },
    {
      "entropy": 5.154292154312134,
      "epoch": 3.00836820083682,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044051256560907374,
      "loss": 4.7969,
      "mean_token_accuracy": 0.20163721293210984,
      "num_tokens": 6645792.0,
      "step": 3595
    },
    {
      "entropy": 5.330294179916382,
      "epoch": 3.01255230125523,
      "grad_norm": 1.25,
      "learning_rate": 0.00044029375223088394,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2003510445356369,
      "num_tokens": 6654850.0,
      "step": 3600
    },
    {
      "entropy": 5.138308477401734,
      "epoch": 3.01673640167364,
      "grad_norm": 1.125,
      "learning_rate": 0.00044007459870472706,
      "loss": 4.8384,
      "mean_token_accuracy": 0.206855471432209,
      "num_tokens": 6664524.0,
      "step": 3605
    },
    {
      "entropy": 5.0820845603942875,
      "epoch": 3.02092050209205,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004398551054815859,
      "loss": 4.7731,
      "mean_token_accuracy": 0.21012527942657472,
      "num_tokens": 6674330.0,
      "step": 3610
    },
    {
      "entropy": 5.248129415512085,
      "epoch": 3.0251046025104604,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004396352730131422,
      "loss": 4.8629,
      "mean_token_accuracy": 0.20349862724542617,
      "num_tokens": 6683121.0,
      "step": 3615
    },
    {
      "entropy": 5.151824855804444,
      "epoch": 3.0292887029288704,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004394151017517758,
      "loss": 4.8044,
      "mean_token_accuracy": 0.20431603044271468,
      "num_tokens": 6692566.0,
      "step": 3620
    },
    {
      "entropy": 5.2952758312225345,
      "epoch": 3.0334728033472804,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004391945921505638,
      "loss": 4.9109,
      "mean_token_accuracy": 0.19832884669303893,
      "num_tokens": 6701577.0,
      "step": 3625
    },
    {
      "entropy": 5.118036508560181,
      "epoch": 3.0376569037656904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043897374466327944,
      "loss": 4.7795,
      "mean_token_accuracy": 0.21047478467226027,
      "num_tokens": 6710672.0,
      "step": 3630
    },
    {
      "entropy": 5.161023855209351,
      "epoch": 3.0418410041841004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043875255974439145,
      "loss": 4.7791,
      "mean_token_accuracy": 0.2087192416191101,
      "num_tokens": 6719412.0,
      "step": 3635
    },
    {
      "entropy": 5.07971568107605,
      "epoch": 3.0460251046025104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004385310378490627,
      "loss": 4.766,
      "mean_token_accuracy": 0.21152466386556626,
      "num_tokens": 6729169.0,
      "step": 3640
    },
    {
      "entropy": 5.153553295135498,
      "epoch": 3.0502092050209204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043830917943314977,
      "loss": 4.7201,
      "mean_token_accuracy": 0.2165022686123848,
      "num_tokens": 6738549.0,
      "step": 3645
    },
    {
      "entropy": 5.1479654788970945,
      "epoch": 3.0543933054393304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043808698495320154,
      "loss": 4.781,
      "mean_token_accuracy": 0.20508224815130233,
      "num_tokens": 6747909.0,
      "step": 3650
    },
    {
      "entropy": 5.120273447036743,
      "epoch": 3.0585774058577404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043786445486645846,
      "loss": 4.8406,
      "mean_token_accuracy": 0.21035999804735184,
      "num_tokens": 6757125.0,
      "step": 3655
    },
    {
      "entropy": 5.2254575252532955,
      "epoch": 3.062761506276151,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004376415896308518,
      "loss": 4.8399,
      "mean_token_accuracy": 0.20878269672393798,
      "num_tokens": 6765774.0,
      "step": 3660
    },
    {
      "entropy": 5.198799705505371,
      "epoch": 3.066945606694561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043741838970500234,
      "loss": 4.8817,
      "mean_token_accuracy": 0.20897659063339233,
      "num_tokens": 6774980.0,
      "step": 3665
    },
    {
      "entropy": 5.18216233253479,
      "epoch": 3.071129707112971,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004371948555482196,
      "loss": 4.8892,
      "mean_token_accuracy": 0.20027927160263062,
      "num_tokens": 6784187.0,
      "step": 3670
    },
    {
      "entropy": 5.192535448074341,
      "epoch": 3.075313807531381,
      "grad_norm": 1.15625,
      "learning_rate": 0.000436970987620501,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2044747143983841,
      "num_tokens": 6792953.0,
      "step": 3675
    },
    {
      "entropy": 5.162596225738525,
      "epoch": 3.079497907949791,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043674678638253066,
      "loss": 4.7624,
      "mean_token_accuracy": 0.21232210844755173,
      "num_tokens": 6803004.0,
      "step": 3680
    },
    {
      "entropy": 5.125282239913941,
      "epoch": 3.083682008368201,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004365222522956788,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2045421451330185,
      "num_tokens": 6811883.0,
      "step": 3685
    },
    {
      "entropy": 5.143026304244995,
      "epoch": 3.087866108786611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362973858220004,
      "loss": 4.8514,
      "mean_token_accuracy": 0.21020737737417222,
      "num_tokens": 6820327.0,
      "step": 3690
    },
    {
      "entropy": 5.145738220214843,
      "epoch": 3.092050209205021,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043607218742423447,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2086082473397255,
      "num_tokens": 6829412.0,
      "step": 3695
    },
    {
      "entropy": 5.110401916503906,
      "epoch": 3.096234309623431,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043584665756580323,
      "loss": 4.8135,
      "mean_token_accuracy": 0.20601451992988587,
      "num_tokens": 6838997.0,
      "step": 3700
    },
    {
      "entropy": 5.219425630569458,
      "epoch": 3.100418410041841,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004356207967108106,
      "loss": 4.8808,
      "mean_token_accuracy": 0.1984048917889595,
      "num_tokens": 6848957.0,
      "step": 3705
    },
    {
      "entropy": 5.124497985839843,
      "epoch": 3.104602510460251,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004353946053240422,
      "loss": 4.9201,
      "mean_token_accuracy": 0.198494029045105,
      "num_tokens": 6859688.0,
      "step": 3710
    },
    {
      "entropy": 5.228004789352417,
      "epoch": 3.1087866108786613,
      "grad_norm": 1.125,
      "learning_rate": 0.0004351680838709635,
      "loss": 4.9495,
      "mean_token_accuracy": 0.19453879296779633,
      "num_tokens": 6869149.0,
      "step": 3715
    },
    {
      "entropy": 5.217283868789673,
      "epoch": 3.1129707112970713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043494123281771915,
      "loss": 4.9036,
      "mean_token_accuracy": 0.19922375530004502,
      "num_tokens": 6879432.0,
      "step": 3720
    },
    {
      "entropy": 5.145452165603638,
      "epoch": 3.1171548117154813,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043471405263113214,
      "loss": 4.889,
      "mean_token_accuracy": 0.20554056465625764,
      "num_tokens": 6888398.0,
      "step": 3725
    },
    {
      "entropy": 5.200772094726562,
      "epoch": 3.1213389121338913,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004344865437787027,
      "loss": 4.8063,
      "mean_token_accuracy": 0.20465275943279265,
      "num_tokens": 6898082.0,
      "step": 3730
    },
    {
      "entropy": 5.097585439682007,
      "epoch": 3.1255230125523012,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043425870672860774,
      "loss": 4.7834,
      "mean_token_accuracy": 0.21189208179712296,
      "num_tokens": 6907693.0,
      "step": 3735
    },
    {
      "entropy": 5.268562841415405,
      "epoch": 3.1297071129707112,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004340305419496989,
      "loss": 4.9823,
      "mean_token_accuracy": 0.19338904172182084,
      "num_tokens": 6917095.0,
      "step": 3740
    },
    {
      "entropy": 5.17300009727478,
      "epoch": 3.1338912133891212,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004338020499115029,
      "loss": 4.9227,
      "mean_token_accuracy": 0.19877261221408843,
      "num_tokens": 6926952.0,
      "step": 3745
    },
    {
      "entropy": 5.185789775848389,
      "epoch": 3.1380753138075312,
      "grad_norm": 1.25,
      "learning_rate": 0.0004335732310842196,
      "loss": 4.8669,
      "mean_token_accuracy": 0.21069399565458297,
      "num_tokens": 6936704.0,
      "step": 3750
    },
    {
      "entropy": 5.126188707351685,
      "epoch": 3.1422594142259412,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004333440859387212,
      "loss": 4.851,
      "mean_token_accuracy": 0.20845145583152772,
      "num_tokens": 6946614.0,
      "step": 3755
    },
    {
      "entropy": 5.220112037658692,
      "epoch": 3.1464435146443517,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043311461494655165,
      "loss": 4.86,
      "mean_token_accuracy": 0.21031337976455688,
      "num_tokens": 6955980.0,
      "step": 3760
    },
    {
      "entropy": 5.23923077583313,
      "epoch": 3.1506276150627617,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004328848185799254,
      "loss": 4.86,
      "mean_token_accuracy": 0.2033395618200302,
      "num_tokens": 6964108.0,
      "step": 3765
    },
    {
      "entropy": 5.147957849502563,
      "epoch": 3.1548117154811717,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004326546973117264,
      "loss": 4.8568,
      "mean_token_accuracy": 0.21318745166063308,
      "num_tokens": 6972895.0,
      "step": 3770
    },
    {
      "entropy": 5.144702339172364,
      "epoch": 3.1589958158995817,
      "grad_norm": 1.25,
      "learning_rate": 0.00043242425161550723,
      "loss": 4.8673,
      "mean_token_accuracy": 0.20916283875703812,
      "num_tokens": 6982222.0,
      "step": 3775
    },
    {
      "entropy": 5.250490617752075,
      "epoch": 3.1631799163179917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043219348196548807,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2065814509987831,
      "num_tokens": 6990820.0,
      "step": 3780
    },
    {
      "entropy": 5.1154365062713625,
      "epoch": 3.1673640167364017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004319623888365557,
      "loss": 4.8932,
      "mean_token_accuracy": 0.20653683990240096,
      "num_tokens": 7000846.0,
      "step": 3785
    },
    {
      "entropy": 5.156445837020874,
      "epoch": 3.1715481171548117,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043173097270426275,
      "loss": 4.818,
      "mean_token_accuracy": 0.19972968697547913,
      "num_tokens": 7009086.0,
      "step": 3790
    },
    {
      "entropy": 5.341412830352783,
      "epoch": 3.1757322175732217,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043149923404482634,
      "loss": 4.8918,
      "mean_token_accuracy": 0.20719874650239944,
      "num_tokens": 7017752.0,
      "step": 3795
    },
    {
      "entropy": 5.087004375457764,
      "epoch": 3.1799163179916317,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043126717333512733,
      "loss": 4.8702,
      "mean_token_accuracy": 0.21366522163152696,
      "num_tokens": 7026535.0,
      "step": 3800
    },
    {
      "entropy": 5.183447265625,
      "epoch": 3.1841004184100417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004310347910527094,
      "loss": 4.9067,
      "mean_token_accuracy": 0.20852705091238022,
      "num_tokens": 7035462.0,
      "step": 3805
    },
    {
      "entropy": 5.155086708068848,
      "epoch": 3.1882845188284517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004308020876757778,
      "loss": 4.8141,
      "mean_token_accuracy": 0.20730070620775223,
      "num_tokens": 7043978.0,
      "step": 3810
    },
    {
      "entropy": 5.098757219314575,
      "epoch": 3.192468619246862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000430569063683199,
      "loss": 4.8421,
      "mean_token_accuracy": 0.20978484451770782,
      "num_tokens": 7053757.0,
      "step": 3815
    },
    {
      "entropy": 5.060821485519409,
      "epoch": 3.196652719665272,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004303357195544986,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2038672685623169,
      "num_tokens": 7062491.0,
      "step": 3820
    },
    {
      "entropy": 5.155342483520508,
      "epoch": 3.200836820083682,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004301020557698615,
      "loss": 4.8104,
      "mean_token_accuracy": 0.21007771641016007,
      "num_tokens": 7072157.0,
      "step": 3825
    },
    {
      "entropy": 5.083573055267334,
      "epoch": 3.205020920502092,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042986807281013013,
      "loss": 4.787,
      "mean_token_accuracy": 0.2048204466700554,
      "num_tokens": 7080631.0,
      "step": 3830
    },
    {
      "entropy": 5.189060211181641,
      "epoch": 3.209205020920502,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004296337711568037,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2146065503358841,
      "num_tokens": 7089944.0,
      "step": 3835
    },
    {
      "entropy": 5.165537309646607,
      "epoch": 3.213389121338912,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042939915129203745,
      "loss": 4.9119,
      "mean_token_accuracy": 0.20280353128910064,
      "num_tokens": 7099190.0,
      "step": 3840
    },
    {
      "entropy": 5.042298698425293,
      "epoch": 3.217573221757322,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004291642136986413,
      "loss": 4.8578,
      "mean_token_accuracy": 0.20252264589071273,
      "num_tokens": 7107944.0,
      "step": 3845
    },
    {
      "entropy": 5.1508770942687985,
      "epoch": 3.221757322175732,
      "grad_norm": 1.1875,
      "learning_rate": 0.000428928958860079,
      "loss": 4.7282,
      "mean_token_accuracy": 0.2113970160484314,
      "num_tokens": 7116953.0,
      "step": 3850
    },
    {
      "entropy": 5.040340375900269,
      "epoch": 3.225941422594142,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004286933872604673,
      "loss": 4.7871,
      "mean_token_accuracy": 0.20862122923135756,
      "num_tokens": 7126236.0,
      "step": 3855
    },
    {
      "entropy": 5.137906646728515,
      "epoch": 3.2301255230125525,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004284574993845746,
      "loss": 4.8376,
      "mean_token_accuracy": 0.20907163321971894,
      "num_tokens": 7136286.0,
      "step": 3860
    },
    {
      "entropy": 5.138406753540039,
      "epoch": 3.2343096234309625,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042822129571782025,
      "loss": 4.9046,
      "mean_token_accuracy": 0.2057375729084015,
      "num_tokens": 7145551.0,
      "step": 3865
    },
    {
      "entropy": 5.142768621444702,
      "epoch": 3.2384937238493725,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042798477674627343,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2094752386212349,
      "num_tokens": 7155486.0,
      "step": 3870
    },
    {
      "entropy": 5.1550507068634035,
      "epoch": 3.2426778242677825,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042774794295665213,
      "loss": 4.8394,
      "mean_token_accuracy": 0.20709673911333085,
      "num_tokens": 7163984.0,
      "step": 3875
    },
    {
      "entropy": 5.073590469360352,
      "epoch": 3.2468619246861925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004275107948363223,
      "loss": 4.8255,
      "mean_token_accuracy": 0.20394547134637833,
      "num_tokens": 7172854.0,
      "step": 3880
    },
    {
      "entropy": 5.051211786270142,
      "epoch": 3.2510460251046025,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004272733328732965,
      "loss": 4.8097,
      "mean_token_accuracy": 0.20123495161533356,
      "num_tokens": 7182202.0,
      "step": 3885
    },
    {
      "entropy": 5.189564371109009,
      "epoch": 3.2552301255230125,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042703555755623337,
      "loss": 4.9453,
      "mean_token_accuracy": 0.19725567251443862,
      "num_tokens": 7192062.0,
      "step": 3890
    },
    {
      "entropy": 5.240469741821289,
      "epoch": 3.2594142259414225,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267974693744362,
      "loss": 4.9448,
      "mean_token_accuracy": 0.19944138526916505,
      "num_tokens": 7201766.0,
      "step": 3895
    },
    {
      "entropy": 5.149724340438842,
      "epoch": 3.2635983263598325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004265590688178523,
      "loss": 4.8514,
      "mean_token_accuracy": 0.20332693308591843,
      "num_tokens": 7211361.0,
      "step": 3900
    },
    {
      "entropy": 5.131576490402222,
      "epoch": 3.2677824267782425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004263203563770716,
      "loss": 4.8244,
      "mean_token_accuracy": 0.211550971865654,
      "num_tokens": 7220221.0,
      "step": 3905
    },
    {
      "entropy": 5.139615726470947,
      "epoch": 3.2719665271966525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000426081332543326,
      "loss": 4.9433,
      "mean_token_accuracy": 0.1962783083319664,
      "num_tokens": 7230134.0,
      "step": 3910
    },
    {
      "entropy": 5.044312191009522,
      "epoch": 3.276150627615063,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004258419978084881,
      "loss": 4.7615,
      "mean_token_accuracy": 0.2103862628340721,
      "num_tokens": 7238713.0,
      "step": 3915
    },
    {
      "entropy": 5.153055715560913,
      "epoch": 3.280334728033473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004256023526650702,
      "loss": 4.8643,
      "mean_token_accuracy": 0.20162613242864608,
      "num_tokens": 7247603.0,
      "step": 3920
    },
    {
      "entropy": 5.111010646820068,
      "epoch": 3.284518828451883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004253623976062236,
      "loss": 4.9139,
      "mean_token_accuracy": 0.20576744973659516,
      "num_tokens": 7257452.0,
      "step": 3925
    },
    {
      "entropy": 5.260893726348877,
      "epoch": 3.288702928870293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004251221331257372,
      "loss": 4.9114,
      "mean_token_accuracy": 0.20624379366636275,
      "num_tokens": 7266401.0,
      "step": 3930
    },
    {
      "entropy": 5.113575983047485,
      "epoch": 3.292887029288703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004248815597180366,
      "loss": 4.8568,
      "mean_token_accuracy": 0.20980830490589142,
      "num_tokens": 7275836.0,
      "step": 3935
    },
    {
      "entropy": 5.1434567928314205,
      "epoch": 3.297071129707113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042464067787818333,
      "loss": 4.9104,
      "mean_token_accuracy": 0.20328453928232193,
      "num_tokens": 7284854.0,
      "step": 3940
    },
    {
      "entropy": 5.127180099487305,
      "epoch": 3.301255230125523,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004243994881018735,
      "loss": 4.8693,
      "mean_token_accuracy": 0.20316764414310456,
      "num_tokens": 7293421.0,
      "step": 3945
    },
    {
      "entropy": 5.1008624076843265,
      "epoch": 3.305439330543933,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004241579908854367,
      "loss": 4.8025,
      "mean_token_accuracy": 0.20773349106311798,
      "num_tokens": 7302184.0,
      "step": 3950
    },
    {
      "entropy": 5.18611330986023,
      "epoch": 3.309623430962343,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042391618672583546,
      "loss": 4.9577,
      "mean_token_accuracy": 0.20182376950979233,
      "num_tokens": 7311596.0,
      "step": 3955
    },
    {
      "entropy": 5.101362419128418,
      "epoch": 3.3138075313807533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042367407612066383,
      "loss": 4.816,
      "mean_token_accuracy": 0.21087881773710251,
      "num_tokens": 7319866.0,
      "step": 3960
    },
    {
      "entropy": 5.126628875732422,
      "epoch": 3.3179916317991633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004234316595681468,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2091270938515663,
      "num_tokens": 7329327.0,
      "step": 3965
    },
    {
      "entropy": 5.227209281921387,
      "epoch": 3.3221757322175733,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004231889375671382,
      "loss": 5.0337,
      "mean_token_accuracy": 0.19003052413463592,
      "num_tokens": 7339717.0,
      "step": 3970
    },
    {
      "entropy": 5.191570091247558,
      "epoch": 3.3263598326359833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042294591061712124,
      "loss": 4.8485,
      "mean_token_accuracy": 0.20856435000896453,
      "num_tokens": 7348850.0,
      "step": 3975
    },
    {
      "entropy": 5.077865982055664,
      "epoch": 3.3305439330543933,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042270257921820623,
      "loss": 4.8025,
      "mean_token_accuracy": 0.21396382451057433,
      "num_tokens": 7357081.0,
      "step": 3980
    },
    {
      "entropy": 4.985082817077637,
      "epoch": 3.3347280334728033,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00042245894387112987,
      "loss": 4.8151,
      "mean_token_accuracy": 0.21312166899442672,
      "num_tokens": 7365762.0,
      "step": 3985
    },
    {
      "entropy": 5.124319648742675,
      "epoch": 3.3389121338912133,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004222150050772548,
      "loss": 4.8817,
      "mean_token_accuracy": 0.20862279534339906,
      "num_tokens": 7374870.0,
      "step": 3990
    },
    {
      "entropy": 5.0922142505645756,
      "epoch": 3.3430962343096233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004219707633385677,
      "loss": 4.7662,
      "mean_token_accuracy": 0.20609999001026152,
      "num_tokens": 7383836.0,
      "step": 3995
    },
    {
      "entropy": 5.08710560798645,
      "epoch": 3.3472803347280333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042172621915767886,
      "loss": 4.8635,
      "mean_token_accuracy": 0.20860992074012757,
      "num_tokens": 7394059.0,
      "step": 4000
    },
    {
      "epoch": 3.3472803347280333,
      "eval_entropy": 4.985736576326055,
      "eval_loss": 5.46319055557251,
      "eval_mean_token_accuracy": 0.18651489741292712,
      "eval_num_tokens": 7394059.0,
      "eval_runtime": 2.7561,
      "eval_samples_per_second": 1417.58,
      "eval_steps_per_second": 177.424,
      "step": 4000
    },
    {
      "entropy": 5.0861451625823975,
      "epoch": 3.3514644351464433,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042148137303782077,
      "loss": 4.8307,
      "mean_token_accuracy": 0.21213750094175338,
      "num_tokens": 7402947.0,
      "step": 4005
    },
    {
      "entropy": 5.149941492080688,
      "epoch": 3.3556485355648533,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004212362254828475,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2033420354127884,
      "num_tokens": 7413430.0,
      "step": 4010
    },
    {
      "entropy": 5.202378416061402,
      "epoch": 3.3598326359832638,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042099077699723326,
      "loss": 4.8931,
      "mean_token_accuracy": 0.20540930330753326,
      "num_tokens": 7423239.0,
      "step": 4015
    },
    {
      "entropy": 5.094257879257202,
      "epoch": 3.3640167364016738,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004207450280860716,
      "loss": 4.8709,
      "mean_token_accuracy": 0.20930316001176835,
      "num_tokens": 7432519.0,
      "step": 4020
    },
    {
      "entropy": 5.083430433273316,
      "epoch": 3.3682008368200838,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004204989792550742,
      "loss": 4.8364,
      "mean_token_accuracy": 0.20555925667285918,
      "num_tokens": 7442089.0,
      "step": 4025
    },
    {
      "entropy": 5.190230035781861,
      "epoch": 3.3723849372384938,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042025263101057016,
      "loss": 4.8583,
      "mean_token_accuracy": 0.20592181831598283,
      "num_tokens": 7451059.0,
      "step": 4030
    },
    {
      "entropy": 5.11042799949646,
      "epoch": 3.3765690376569037,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042000598385950427,
      "loss": 4.8144,
      "mean_token_accuracy": 0.21088250875473022,
      "num_tokens": 7458960.0,
      "step": 4035
    },
    {
      "entropy": 5.126355361938477,
      "epoch": 3.3807531380753137,
      "grad_norm": 1.15625,
      "learning_rate": 0.000419759038309437,
      "loss": 5.0011,
      "mean_token_accuracy": 0.19898598790168762,
      "num_tokens": 7468863.0,
      "step": 4040
    },
    {
      "entropy": 5.155329084396362,
      "epoch": 3.3849372384937237,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041951179486854253,
      "loss": 4.9566,
      "mean_token_accuracy": 0.20642842054367067,
      "num_tokens": 7477204.0,
      "step": 4045
    },
    {
      "entropy": 5.11786527633667,
      "epoch": 3.3891213389121337,
      "grad_norm": 1.140625,
      "learning_rate": 0.000419264254045608,
      "loss": 4.7906,
      "mean_token_accuracy": 0.20480405390262604,
      "num_tokens": 7487367.0,
      "step": 4050
    },
    {
      "entropy": 4.98615870475769,
      "epoch": 3.393305439330544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004190164163500326,
      "loss": 4.7855,
      "mean_token_accuracy": 0.2164631724357605,
      "num_tokens": 7495808.0,
      "step": 4055
    },
    {
      "entropy": 5.120810890197754,
      "epoch": 3.397489539748954,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004187682822918266,
      "loss": 4.9104,
      "mean_token_accuracy": 0.1951922059059143,
      "num_tokens": 7506119.0,
      "step": 4060
    },
    {
      "entropy": 5.126148271560669,
      "epoch": 3.401673640167364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041851985238161,
      "loss": 4.8846,
      "mean_token_accuracy": 0.20199777334928512,
      "num_tokens": 7514931.0,
      "step": 4065
    },
    {
      "entropy": 5.0606081008911135,
      "epoch": 3.405857740585774,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041827112713061155,
      "loss": 4.6942,
      "mean_token_accuracy": 0.22149741649627686,
      "num_tokens": 7523753.0,
      "step": 4070
    },
    {
      "entropy": 5.064794969558716,
      "epoch": 3.410041841004184,
      "grad_norm": 1.25,
      "learning_rate": 0.00041802210705066775,
      "loss": 4.7682,
      "mean_token_accuracy": 0.21452270895242692,
      "num_tokens": 7532447.0,
      "step": 4075
    },
    {
      "entropy": 5.082556819915771,
      "epoch": 3.414225941422594,
      "grad_norm": 1.28125,
      "learning_rate": 0.000417772792654222,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20070126205682753,
      "num_tokens": 7541388.0,
      "step": 4080
    },
    {
      "entropy": 5.150190734863282,
      "epoch": 3.418410041841004,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041752318445432323,
      "loss": 4.9239,
      "mean_token_accuracy": 0.19895025342702866,
      "num_tokens": 7550422.0,
      "step": 4085
    },
    {
      "entropy": 5.1157999515533445,
      "epoch": 3.422594142259414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000417273282964625,
      "loss": 4.7817,
      "mean_token_accuracy": 0.21615640223026275,
      "num_tokens": 7559522.0,
      "step": 4090
    },
    {
      "entropy": 5.112621021270752,
      "epoch": 3.426778242677824,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004170230886993844,
      "loss": 4.902,
      "mean_token_accuracy": 0.19627136141061782,
      "num_tokens": 7568785.0,
      "step": 4095
    },
    {
      "entropy": 5.065780305862427,
      "epoch": 3.430962343096234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041677260217346093,
      "loss": 4.8267,
      "mean_token_accuracy": 0.21189531534910203,
      "num_tokens": 7579013.0,
      "step": 4100
    },
    {
      "entropy": 5.111838912963867,
      "epoch": 3.435146443514644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041652182390231556,
      "loss": 4.8605,
      "mean_token_accuracy": 0.20376810282468796,
      "num_tokens": 7588577.0,
      "step": 4105
    },
    {
      "entropy": 5.090034055709839,
      "epoch": 3.439330543933054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041627075440200977,
      "loss": 4.7785,
      "mean_token_accuracy": 0.2134248748421669,
      "num_tokens": 7597702.0,
      "step": 4110
    },
    {
      "entropy": 5.063868522644043,
      "epoch": 3.4435146443514646,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004160193941892043,
      "loss": 4.8422,
      "mean_token_accuracy": 0.20544036775827407,
      "num_tokens": 7607156.0,
      "step": 4115
    },
    {
      "entropy": 5.107327699661255,
      "epoch": 3.4476987447698746,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004157677437811578,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2133098989725113,
      "num_tokens": 7615936.0,
      "step": 4120
    },
    {
      "entropy": 5.157761907577514,
      "epoch": 3.4518828451882846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004155158036957265,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2041935294866562,
      "num_tokens": 7626568.0,
      "step": 4125
    },
    {
      "entropy": 5.1917156219482425,
      "epoch": 3.4560669456066946,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004152635744513626,
      "loss": 4.8458,
      "mean_token_accuracy": 0.21043034940958022,
      "num_tokens": 7635102.0,
      "step": 4130
    },
    {
      "entropy": 5.013117122650146,
      "epoch": 3.4602510460251046,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004150110565671133,
      "loss": 4.7563,
      "mean_token_accuracy": 0.22028307914733886,
      "num_tokens": 7644204.0,
      "step": 4135
    },
    {
      "entropy": 5.116233921051025,
      "epoch": 3.4644351464435146,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041475825056261985,
      "loss": 4.8843,
      "mean_token_accuracy": 0.20191183984279631,
      "num_tokens": 7653585.0,
      "step": 4140
    },
    {
      "entropy": 5.189920139312744,
      "epoch": 3.4686192468619246,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004145051569581162,
      "loss": 4.8768,
      "mean_token_accuracy": 0.19981294125318527,
      "num_tokens": 7662734.0,
      "step": 4145
    },
    {
      "entropy": 5.151976680755615,
      "epoch": 3.4728033472803346,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004142517762744286,
      "loss": 4.8322,
      "mean_token_accuracy": 0.2101568877696991,
      "num_tokens": 7671682.0,
      "step": 4150
    },
    {
      "entropy": 5.149965047836304,
      "epoch": 3.476987447698745,
      "grad_norm": 1.25,
      "learning_rate": 0.0004139981090329735,
      "loss": 4.9305,
      "mean_token_accuracy": 0.20690995901823045,
      "num_tokens": 7681551.0,
      "step": 4155
    },
    {
      "entropy": 5.124179649353027,
      "epoch": 3.481171548117155,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004137441557557575,
      "loss": 4.9015,
      "mean_token_accuracy": 0.20703185349702835,
      "num_tokens": 7690943.0,
      "step": 4160
    },
    {
      "entropy": 5.073544120788574,
      "epoch": 3.485355648535565,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041348991696537533,
      "loss": 4.793,
      "mean_token_accuracy": 0.21441176235675813,
      "num_tokens": 7700376.0,
      "step": 4165
    },
    {
      "entropy": 5.148408889770508,
      "epoch": 3.489539748953975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004132353931850098,
      "loss": 4.9043,
      "mean_token_accuracy": 0.19997784048318862,
      "num_tokens": 7709845.0,
      "step": 4170
    },
    {
      "entropy": 5.187413167953491,
      "epoch": 3.493723849372385,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004129805849384299,
      "loss": 4.9811,
      "mean_token_accuracy": 0.20034596025943757,
      "num_tokens": 7719908.0,
      "step": 4175
    },
    {
      "entropy": 5.0858958721160885,
      "epoch": 3.497907949790795,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004127254927499898,
      "loss": 4.8971,
      "mean_token_accuracy": 0.20275071114301682,
      "num_tokens": 7729543.0,
      "step": 4180
    },
    {
      "entropy": 5.146850728988648,
      "epoch": 3.502092050209205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004124701171446286,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2049371689558029,
      "num_tokens": 7739371.0,
      "step": 4185
    },
    {
      "entropy": 5.10733208656311,
      "epoch": 3.506276150627615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041221445864786794,
      "loss": 4.8187,
      "mean_token_accuracy": 0.20833780020475387,
      "num_tokens": 7748027.0,
      "step": 4190
    },
    {
      "entropy": 5.0194145202636715,
      "epoch": 3.510460251046025,
      "grad_norm": 1.171875,
      "learning_rate": 0.000411958517785812,
      "loss": 4.7581,
      "mean_token_accuracy": 0.21444778442382811,
      "num_tokens": 7756588.0,
      "step": 4195
    },
    {
      "entropy": 5.1101737976074215,
      "epoch": 3.514644351464435,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041170229508514584,
      "loss": 4.7468,
      "mean_token_accuracy": 0.20663689076900482,
      "num_tokens": 7765222.0,
      "step": 4200
    },
    {
      "entropy": 5.092128467559815,
      "epoch": 3.518828451882845,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004114457910731347,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2106161281466484,
      "num_tokens": 7774139.0,
      "step": 4205
    },
    {
      "entropy": 5.026708507537842,
      "epoch": 3.523012552301255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004111890062776226,
      "loss": 4.7452,
      "mean_token_accuracy": 0.20863335132598876,
      "num_tokens": 7783597.0,
      "step": 4210
    },
    {
      "entropy": 5.045945262908935,
      "epoch": 3.5271966527196654,
      "grad_norm": 1.25,
      "learning_rate": 0.0004109319412270314,
      "loss": 4.8409,
      "mean_token_accuracy": 0.21286385655403137,
      "num_tokens": 7792711.0,
      "step": 4215
    },
    {
      "entropy": 5.2014275074005125,
      "epoch": 3.5313807531380754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004106745964503594,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2045663923025131,
      "num_tokens": 7801878.0,
      "step": 4220
    },
    {
      "entropy": 5.030341577529907,
      "epoch": 3.5355648535564854,
      "grad_norm": 1.109375,
      "learning_rate": 0.000410416972477181,
      "loss": 4.798,
      "mean_token_accuracy": 0.21313095539808274,
      "num_tokens": 7810484.0,
      "step": 4225
    },
    {
      "entropy": 5.141058349609375,
      "epoch": 3.5397489539748954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041015906983764495,
      "loss": 4.9184,
      "mean_token_accuracy": 0.204750457406044,
      "num_tokens": 7820521.0,
      "step": 4230
    },
    {
      "entropy": 5.168326950073242,
      "epoch": 3.5439330543933054,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040990088906247326,
      "loss": 4.8305,
      "mean_token_accuracy": 0.20982770025730133,
      "num_tokens": 7829218.0,
      "step": 4235
    },
    {
      "entropy": 5.05530309677124,
      "epoch": 3.5481171548117154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004096424306829605,
      "loss": 4.8607,
      "mean_token_accuracy": 0.20657587945461273,
      "num_tokens": 7837795.0,
      "step": 4240
    },
    {
      "entropy": 5.165108299255371,
      "epoch": 3.5523012552301254,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004093836952309726,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2080150991678238,
      "num_tokens": 7846973.0,
      "step": 4245
    },
    {
      "entropy": 5.1759199619293215,
      "epoch": 3.5564853556485354,
      "grad_norm": 1.25,
      "learning_rate": 0.0004091246832389453,
      "loss": 4.9133,
      "mean_token_accuracy": 0.20285059064626693,
      "num_tokens": 7856627.0,
      "step": 4250
    },
    {
      "entropy": 5.04175763130188,
      "epoch": 3.560669456066946,
      "grad_norm": 1.125,
      "learning_rate": 0.0004088653952398837,
      "loss": 4.8365,
      "mean_token_accuracy": 0.2079515963792801,
      "num_tokens": 7866308.0,
      "step": 4255
    },
    {
      "entropy": 5.077424430847168,
      "epoch": 3.564853556485356,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040860583176736097,
      "loss": 4.8262,
      "mean_token_accuracy": 0.20950917750597,
      "num_tokens": 7875350.0,
      "step": 4260
    },
    {
      "entropy": 5.077731943130493,
      "epoch": 3.569037656903766,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004083459933555169,
      "loss": 4.8298,
      "mean_token_accuracy": 0.21197184771299363,
      "num_tokens": 7884276.0,
      "step": 4265
    },
    {
      "entropy": 5.107650852203369,
      "epoch": 3.573221757322176,
      "grad_norm": 1.125,
      "learning_rate": 0.0004080858805390571,
      "loss": 4.8995,
      "mean_token_accuracy": 0.20368795841932297,
      "num_tokens": 7893836.0,
      "step": 4270
    },
    {
      "entropy": 5.1801831245422365,
      "epoch": 3.577405857740586,
      "grad_norm": 1.1875,
      "learning_rate": 0.000407825493853252,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2066332772374153,
      "num_tokens": 7903249.0,
      "step": 4275
    },
    {
      "entropy": 5.120818758010865,
      "epoch": 3.581589958158996,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040756483383393556,
      "loss": 4.9225,
      "mean_token_accuracy": 0.203904427587986,
      "num_tokens": 7912217.0,
      "step": 4280
    },
    {
      "entropy": 5.086220693588257,
      "epoch": 3.585774058577406,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004073039010175041,
      "loss": 4.9629,
      "mean_token_accuracy": 0.2006915271282196,
      "num_tokens": 7922241.0,
      "step": 4285
    },
    {
      "entropy": 5.186107063293457,
      "epoch": 3.589958158995816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040704269594091556,
      "loss": 4.871,
      "mean_token_accuracy": 0.2071731850504875,
      "num_tokens": 7931866.0,
      "step": 4290
    },
    {
      "entropy": 5.1948097229003904,
      "epoch": 3.594142259414226,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004067812191416878,
      "loss": 4.8986,
      "mean_token_accuracy": 0.20188879817724228,
      "num_tokens": 7941787.0,
      "step": 4295
    },
    {
      "entropy": 5.117114782333374,
      "epoch": 3.598326359832636,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004065194711578982,
      "loss": 4.9136,
      "mean_token_accuracy": 0.20612358748912812,
      "num_tokens": 7950832.0,
      "step": 4300
    },
    {
      "entropy": 5.1436402797698975,
      "epoch": 3.602510460251046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004062574525281819,
      "loss": 4.8968,
      "mean_token_accuracy": 0.20418329536914825,
      "num_tokens": 7960059.0,
      "step": 4305
    },
    {
      "entropy": 5.134668636322021,
      "epoch": 3.606694560669456,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040599516379173113,
      "loss": 4.8861,
      "mean_token_accuracy": 0.20005105286836625,
      "num_tokens": 7969411.0,
      "step": 4310
    },
    {
      "entropy": 5.116720819473267,
      "epoch": 3.6108786610878663,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040573260548829417,
      "loss": 4.8842,
      "mean_token_accuracy": 0.20714796483516693,
      "num_tokens": 7978446.0,
      "step": 4315
    },
    {
      "entropy": 5.015983247756958,
      "epoch": 3.6150627615062763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004054697781581734,
      "loss": 4.7564,
      "mean_token_accuracy": 0.20707345455884935,
      "num_tokens": 7987117.0,
      "step": 4320
    },
    {
      "entropy": 5.107610368728638,
      "epoch": 3.6192468619246863,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004052066823422255,
      "loss": 4.8457,
      "mean_token_accuracy": 0.20731030106544496,
      "num_tokens": 7995989.0,
      "step": 4325
    },
    {
      "entropy": 5.082603549957275,
      "epoch": 3.6234309623430963,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004049433185818593,
      "loss": 4.9229,
      "mean_token_accuracy": 0.20139078050851822,
      "num_tokens": 8005486.0,
      "step": 4330
    },
    {
      "entropy": 5.078189849853516,
      "epoch": 3.6276150627615062,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040467968741903507,
      "loss": 4.7734,
      "mean_token_accuracy": 0.21146121770143508,
      "num_tokens": 8014853.0,
      "step": 4335
    },
    {
      "entropy": 5.125499629974366,
      "epoch": 3.6317991631799162,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004044157893962633,
      "loss": 4.8429,
      "mean_token_accuracy": 0.21341473758220672,
      "num_tokens": 8023931.0,
      "step": 4340
    },
    {
      "entropy": 5.103664350509644,
      "epoch": 3.6359832635983262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004041516250566037,
      "loss": 4.8827,
      "mean_token_accuracy": 0.21109170764684676,
      "num_tokens": 8032659.0,
      "step": 4345
    },
    {
      "entropy": 5.145339918136597,
      "epoch": 3.6401673640167362,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040388719494366404,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20322352200746535,
      "num_tokens": 8041857.0,
      "step": 4350
    },
    {
      "entropy": 5.213011121749878,
      "epoch": 3.6443514644351467,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040362249960159904,
      "loss": 4.9744,
      "mean_token_accuracy": 0.20129672586917877,
      "num_tokens": 8051186.0,
      "step": 4355
    },
    {
      "entropy": 5.097787809371948,
      "epoch": 3.6485355648535567,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040335753957510906,
      "loss": 4.8951,
      "mean_token_accuracy": 0.19750130027532578,
      "num_tokens": 8060396.0,
      "step": 4360
    },
    {
      "entropy": 5.073494338989258,
      "epoch": 3.6527196652719667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040309231540943937,
      "loss": 4.8079,
      "mean_token_accuracy": 0.21439304649829866,
      "num_tokens": 8068106.0,
      "step": 4365
    },
    {
      "entropy": 5.120754909515381,
      "epoch": 3.6569037656903767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004028268276503784,
      "loss": 4.8483,
      "mean_token_accuracy": 0.2051659792661667,
      "num_tokens": 8077878.0,
      "step": 4370
    },
    {
      "entropy": 5.1757666110992435,
      "epoch": 3.6610878661087867,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004025610768442576,
      "loss": 4.9738,
      "mean_token_accuracy": 0.1989295795559883,
      "num_tokens": 8086968.0,
      "step": 4375
    },
    {
      "entropy": 5.131380891799926,
      "epoch": 3.6652719665271967,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004022950635379492,
      "loss": 4.7689,
      "mean_token_accuracy": 0.22497869580984114,
      "num_tokens": 8095999.0,
      "step": 4380
    },
    {
      "entropy": 5.033745956420899,
      "epoch": 3.6694560669456067,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040202878827886587,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2179463043808937,
      "num_tokens": 8105376.0,
      "step": 4385
    },
    {
      "entropy": 5.106619548797608,
      "epoch": 3.6736401673640167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004017622516149592,
      "loss": 4.8912,
      "mean_token_accuracy": 0.20658889710903167,
      "num_tokens": 8115024.0,
      "step": 4390
    },
    {
      "entropy": 5.121318626403808,
      "epoch": 3.6778242677824267,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004014954540947189,
      "loss": 4.8889,
      "mean_token_accuracy": 0.20341428220272065,
      "num_tokens": 8124306.0,
      "step": 4395
    },
    {
      "entropy": 5.090257740020752,
      "epoch": 3.6820083682008367,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040122839626717137,
      "loss": 4.8792,
      "mean_token_accuracy": 0.21472734659910203,
      "num_tokens": 8133151.0,
      "step": 4400
    },
    {
      "entropy": 5.089275932312011,
      "epoch": 3.6861924686192467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004009610786818786,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2134702667593956,
      "num_tokens": 8141745.0,
      "step": 4405
    },
    {
      "entropy": 5.059766674041748,
      "epoch": 3.6903765690376567,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040069350188893735,
      "loss": 4.7383,
      "mean_token_accuracy": 0.215824893116951,
      "num_tokens": 8150884.0,
      "step": 4410
    },
    {
      "entropy": 5.093609046936035,
      "epoch": 3.694560669456067,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040042566643897745,
      "loss": 4.8414,
      "mean_token_accuracy": 0.20642719566822051,
      "num_tokens": 8160138.0,
      "step": 4415
    },
    {
      "entropy": 5.150986671447754,
      "epoch": 3.698744769874477,
      "grad_norm": 1.25,
      "learning_rate": 0.0004001575728831613,
      "loss": 4.9199,
      "mean_token_accuracy": 0.20834485590457916,
      "num_tokens": 8169655.0,
      "step": 4420
    },
    {
      "entropy": 5.213423728942871,
      "epoch": 3.702928870292887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003998892217731824,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21138543933629989,
      "num_tokens": 8178846.0,
      "step": 4425
    },
    {
      "entropy": 5.066442012786865,
      "epoch": 3.707112970711297,
      "grad_norm": 1.203125,
      "learning_rate": 0.000399620613661264,
      "loss": 4.833,
      "mean_token_accuracy": 0.2046512633562088,
      "num_tokens": 8187525.0,
      "step": 4430
    },
    {
      "entropy": 5.121527099609375,
      "epoch": 3.711297071129707,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003993517491001586,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20239897221326827,
      "num_tokens": 8196855.0,
      "step": 4435
    },
    {
      "entropy": 5.039672756195069,
      "epoch": 3.715481171548117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003990826286431462,
      "loss": 4.8132,
      "mean_token_accuracy": 0.204982253909111,
      "num_tokens": 8205530.0,
      "step": 4440
    },
    {
      "entropy": 5.113843774795532,
      "epoch": 3.719665271966527,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003988132528440333,
      "loss": 4.8777,
      "mean_token_accuracy": 0.2079408660531044,
      "num_tokens": 8215066.0,
      "step": 4445
    },
    {
      "entropy": 5.095618438720703,
      "epoch": 3.723849372384937,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003985436222571521,
      "loss": 4.8609,
      "mean_token_accuracy": 0.20079515129327774,
      "num_tokens": 8224585.0,
      "step": 4450
    },
    {
      "entropy": 5.1642131328582765,
      "epoch": 3.7280334728033475,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000398273737437359,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20152941048145295,
      "num_tokens": 8233334.0,
      "step": 4455
    },
    {
      "entropy": 5.148773813247681,
      "epoch": 3.7322175732217575,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039800359894003355,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2125294029712677,
      "num_tokens": 8241636.0,
      "step": 4460
    },
    {
      "entropy": 5.136564540863037,
      "epoch": 3.7364016736401675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003977332073210774,
      "loss": 4.9004,
      "mean_token_accuracy": 0.2041999727487564,
      "num_tokens": 8250273.0,
      "step": 4465
    },
    {
      "entropy": 5.139496755599976,
      "epoch": 3.7405857740585775,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003974625631369129,
      "loss": 4.9344,
      "mean_token_accuracy": 0.1999513953924179,
      "num_tokens": 8260349.0,
      "step": 4470
    },
    {
      "entropy": 5.108516979217529,
      "epoch": 3.7447698744769875,
      "grad_norm": 1.125,
      "learning_rate": 0.0003971916669444824,
      "loss": 4.929,
      "mean_token_accuracy": 0.20351004600524902,
      "num_tokens": 8270505.0,
      "step": 4475
    },
    {
      "entropy": 5.197675275802612,
      "epoch": 3.7489539748953975,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039692051930124663,
      "loss": 4.8964,
      "mean_token_accuracy": 0.20123675167560579,
      "num_tokens": 8278651.0,
      "step": 4480
    },
    {
      "entropy": 5.111096525192261,
      "epoch": 3.7531380753138075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003966491207651838,
      "loss": 4.8475,
      "mean_token_accuracy": 0.21011456847190857,
      "num_tokens": 8287173.0,
      "step": 4485
    },
    {
      "entropy": 5.171531820297242,
      "epoch": 3.7573221757322175,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039637747189478866,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2036084309220314,
      "num_tokens": 8295785.0,
      "step": 4490
    },
    {
      "entropy": 5.143098258972168,
      "epoch": 3.7615062761506275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003961055732490708,
      "loss": 4.9552,
      "mean_token_accuracy": 0.2031768247485161,
      "num_tokens": 8305534.0,
      "step": 4495
    },
    {
      "entropy": 5.098934125900269,
      "epoch": 3.7656903765690375,
      "grad_norm": 1.25,
      "learning_rate": 0.00039583342538755374,
      "loss": 4.844,
      "mean_token_accuracy": 0.2070768177509308,
      "num_tokens": 8314232.0,
      "step": 4500
    },
    {
      "epoch": 3.7656903765690375,
      "eval_entropy": 4.91763174119415,
      "eval_loss": 5.411562919616699,
      "eval_mean_token_accuracy": 0.19110058825501147,
      "eval_num_tokens": 8314232.0,
      "eval_runtime": 2.956,
      "eval_samples_per_second": 1321.726,
      "eval_steps_per_second": 165.427,
      "step": 4500
    },
    {
      "entropy": 5.1339836597442625,
      "epoch": 3.7698744769874475,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039556102887027423,
      "loss": 4.9124,
      "mean_token_accuracy": 0.1987938776612282,
      "num_tokens": 8323230.0,
      "step": 4505
    },
    {
      "entropy": 5.180046796798706,
      "epoch": 3.7740585774058575,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039528838425778053,
      "loss": 4.9411,
      "mean_token_accuracy": 0.20595014244318008,
      "num_tokens": 8333072.0,
      "step": 4510
    },
    {
      "entropy": 5.1110107421875,
      "epoch": 3.778242677824268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003950154921111313,
      "loss": 4.9007,
      "mean_token_accuracy": 0.20191781520843505,
      "num_tokens": 8342953.0,
      "step": 4515
    },
    {
      "entropy": 5.114612770080567,
      "epoch": 3.782426778242678,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039474235299189484,
      "loss": 4.8899,
      "mean_token_accuracy": 0.20523059368133545,
      "num_tokens": 8351729.0,
      "step": 4520
    },
    {
      "entropy": 5.192487192153931,
      "epoch": 3.786610878661088,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003944689674621474,
      "loss": 4.9468,
      "mean_token_accuracy": 0.20631079077720643,
      "num_tokens": 8360929.0,
      "step": 4525
    },
    {
      "entropy": 5.154665756225586,
      "epoch": 3.790794979079498,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039419533608447244,
      "loss": 4.9408,
      "mean_token_accuracy": 0.20518142879009246,
      "num_tokens": 8369506.0,
      "step": 4530
    },
    {
      "entropy": 5.153623580932617,
      "epoch": 3.794979079497908,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003939214594219595,
      "loss": 4.8527,
      "mean_token_accuracy": 0.20913135707378389,
      "num_tokens": 8378583.0,
      "step": 4535
    },
    {
      "entropy": 5.1313371658325195,
      "epoch": 3.799163179916318,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039364733803820263,
      "loss": 4.8364,
      "mean_token_accuracy": 0.21072372943162918,
      "num_tokens": 8388078.0,
      "step": 4540
    },
    {
      "entropy": 5.020518255233765,
      "epoch": 3.803347280334728,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003933729724972996,
      "loss": 4.8356,
      "mean_token_accuracy": 0.20935306698083878,
      "num_tokens": 8397086.0,
      "step": 4545
    },
    {
      "entropy": 5.069676208496094,
      "epoch": 3.8075313807531384,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039309836336385044,
      "loss": 4.829,
      "mean_token_accuracy": 0.22389160692691804,
      "num_tokens": 8405621.0,
      "step": 4550
    },
    {
      "entropy": 5.102335405349732,
      "epoch": 3.8117154811715483,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003928235112029568,
      "loss": 4.9006,
      "mean_token_accuracy": 0.2056551069021225,
      "num_tokens": 8415220.0,
      "step": 4555
    },
    {
      "entropy": 5.144475269317627,
      "epoch": 3.8158995815899583,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003925484165802201,
      "loss": 4.9142,
      "mean_token_accuracy": 0.20348594039678575,
      "num_tokens": 8424719.0,
      "step": 4560
    },
    {
      "entropy": 5.063912963867187,
      "epoch": 3.8200836820083683,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003922730800617408,
      "loss": 4.85,
      "mean_token_accuracy": 0.20849055647850037,
      "num_tokens": 8433844.0,
      "step": 4565
    },
    {
      "entropy": 5.007762098312378,
      "epoch": 3.8242677824267783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003919975022141174,
      "loss": 4.7826,
      "mean_token_accuracy": 0.21511289924383165,
      "num_tokens": 8442700.0,
      "step": 4570
    },
    {
      "entropy": 5.22186312675476,
      "epoch": 3.8284518828451883,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039172168360444463,
      "loss": 5.0249,
      "mean_token_accuracy": 0.19310083240270615,
      "num_tokens": 8452246.0,
      "step": 4575
    },
    {
      "entropy": 5.129202032089234,
      "epoch": 3.8326359832635983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039144562480031297,
      "loss": 4.8596,
      "mean_token_accuracy": 0.21053662449121474,
      "num_tokens": 8461977.0,
      "step": 4580
    },
    {
      "entropy": 5.026481199264526,
      "epoch": 3.8368200836820083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039116932636980685,
      "loss": 4.7689,
      "mean_token_accuracy": 0.20971563160419465,
      "num_tokens": 8470124.0,
      "step": 4585
    },
    {
      "entropy": 5.091988611221313,
      "epoch": 3.8410041841004183,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039089278888150417,
      "loss": 4.9136,
      "mean_token_accuracy": 0.19794375747442244,
      "num_tokens": 8479408.0,
      "step": 4590
    },
    {
      "entropy": 5.071839761734009,
      "epoch": 3.8451882845188283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003906160129044746,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2049538791179657,
      "num_tokens": 8488239.0,
      "step": 4595
    },
    {
      "entropy": 5.197607517242432,
      "epoch": 3.8493723849372383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003903389990082787,
      "loss": 4.9037,
      "mean_token_accuracy": 0.19678782671689987,
      "num_tokens": 8497690.0,
      "step": 4600
    },
    {
      "entropy": 5.194812679290772,
      "epoch": 3.8535564853556483,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003900617477629663,
      "loss": 4.9648,
      "mean_token_accuracy": 0.2004719316959381,
      "num_tokens": 8506974.0,
      "step": 4605
    },
    {
      "entropy": 5.1031715869903564,
      "epoch": 3.8577405857740583,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003897842597390761,
      "loss": 4.907,
      "mean_token_accuracy": 0.20890475660562516,
      "num_tokens": 8516066.0,
      "step": 4610
    },
    {
      "entropy": 5.0752955913543705,
      "epoch": 3.8619246861924688,
      "grad_norm": 1.25,
      "learning_rate": 0.0003895065355076338,
      "loss": 4.8654,
      "mean_token_accuracy": 0.20371763408184052,
      "num_tokens": 8524747.0,
      "step": 4615
    },
    {
      "entropy": 5.065091371536255,
      "epoch": 3.8661087866108788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003892285756401511,
      "loss": 4.8698,
      "mean_token_accuracy": 0.21580760478973388,
      "num_tokens": 8534274.0,
      "step": 4620
    },
    {
      "entropy": 5.0911235332489015,
      "epoch": 3.8702928870292888,
      "grad_norm": 1.125,
      "learning_rate": 0.0003889503807086248,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2059857204556465,
      "num_tokens": 8543453.0,
      "step": 4625
    },
    {
      "entropy": 5.114228630065918,
      "epoch": 3.8744769874476988,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038867195128553536,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2071972444653511,
      "num_tokens": 8553190.0,
      "step": 4630
    },
    {
      "entropy": 5.14755859375,
      "epoch": 3.8786610878661087,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003883932879438457,
      "loss": 4.8298,
      "mean_token_accuracy": 0.20431060791015626,
      "num_tokens": 8562814.0,
      "step": 4635
    },
    {
      "entropy": 5.028541326522827,
      "epoch": 3.8828451882845187,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003881143912570002,
      "loss": 4.842,
      "mean_token_accuracy": 0.20079325437545775,
      "num_tokens": 8572131.0,
      "step": 4640
    },
    {
      "entropy": 5.078951597213745,
      "epoch": 3.8870292887029287,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003878352617989235,
      "loss": 4.9203,
      "mean_token_accuracy": 0.1968255579471588,
      "num_tokens": 8582325.0,
      "step": 4645
    },
    {
      "entropy": 5.1008062839508055,
      "epoch": 3.891213389121339,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003875559001440191,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21458019614219664,
      "num_tokens": 8591704.0,
      "step": 4650
    },
    {
      "entropy": 5.137116861343384,
      "epoch": 3.895397489539749,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003872763068671682,
      "loss": 4.8893,
      "mean_token_accuracy": 0.20066926181316375,
      "num_tokens": 8601136.0,
      "step": 4655
    },
    {
      "entropy": 5.052433252334595,
      "epoch": 3.899581589958159,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003869964825437291,
      "loss": 4.8619,
      "mean_token_accuracy": 0.20738904625177385,
      "num_tokens": 8610491.0,
      "step": 4660
    },
    {
      "entropy": 5.124885940551758,
      "epoch": 3.903765690376569,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038671642774953507,
      "loss": 4.8947,
      "mean_token_accuracy": 0.20506638288497925,
      "num_tokens": 8619199.0,
      "step": 4665
    },
    {
      "entropy": 5.103211355209351,
      "epoch": 3.907949790794979,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003864361430608939,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20372416228055953,
      "num_tokens": 8629291.0,
      "step": 4670
    },
    {
      "entropy": 5.114092493057251,
      "epoch": 3.912133891213389,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003861556290545864,
      "loss": 4.9198,
      "mean_token_accuracy": 0.20413174778223037,
      "num_tokens": 8639212.0,
      "step": 4675
    },
    {
      "entropy": 5.108756685256958,
      "epoch": 3.916317991631799,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038587488630786534,
      "loss": 4.8623,
      "mean_token_accuracy": 0.20985756069421768,
      "num_tokens": 8648351.0,
      "step": 4680
    },
    {
      "entropy": 5.134943580627441,
      "epoch": 3.920502092050209,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038559391539845413,
      "loss": 4.9085,
      "mean_token_accuracy": 0.202670918405056,
      "num_tokens": 8657267.0,
      "step": 4685
    },
    {
      "entropy": 5.131587409973145,
      "epoch": 3.924686192468619,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038531271690454567,
      "loss": 5.0421,
      "mean_token_accuracy": 0.19898991137742997,
      "num_tokens": 8666982.0,
      "step": 4690
    },
    {
      "entropy": 5.200097894668579,
      "epoch": 3.928870292887029,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003850312914048015,
      "loss": 4.8429,
      "mean_token_accuracy": 0.2136537253856659,
      "num_tokens": 8676067.0,
      "step": 4695
    },
    {
      "entropy": 5.112923431396484,
      "epoch": 3.933054393305439,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003847496394783496,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21538768112659454,
      "num_tokens": 8685106.0,
      "step": 4700
    },
    {
      "entropy": 5.087848043441772,
      "epoch": 3.937238493723849,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003844677617047848,
      "loss": 4.9409,
      "mean_token_accuracy": 0.1996578350663185,
      "num_tokens": 8695203.0,
      "step": 4705
    },
    {
      "entropy": 5.042820072174072,
      "epoch": 3.941422594142259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000384185658664166,
      "loss": 4.7758,
      "mean_token_accuracy": 0.21243813037872314,
      "num_tokens": 8703700.0,
      "step": 4710
    },
    {
      "entropy": 5.169864368438721,
      "epoch": 3.9456066945606696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038390333093701596,
      "loss": 4.9168,
      "mean_token_accuracy": 0.2031609058380127,
      "num_tokens": 8713393.0,
      "step": 4715
    },
    {
      "entropy": 5.109617900848389,
      "epoch": 3.9497907949790796,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038362077910431973,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2068963512778282,
      "num_tokens": 8722263.0,
      "step": 4720
    },
    {
      "entropy": 5.067483520507812,
      "epoch": 3.9539748953974896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003833380037475236,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2049105256795883,
      "num_tokens": 8732190.0,
      "step": 4725
    },
    {
      "entropy": 5.185487985610962,
      "epoch": 3.9581589958158996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038305500544853375,
      "loss": 4.8458,
      "mean_token_accuracy": 0.21297864764928817,
      "num_tokens": 8740995.0,
      "step": 4730
    },
    {
      "entropy": 5.216033840179444,
      "epoch": 3.9623430962343096,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038277178478971527,
      "loss": 4.9665,
      "mean_token_accuracy": 0.1954173520207405,
      "num_tokens": 8750768.0,
      "step": 4735
    },
    {
      "entropy": 5.052295589447022,
      "epoch": 3.9665271966527196,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003824883423538905,
      "loss": 4.8918,
      "mean_token_accuracy": 0.2058481052517891,
      "num_tokens": 8760070.0,
      "step": 4740
    },
    {
      "entropy": 5.0752264022827145,
      "epoch": 3.9707112970711296,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038220467872433875,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2065647304058075,
      "num_tokens": 8769313.0,
      "step": 4745
    },
    {
      "entropy": 5.098513650894165,
      "epoch": 3.97489539748954,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003819207944847939,
      "loss": 4.8416,
      "mean_token_accuracy": 0.2117307260632515,
      "num_tokens": 8778328.0,
      "step": 4750
    },
    {
      "entropy": 5.116994905471802,
      "epoch": 3.97907949790795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038163669021944413,
      "loss": 4.868,
      "mean_token_accuracy": 0.21231727302074432,
      "num_tokens": 8787408.0,
      "step": 4755
    },
    {
      "entropy": 5.121054029464721,
      "epoch": 3.98326359832636,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003813523665129304,
      "loss": 4.8126,
      "mean_token_accuracy": 0.2095250591635704,
      "num_tokens": 8796781.0,
      "step": 4760
    },
    {
      "entropy": 5.096701717376709,
      "epoch": 3.98744769874477,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003810678239503452,
      "loss": 4.8814,
      "mean_token_accuracy": 0.20449732095003129,
      "num_tokens": 8806965.0,
      "step": 4765
    },
    {
      "entropy": 5.084413623809814,
      "epoch": 3.99163179916318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003807830631172312,
      "loss": 4.8921,
      "mean_token_accuracy": 0.2022484600543976,
      "num_tokens": 8817348.0,
      "step": 4770
    },
    {
      "entropy": 5.101122140884399,
      "epoch": 3.99581589958159,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038049808459958065,
      "loss": 4.9295,
      "mean_token_accuracy": 0.20652856975793837,
      "num_tokens": 8827304.0,
      "step": 4775
    },
    {
      "entropy": 5.091573047637939,
      "epoch": 4.0,
      "grad_norm": 1.125,
      "learning_rate": 0.00038021288898383343,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2077145606279373,
      "num_tokens": 8837288.0,
      "step": 4780
    },
    {
      "entropy": 5.127749586105347,
      "epoch": 4.00418410041841,
      "grad_norm": 1.125,
      "learning_rate": 0.0003799274768568761,
      "loss": 4.6296,
      "mean_token_accuracy": 0.22858476340770723,
      "num_tokens": 8846447.0,
      "step": 4785
    },
    {
      "entropy": 5.056386661529541,
      "epoch": 4.00836820083682,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037964184880604125,
      "loss": 4.5441,
      "mean_token_accuracy": 0.23339543342590333,
      "num_tokens": 8855990.0,
      "step": 4790
    },
    {
      "entropy": 5.014510631561279,
      "epoch": 4.01255230125523,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037935600541910523,
      "loss": 4.5512,
      "mean_token_accuracy": 0.22543267756700516,
      "num_tokens": 8864949.0,
      "step": 4795
    },
    {
      "entropy": 5.000003862380981,
      "epoch": 4.01673640167364,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003790699472842879,
      "loss": 4.4939,
      "mean_token_accuracy": 0.23468135446310043,
      "num_tokens": 8873501.0,
      "step": 4800
    },
    {
      "entropy": 5.025322151184082,
      "epoch": 4.02092050209205,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037878367499025086,
      "loss": 4.6044,
      "mean_token_accuracy": 0.22417306303977966,
      "num_tokens": 8882920.0,
      "step": 4805
    },
    {
      "entropy": 5.038575887680054,
      "epoch": 4.02510460251046,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003784971891260965,
      "loss": 4.5254,
      "mean_token_accuracy": 0.22962529212236404,
      "num_tokens": 8893617.0,
      "step": 4810
    },
    {
      "entropy": 5.070232295989991,
      "epoch": 4.02928870292887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037821049028136664,
      "loss": 4.5512,
      "mean_token_accuracy": 0.2259914293885231,
      "num_tokens": 8902558.0,
      "step": 4815
    },
    {
      "entropy": 5.077103900909424,
      "epoch": 4.03347280334728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003779235790460414,
      "loss": 4.573,
      "mean_token_accuracy": 0.2270744636654854,
      "num_tokens": 8912477.0,
      "step": 4820
    },
    {
      "entropy": 5.044887256622315,
      "epoch": 4.03765690376569,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003776364560105379,
      "loss": 4.5799,
      "mean_token_accuracy": 0.2272309586405754,
      "num_tokens": 8922313.0,
      "step": 4825
    },
    {
      "entropy": 4.988963603973389,
      "epoch": 4.0418410041841,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037734912176570935,
      "loss": 4.5391,
      "mean_token_accuracy": 0.2340757817029953,
      "num_tokens": 8932011.0,
      "step": 4830
    },
    {
      "entropy": 5.025544023513794,
      "epoch": 4.046025104602511,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037706157690284335,
      "loss": 4.5751,
      "mean_token_accuracy": 0.22412413209676743,
      "num_tokens": 8941022.0,
      "step": 4835
    },
    {
      "entropy": 5.0532307624816895,
      "epoch": 4.050209205020921,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037677382201366103,
      "loss": 4.5941,
      "mean_token_accuracy": 0.22880506813526152,
      "num_tokens": 8950825.0,
      "step": 4840
    },
    {
      "entropy": 5.049771642684936,
      "epoch": 4.054393305439331,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037648585769031555,
      "loss": 4.6131,
      "mean_token_accuracy": 0.21904713809490203,
      "num_tokens": 8961035.0,
      "step": 4845
    },
    {
      "entropy": 5.098345184326172,
      "epoch": 4.058577405857741,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003761976845253915,
      "loss": 4.6507,
      "mean_token_accuracy": 0.2194043755531311,
      "num_tokens": 8970535.0,
      "step": 4850
    },
    {
      "entropy": 4.985476875305176,
      "epoch": 4.062761506276151,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003759093031119025,
      "loss": 4.5822,
      "mean_token_accuracy": 0.22069828063249589,
      "num_tokens": 8979404.0,
      "step": 4855
    },
    {
      "entropy": 5.081757593154907,
      "epoch": 4.066945606694561,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037562071404329147,
      "loss": 4.5815,
      "mean_token_accuracy": 0.23617115765810012,
      "num_tokens": 8989164.0,
      "step": 4860
    },
    {
      "entropy": 5.132647085189819,
      "epoch": 4.071129707112971,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037533191791342817,
      "loss": 4.6384,
      "mean_token_accuracy": 0.21816497445106506,
      "num_tokens": 8998630.0,
      "step": 4865
    },
    {
      "entropy": 5.041156482696533,
      "epoch": 4.075313807531381,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037504291531660866,
      "loss": 4.5583,
      "mean_token_accuracy": 0.2322463497519493,
      "num_tokens": 9008235.0,
      "step": 4870
    },
    {
      "entropy": 5.041400671005249,
      "epoch": 4.079497907949791,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003747537068475538,
      "loss": 4.676,
      "mean_token_accuracy": 0.2232330098748207,
      "num_tokens": 9018175.0,
      "step": 4875
    },
    {
      "entropy": 5.117913818359375,
      "epoch": 4.083682008368201,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003744642931014082,
      "loss": 4.6922,
      "mean_token_accuracy": 0.21891210973262787,
      "num_tokens": 9028159.0,
      "step": 4880
    },
    {
      "entropy": 4.971291971206665,
      "epoch": 4.087866108786611,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003741746746737386,
      "loss": 4.6022,
      "mean_token_accuracy": 0.2239283248782158,
      "num_tokens": 9038263.0,
      "step": 4885
    },
    {
      "entropy": 5.066917133331299,
      "epoch": 4.092050209205021,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003738848521605333,
      "loss": 4.5974,
      "mean_token_accuracy": 0.22447667866945267,
      "num_tokens": 9048023.0,
      "step": 4890
    },
    {
      "entropy": 5.051163101196289,
      "epoch": 4.096234309623431,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003735948261582005,
      "loss": 4.6578,
      "mean_token_accuracy": 0.21464162617921828,
      "num_tokens": 9056988.0,
      "step": 4895
    },
    {
      "entropy": 5.029681873321533,
      "epoch": 4.100418410041841,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000373304597263567,
      "loss": 4.628,
      "mean_token_accuracy": 0.22242659330368042,
      "num_tokens": 9066595.0,
      "step": 4900
    },
    {
      "entropy": 5.066319227218628,
      "epoch": 4.104602510460251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037301416607387716,
      "loss": 4.6563,
      "mean_token_accuracy": 0.22113464921712875,
      "num_tokens": 9075034.0,
      "step": 4905
    },
    {
      "entropy": 4.993513917922973,
      "epoch": 4.108786610878661,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037272353318679185,
      "loss": 4.5948,
      "mean_token_accuracy": 0.22850108593702317,
      "num_tokens": 9084574.0,
      "step": 4910
    },
    {
      "entropy": 4.98863115310669,
      "epoch": 4.112970711297071,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003724326992003867,
      "loss": 4.5698,
      "mean_token_accuracy": 0.22540607750415803,
      "num_tokens": 9093591.0,
      "step": 4915
    },
    {
      "entropy": 5.04891004562378,
      "epoch": 4.117154811715481,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003721416647131512,
      "loss": 4.59,
      "mean_token_accuracy": 0.2303481638431549,
      "num_tokens": 9102439.0,
      "step": 4920
    },
    {
      "entropy": 4.949215269088745,
      "epoch": 4.121338912133891,
      "grad_norm": 1.25,
      "learning_rate": 0.0003718504303239876,
      "loss": 4.4807,
      "mean_token_accuracy": 0.23733771443367005,
      "num_tokens": 9111363.0,
      "step": 4925
    },
    {
      "entropy": 5.053757667541504,
      "epoch": 4.125523012552302,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003715589966322097,
      "loss": 4.6627,
      "mean_token_accuracy": 0.2233867108821869,
      "num_tokens": 9120417.0,
      "step": 4930
    },
    {
      "entropy": 5.102218723297119,
      "epoch": 4.129707112970712,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003712673642375409,
      "loss": 4.6351,
      "mean_token_accuracy": 0.21361656337976456,
      "num_tokens": 9130622.0,
      "step": 4935
    },
    {
      "entropy": 4.996326017379761,
      "epoch": 4.133891213389122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003709755337401138,
      "loss": 4.6061,
      "mean_token_accuracy": 0.2224092349410057,
      "num_tokens": 9139992.0,
      "step": 4940
    },
    {
      "entropy": 4.954313564300537,
      "epoch": 4.138075313807532,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037068350574046867,
      "loss": 4.5954,
      "mean_token_accuracy": 0.2263863727450371,
      "num_tokens": 9149883.0,
      "step": 4945
    },
    {
      "entropy": 5.00340518951416,
      "epoch": 4.142259414225942,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037039128083955227,
      "loss": 4.5567,
      "mean_token_accuracy": 0.23284687846899033,
      "num_tokens": 9159633.0,
      "step": 4950
    },
    {
      "entropy": 5.013108205795288,
      "epoch": 4.146443514644352,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003700988596387163,
      "loss": 4.5665,
      "mean_token_accuracy": 0.22669409215450287,
      "num_tokens": 9167780.0,
      "step": 4955
    },
    {
      "entropy": 5.017022705078125,
      "epoch": 4.150627615062762,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003698062427397168,
      "loss": 4.596,
      "mean_token_accuracy": 0.22676117569208146,
      "num_tokens": 9176808.0,
      "step": 4960
    },
    {
      "entropy": 5.0242228507995605,
      "epoch": 4.154811715481172,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036951343074471217,
      "loss": 4.6018,
      "mean_token_accuracy": 0.22619294226169587,
      "num_tokens": 9186097.0,
      "step": 4965
    },
    {
      "entropy": 5.026171493530273,
      "epoch": 4.158995815899582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003692204242562624,
      "loss": 4.6249,
      "mean_token_accuracy": 0.21765639185905455,
      "num_tokens": 9194428.0,
      "step": 4970
    },
    {
      "entropy": 5.0025266170501705,
      "epoch": 4.163179916317992,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003689272238773278,
      "loss": 4.6335,
      "mean_token_accuracy": 0.2212685227394104,
      "num_tokens": 9203156.0,
      "step": 4975
    },
    {
      "entropy": 5.00650429725647,
      "epoch": 4.167364016736402,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036863383021126756,
      "loss": 4.6364,
      "mean_token_accuracy": 0.2184078186750412,
      "num_tokens": 9212102.0,
      "step": 4980
    },
    {
      "entropy": 5.009310245513916,
      "epoch": 4.171548117154812,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036834024386183875,
      "loss": 4.6296,
      "mean_token_accuracy": 0.22262282073497772,
      "num_tokens": 9221507.0,
      "step": 4985
    },
    {
      "entropy": 5.03646674156189,
      "epoch": 4.175732217573222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003680464654331947,
      "loss": 4.5685,
      "mean_token_accuracy": 0.230113685131073,
      "num_tokens": 9230543.0,
      "step": 4990
    },
    {
      "entropy": 5.064976406097412,
      "epoch": 4.179916317991632,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003677524955298844,
      "loss": 4.6704,
      "mean_token_accuracy": 0.2220829501748085,
      "num_tokens": 9240092.0,
      "step": 4995
    },
    {
      "entropy": 5.0096848487854,
      "epoch": 4.184100418410042,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003674583347568505,
      "loss": 4.6337,
      "mean_token_accuracy": 0.2181374341249466,
      "num_tokens": 9250145.0,
      "step": 5000
    },
    {
      "epoch": 4.184100418410042,
      "eval_entropy": 4.8839097281418935,
      "eval_loss": 5.404016494750977,
      "eval_mean_token_accuracy": 0.19186635220952805,
      "eval_num_tokens": 9250145.0,
      "eval_runtime": 2.7597,
      "eval_samples_per_second": 1415.754,
      "eval_steps_per_second": 177.196,
      "step": 5000
    },
    {
      "entropy": 5.024817848205567,
      "epoch": 4.188284518828452,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036716398371942866,
      "loss": 4.5989,
      "mean_token_accuracy": 0.22544457763433456,
      "num_tokens": 9259643.0,
      "step": 5005
    },
    {
      "entropy": 5.045625019073486,
      "epoch": 4.192468619246862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003668694430233459,
      "loss": 4.6235,
      "mean_token_accuracy": 0.22249772399663925,
      "num_tokens": 9268471.0,
      "step": 5010
    },
    {
      "entropy": 5.036555576324463,
      "epoch": 4.196652719665272,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003665747132747198,
      "loss": 4.5753,
      "mean_token_accuracy": 0.22026388943195344,
      "num_tokens": 9276916.0,
      "step": 5015
    },
    {
      "entropy": 4.919180202484131,
      "epoch": 4.200836820083682,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036627979508005654,
      "loss": 4.5937,
      "mean_token_accuracy": 0.22548947632312774,
      "num_tokens": 9285867.0,
      "step": 5020
    },
    {
      "entropy": 5.073176574707031,
      "epoch": 4.205020920502092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003659846890462506,
      "loss": 4.7031,
      "mean_token_accuracy": 0.2204178974032402,
      "num_tokens": 9295044.0,
      "step": 5025
    },
    {
      "entropy": 5.1380829334259035,
      "epoch": 4.209205020920502,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003656893957805826,
      "loss": 4.7088,
      "mean_token_accuracy": 0.22122974395751954,
      "num_tokens": 9303744.0,
      "step": 5030
    },
    {
      "entropy": 5.020355606079102,
      "epoch": 4.2133891213389125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003653939158907186,
      "loss": 4.5924,
      "mean_token_accuracy": 0.22269472479820251,
      "num_tokens": 9312996.0,
      "step": 5035
    },
    {
      "entropy": 4.980009078979492,
      "epoch": 4.2175732217573225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003650982499847088,
      "loss": 4.6009,
      "mean_token_accuracy": 0.2312599629163742,
      "num_tokens": 9321912.0,
      "step": 5040
    },
    {
      "entropy": 5.018081760406494,
      "epoch": 4.2217573221757325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000364802398670986,
      "loss": 4.6189,
      "mean_token_accuracy": 0.22419243007898332,
      "num_tokens": 9331798.0,
      "step": 5045
    },
    {
      "entropy": 4.992781448364258,
      "epoch": 4.2259414225941425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036450636255836474,
      "loss": 4.6671,
      "mean_token_accuracy": 0.21717500537633896,
      "num_tokens": 9342064.0,
      "step": 5050
    },
    {
      "entropy": 5.078842115402222,
      "epoch": 4.2301255230125525,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036421014225603964,
      "loss": 4.5761,
      "mean_token_accuracy": 0.22376060783863067,
      "num_tokens": 9351494.0,
      "step": 5055
    },
    {
      "entropy": 4.899067115783692,
      "epoch": 4.2343096234309625,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036391373837358455,
      "loss": 4.5336,
      "mean_token_accuracy": 0.23086315095424653,
      "num_tokens": 9360683.0,
      "step": 5060
    },
    {
      "entropy": 4.962994050979614,
      "epoch": 4.2384937238493725,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036361715152095095,
      "loss": 4.621,
      "mean_token_accuracy": 0.22050356566905976,
      "num_tokens": 9370784.0,
      "step": 5065
    },
    {
      "entropy": 5.07359504699707,
      "epoch": 4.2426778242677825,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003633203823084669,
      "loss": 4.6723,
      "mean_token_accuracy": 0.22397183179855346,
      "num_tokens": 9380673.0,
      "step": 5070
    },
    {
      "entropy": 5.03209867477417,
      "epoch": 4.2468619246861925,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036302343134683566,
      "loss": 4.6556,
      "mean_token_accuracy": 0.22542175203561782,
      "num_tokens": 9390260.0,
      "step": 5075
    },
    {
      "entropy": 5.101150369644165,
      "epoch": 4.2510460251046025,
      "grad_norm": 1.25,
      "learning_rate": 0.0003627262992471348,
      "loss": 4.7013,
      "mean_token_accuracy": 0.2191369652748108,
      "num_tokens": 9400643.0,
      "step": 5080
    },
    {
      "entropy": 4.979807043075562,
      "epoch": 4.2552301255230125,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003624289866208142,
      "loss": 4.5681,
      "mean_token_accuracy": 0.22356537431478501,
      "num_tokens": 9408898.0,
      "step": 5085
    },
    {
      "entropy": 4.990454626083374,
      "epoch": 4.2594142259414225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036213149407969547,
      "loss": 4.6421,
      "mean_token_accuracy": 0.2180890902876854,
      "num_tokens": 9418024.0,
      "step": 5090
    },
    {
      "entropy": 5.1459509372711185,
      "epoch": 4.2635983263598325,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003618338222359705,
      "loss": 4.7253,
      "mean_token_accuracy": 0.211072239279747,
      "num_tokens": 9427548.0,
      "step": 5095
    },
    {
      "entropy": 5.031407833099365,
      "epoch": 4.2677824267782425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036153597170220005,
      "loss": 4.5436,
      "mean_token_accuracy": 0.22402570843696595,
      "num_tokens": 9436370.0,
      "step": 5100
    },
    {
      "entropy": 5.033519315719604,
      "epoch": 4.2719665271966525,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003612379430913126,
      "loss": 4.6129,
      "mean_token_accuracy": 0.21965315490961074,
      "num_tokens": 9445506.0,
      "step": 5105
    },
    {
      "entropy": 5.0202953815460205,
      "epoch": 4.2761506276150625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003609397370166031,
      "loss": 4.547,
      "mean_token_accuracy": 0.23238158524036406,
      "num_tokens": 9455288.0,
      "step": 5110
    },
    {
      "entropy": 4.938917922973633,
      "epoch": 4.2803347280334725,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003606413540917319,
      "loss": 4.6767,
      "mean_token_accuracy": 0.22185649573802949,
      "num_tokens": 9464366.0,
      "step": 5115
    },
    {
      "entropy": 4.981463098526001,
      "epoch": 4.2845188284518825,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003603427949307228,
      "loss": 4.5957,
      "mean_token_accuracy": 0.2296289473772049,
      "num_tokens": 9474205.0,
      "step": 5120
    },
    {
      "entropy": 4.9770636558532715,
      "epoch": 4.2887029288702925,
      "grad_norm": 1.125,
      "learning_rate": 0.00036004406014796274,
      "loss": 4.5906,
      "mean_token_accuracy": 0.22608524858951567,
      "num_tokens": 9483181.0,
      "step": 5125
    },
    {
      "entropy": 4.976647901535034,
      "epoch": 4.292887029288703,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003597451503581998,
      "loss": 4.5965,
      "mean_token_accuracy": 0.22682426422834395,
      "num_tokens": 9492394.0,
      "step": 5130
    },
    {
      "entropy": 5.072396516799927,
      "epoch": 4.297071129707113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035944606617654236,
      "loss": 4.6818,
      "mean_token_accuracy": 0.21135516166687013,
      "num_tokens": 9502184.0,
      "step": 5135
    },
    {
      "entropy": 4.970218467712402,
      "epoch": 4.301255230125523,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035914680821845746,
      "loss": 4.6521,
      "mean_token_accuracy": 0.22258177250623704,
      "num_tokens": 9511399.0,
      "step": 5140
    },
    {
      "entropy": 5.029709339141846,
      "epoch": 4.305439330543933,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035884737709976995,
      "loss": 4.6562,
      "mean_token_accuracy": 0.227021224796772,
      "num_tokens": 9520710.0,
      "step": 5145
    },
    {
      "entropy": 4.999623441696167,
      "epoch": 4.309623430962343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035854777343666093,
      "loss": 4.528,
      "mean_token_accuracy": 0.2403809756040573,
      "num_tokens": 9529992.0,
      "step": 5150
    },
    {
      "entropy": 4.950124216079712,
      "epoch": 4.313807531380753,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003582479978456665,
      "loss": 4.5511,
      "mean_token_accuracy": 0.23041269630193711,
      "num_tokens": 9538279.0,
      "step": 5155
    },
    {
      "entropy": 4.979808139801025,
      "epoch": 4.317991631799163,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003579480509436766,
      "loss": 4.5346,
      "mean_token_accuracy": 0.22957701981067657,
      "num_tokens": 9547425.0,
      "step": 5160
    },
    {
      "entropy": 4.970987844467163,
      "epoch": 4.322175732217573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035764793334793393,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2294863075017929,
      "num_tokens": 9556657.0,
      "step": 5165
    },
    {
      "entropy": 5.0658666610717775,
      "epoch": 4.326359832635983,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035734764567603216,
      "loss": 4.6611,
      "mean_token_accuracy": 0.22321824729442596,
      "num_tokens": 9565989.0,
      "step": 5170
    },
    {
      "entropy": 4.995517873764038,
      "epoch": 4.330543933054393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003570471885459149,
      "loss": 4.6182,
      "mean_token_accuracy": 0.22329700589179993,
      "num_tokens": 9575246.0,
      "step": 5175
    },
    {
      "entropy": 4.950409412384033,
      "epoch": 4.334728033472803,
      "grad_norm": 1.125,
      "learning_rate": 0.0003567465625758748,
      "loss": 4.5388,
      "mean_token_accuracy": 0.23305031210184096,
      "num_tokens": 9586185.0,
      "step": 5180
    },
    {
      "entropy": 4.993305301666259,
      "epoch": 4.338912133891213,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003564457683845516,
      "loss": 4.5495,
      "mean_token_accuracy": 0.23107281774282457,
      "num_tokens": 9595222.0,
      "step": 5185
    },
    {
      "entropy": 5.056625509262085,
      "epoch": 4.343096234309623,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003561448065909315,
      "loss": 4.6905,
      "mean_token_accuracy": 0.22466468065977097,
      "num_tokens": 9604866.0,
      "step": 5190
    },
    {
      "entropy": 4.967604446411133,
      "epoch": 4.347280334728033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003558436778143455,
      "loss": 4.5297,
      "mean_token_accuracy": 0.2330329164862633,
      "num_tokens": 9614522.0,
      "step": 5195
    },
    {
      "entropy": 5.036429595947266,
      "epoch": 4.351464435146443,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003555423826744682,
      "loss": 4.7085,
      "mean_token_accuracy": 0.21232665479183196,
      "num_tokens": 9623308.0,
      "step": 5200
    },
    {
      "entropy": 5.027107048034668,
      "epoch": 4.355648535564853,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035524092179131645,
      "loss": 4.693,
      "mean_token_accuracy": 0.21849237978458405,
      "num_tokens": 9632887.0,
      "step": 5205
    },
    {
      "entropy": 4.985956048965454,
      "epoch": 4.359832635983263,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003549392957852485,
      "loss": 4.6535,
      "mean_token_accuracy": 0.22133904844522476,
      "num_tokens": 9643218.0,
      "step": 5210
    },
    {
      "entropy": 5.01773476600647,
      "epoch": 4.364016736401673,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003546375052769619,
      "loss": 4.6584,
      "mean_token_accuracy": 0.2160348728299141,
      "num_tokens": 9653175.0,
      "step": 5215
    },
    {
      "entropy": 4.973669195175171,
      "epoch": 4.368200836820083,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003543355508874932,
      "loss": 4.561,
      "mean_token_accuracy": 0.22843843400478364,
      "num_tokens": 9662431.0,
      "step": 5220
    },
    {
      "entropy": 5.0396506786346436,
      "epoch": 4.372384937238493,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003540334332382159,
      "loss": 4.6049,
      "mean_token_accuracy": 0.22674299478530885,
      "num_tokens": 9671191.0,
      "step": 5225
    },
    {
      "entropy": 4.90244722366333,
      "epoch": 4.376569037656903,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003537311529508396,
      "loss": 4.5638,
      "mean_token_accuracy": 0.22337159663438796,
      "num_tokens": 9680637.0,
      "step": 5230
    },
    {
      "entropy": 4.976162052154541,
      "epoch": 4.380753138075314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035342871064740846,
      "loss": 4.5225,
      "mean_token_accuracy": 0.2310200423002243,
      "num_tokens": 9689325.0,
      "step": 5235
    },
    {
      "entropy": 4.999366378784179,
      "epoch": 4.384937238493724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035312610695030017,
      "loss": 4.6383,
      "mean_token_accuracy": 0.2239434614777565,
      "num_tokens": 9698335.0,
      "step": 5240
    },
    {
      "entropy": 4.9663115501403805,
      "epoch": 4.389121338912134,
      "grad_norm": 1.25,
      "learning_rate": 0.0003528233424822245,
      "loss": 4.593,
      "mean_token_accuracy": 0.22254677563905717,
      "num_tokens": 9706647.0,
      "step": 5245
    },
    {
      "entropy": 5.103724622726441,
      "epoch": 4.393305439330544,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000352520417866222,
      "loss": 4.6231,
      "mean_token_accuracy": 0.22940755188465117,
      "num_tokens": 9715941.0,
      "step": 5250
    },
    {
      "entropy": 4.981633567810059,
      "epoch": 4.397489539748954,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003522173337256628,
      "loss": 4.6518,
      "mean_token_accuracy": 0.21918769776821137,
      "num_tokens": 9724693.0,
      "step": 5255
    },
    {
      "entropy": 4.985882234573364,
      "epoch": 4.401673640167364,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003519140906842455,
      "loss": 4.6694,
      "mean_token_accuracy": 0.21712317317724228,
      "num_tokens": 9733314.0,
      "step": 5260
    },
    {
      "entropy": 5.123903465270996,
      "epoch": 4.405857740585774,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003516106893659953,
      "loss": 4.693,
      "mean_token_accuracy": 0.21850016415119172,
      "num_tokens": 9742844.0,
      "step": 5265
    },
    {
      "entropy": 4.985010385513306,
      "epoch": 4.410041841004184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003513071303952635,
      "loss": 4.5888,
      "mean_token_accuracy": 0.22708119601011276,
      "num_tokens": 9752164.0,
      "step": 5270
    },
    {
      "entropy": 4.900617074966431,
      "epoch": 4.414225941422594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003510034143967256,
      "loss": 4.5845,
      "mean_token_accuracy": 0.22353900372982025,
      "num_tokens": 9761483.0,
      "step": 5275
    },
    {
      "entropy": 5.002999877929687,
      "epoch": 4.418410041841004,
      "grad_norm": 1.25,
      "learning_rate": 0.00035069954199538027,
      "loss": 4.5528,
      "mean_token_accuracy": 0.22828604876995087,
      "num_tokens": 9769998.0,
      "step": 5280
    },
    {
      "entropy": 5.072404527664185,
      "epoch": 4.422594142259414,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035039551381654816,
      "loss": 4.628,
      "mean_token_accuracy": 0.22308734506368638,
      "num_tokens": 9778965.0,
      "step": 5285
    },
    {
      "entropy": 5.040173721313477,
      "epoch": 4.426778242677824,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035009133048587024,
      "loss": 4.6381,
      "mean_token_accuracy": 0.22403745204210282,
      "num_tokens": 9788543.0,
      "step": 5290
    },
    {
      "entropy": 4.962851285934448,
      "epoch": 4.430962343096234,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003497869926293069,
      "loss": 4.5158,
      "mean_token_accuracy": 0.22720393687486648,
      "num_tokens": 9797744.0,
      "step": 5295
    },
    {
      "entropy": 4.895706510543823,
      "epoch": 4.435146443514644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003494825008731367,
      "loss": 4.5796,
      "mean_token_accuracy": 0.22817811220884324,
      "num_tokens": 9807528.0,
      "step": 5300
    },
    {
      "entropy": 4.910212659835816,
      "epoch": 4.439330543933054,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034917785584395457,
      "loss": 4.6079,
      "mean_token_accuracy": 0.21853889375925065,
      "num_tokens": 9816078.0,
      "step": 5305
    },
    {
      "entropy": 5.032774496078491,
      "epoch": 4.443514644351464,
      "grad_norm": 1.25,
      "learning_rate": 0.000348873058168671,
      "loss": 4.5618,
      "mean_token_accuracy": 0.2248278096318245,
      "num_tokens": 9825199.0,
      "step": 5310
    },
    {
      "entropy": 5.07826509475708,
      "epoch": 4.447698744769874,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003485681084745106,
      "loss": 4.7152,
      "mean_token_accuracy": 0.2193989485502243,
      "num_tokens": 9835294.0,
      "step": 5315
    },
    {
      "entropy": 4.9364392280578615,
      "epoch": 4.451882845188284,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000348263007389011,
      "loss": 4.5814,
      "mean_token_accuracy": 0.2208610162138939,
      "num_tokens": 9844449.0,
      "step": 5320
    },
    {
      "entropy": 4.966502475738525,
      "epoch": 4.456066945606695,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034795775554002093,
      "loss": 4.6054,
      "mean_token_accuracy": 0.23577258586883545,
      "num_tokens": 9852930.0,
      "step": 5325
    },
    {
      "entropy": 4.98392243385315,
      "epoch": 4.460251046025105,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034765235355569974,
      "loss": 4.5416,
      "mean_token_accuracy": 0.23519645780324935,
      "num_tokens": 9862168.0,
      "step": 5330
    },
    {
      "entropy": 4.985634708404541,
      "epoch": 4.464435146443515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003473468020645155,
      "loss": 4.6888,
      "mean_token_accuracy": 0.22639351189136506,
      "num_tokens": 9871702.0,
      "step": 5335
    },
    {
      "entropy": 5.078781032562256,
      "epoch": 4.468619246861925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003470411016952442,
      "loss": 4.6932,
      "mean_token_accuracy": 0.2196798101067543,
      "num_tokens": 9880502.0,
      "step": 5340
    },
    {
      "entropy": 5.055650043487549,
      "epoch": 4.472803347280335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000346735253076968,
      "loss": 4.6389,
      "mean_token_accuracy": 0.22310524582862853,
      "num_tokens": 9889093.0,
      "step": 5345
    },
    {
      "entropy": 4.9169941425323485,
      "epoch": 4.476987447698745,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003464292568390743,
      "loss": 4.5337,
      "mean_token_accuracy": 0.23203957229852676,
      "num_tokens": 9897113.0,
      "step": 5350
    },
    {
      "entropy": 4.872857284545899,
      "epoch": 4.481171548117155,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003461231136112539,
      "loss": 4.5793,
      "mean_token_accuracy": 0.22614129185676574,
      "num_tokens": 9906478.0,
      "step": 5355
    },
    {
      "entropy": 5.02797155380249,
      "epoch": 4.485355648535565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034581682402350057,
      "loss": 4.632,
      "mean_token_accuracy": 0.21706513166427613,
      "num_tokens": 9916143.0,
      "step": 5360
    },
    {
      "entropy": 5.045882654190064,
      "epoch": 4.489539748953975,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034551038870610885,
      "loss": 4.6058,
      "mean_token_accuracy": 0.22375112771987915,
      "num_tokens": 9925675.0,
      "step": 5365
    },
    {
      "entropy": 4.957107448577881,
      "epoch": 4.493723849372385,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034520380828967346,
      "loss": 4.5428,
      "mean_token_accuracy": 0.23210841417312622,
      "num_tokens": 9934398.0,
      "step": 5370
    },
    {
      "entropy": 4.948310041427613,
      "epoch": 4.497907949790795,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003448970834050877,
      "loss": 4.5963,
      "mean_token_accuracy": 0.22628725022077562,
      "num_tokens": 9943066.0,
      "step": 5375
    },
    {
      "entropy": 4.866590547561645,
      "epoch": 4.502092050209205,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003445902146835419,
      "loss": 4.4997,
      "mean_token_accuracy": 0.23174125850200653,
      "num_tokens": 9951565.0,
      "step": 5380
    },
    {
      "entropy": 4.948875570297242,
      "epoch": 4.506276150627615,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003442832027565227,
      "loss": 4.6335,
      "mean_token_accuracy": 0.22018599212169648,
      "num_tokens": 9960997.0,
      "step": 5385
    },
    {
      "entropy": 5.06408462524414,
      "epoch": 4.510460251046025,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034397604825581115,
      "loss": 4.6398,
      "mean_token_accuracy": 0.22715661227703093,
      "num_tokens": 9970188.0,
      "step": 5390
    },
    {
      "entropy": 4.997549295425415,
      "epoch": 4.514644351464435,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003436687518134819,
      "loss": 4.63,
      "mean_token_accuracy": 0.22984076738357545,
      "num_tokens": 9978749.0,
      "step": 5395
    },
    {
      "entropy": 5.036268424987793,
      "epoch": 4.518828451882845,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003433613140619016,
      "loss": 4.6966,
      "mean_token_accuracy": 0.22537703067064285,
      "num_tokens": 9988107.0,
      "step": 5400
    },
    {
      "entropy": 4.89153733253479,
      "epoch": 4.523012552301255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003430537356337277,
      "loss": 4.5219,
      "mean_token_accuracy": 0.22330804765224457,
      "num_tokens": 9996127.0,
      "step": 5405
    },
    {
      "entropy": 4.989614391326905,
      "epoch": 4.527196652719665,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003427460171619073,
      "loss": 4.6394,
      "mean_token_accuracy": 0.219285224378109,
      "num_tokens": 10005482.0,
      "step": 5410
    },
    {
      "entropy": 5.041315746307373,
      "epoch": 4.531380753138075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034243815927967536,
      "loss": 4.6911,
      "mean_token_accuracy": 0.21671502888202668,
      "num_tokens": 10014765.0,
      "step": 5415
    },
    {
      "entropy": 5.0097565174102785,
      "epoch": 4.535564853556485,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034213016262055396,
      "loss": 4.6227,
      "mean_token_accuracy": 0.21503682881593705,
      "num_tokens": 10022897.0,
      "step": 5420
    },
    {
      "entropy": 5.044774103164673,
      "epoch": 4.539748953974895,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003418220278183507,
      "loss": 4.6959,
      "mean_token_accuracy": 0.2108847975730896,
      "num_tokens": 10032281.0,
      "step": 5425
    },
    {
      "entropy": 4.891918468475342,
      "epoch": 4.543933054393305,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034151375550715753,
      "loss": 4.5326,
      "mean_token_accuracy": 0.23642961382865907,
      "num_tokens": 10041194.0,
      "step": 5430
    },
    {
      "entropy": 4.897344636917114,
      "epoch": 4.548117154811716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034120534632134917,
      "loss": 4.6057,
      "mean_token_accuracy": 0.21950752884149552,
      "num_tokens": 10050062.0,
      "step": 5435
    },
    {
      "entropy": 5.0067027568817135,
      "epoch": 4.552301255230126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034089680089558226,
      "loss": 4.6858,
      "mean_token_accuracy": 0.22794269323348998,
      "num_tokens": 10059747.0,
      "step": 5440
    },
    {
      "entropy": 5.0438157081604,
      "epoch": 4.556485355648536,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034058811986479367,
      "loss": 4.711,
      "mean_token_accuracy": 0.21663242131471633,
      "num_tokens": 10069351.0,
      "step": 5445
    },
    {
      "entropy": 5.003661346435547,
      "epoch": 4.560669456066946,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034027930386419935,
      "loss": 4.6373,
      "mean_token_accuracy": 0.21767364144325257,
      "num_tokens": 10078727.0,
      "step": 5450
    },
    {
      "entropy": 4.928648948669434,
      "epoch": 4.564853556485356,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033997035352929285,
      "loss": 4.5883,
      "mean_token_accuracy": 0.22172482460737228,
      "num_tokens": 10088151.0,
      "step": 5455
    },
    {
      "entropy": 4.917711782455444,
      "epoch": 4.569037656903766,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003396612694958444,
      "loss": 4.5452,
      "mean_token_accuracy": 0.2294139802455902,
      "num_tokens": 10097311.0,
      "step": 5460
    },
    {
      "entropy": 4.9842822551727295,
      "epoch": 4.573221757322176,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033935205239989927,
      "loss": 4.6796,
      "mean_token_accuracy": 0.2146899789571762,
      "num_tokens": 10107474.0,
      "step": 5465
    },
    {
      "entropy": 5.0064507007598875,
      "epoch": 4.577405857740586,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033904270287777645,
      "loss": 4.6116,
      "mean_token_accuracy": 0.23307846784591674,
      "num_tokens": 10116305.0,
      "step": 5470
    },
    {
      "entropy": 5.010985851287842,
      "epoch": 4.581589958158996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003387332215660675,
      "loss": 4.6489,
      "mean_token_accuracy": 0.22137546241283418,
      "num_tokens": 10125947.0,
      "step": 5475
    },
    {
      "entropy": 4.944061613082885,
      "epoch": 4.585774058577406,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033842360910163536,
      "loss": 4.6351,
      "mean_token_accuracy": 0.22578810006380082,
      "num_tokens": 10134510.0,
      "step": 5480
    },
    {
      "entropy": 5.044047784805298,
      "epoch": 4.589958158995816,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003381138661216125,
      "loss": 4.6945,
      "mean_token_accuracy": 0.217996309697628,
      "num_tokens": 10143429.0,
      "step": 5485
    },
    {
      "entropy": 5.028554058074951,
      "epoch": 4.594142259414226,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033780399326340026,
      "loss": 4.6737,
      "mean_token_accuracy": 0.220823672413826,
      "num_tokens": 10153405.0,
      "step": 5490
    },
    {
      "entropy": 5.046740102767944,
      "epoch": 4.598326359832636,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033749399116466723,
      "loss": 4.6279,
      "mean_token_accuracy": 0.23109050989151,
      "num_tokens": 10162334.0,
      "step": 5495
    },
    {
      "entropy": 4.98068675994873,
      "epoch": 4.602510460251046,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003371838604633479,
      "loss": 4.6394,
      "mean_token_accuracy": 0.22259229719638823,
      "num_tokens": 10171604.0,
      "step": 5500
    },
    {
      "epoch": 4.602510460251046,
      "eval_entropy": 4.798924718166422,
      "eval_loss": 5.38128662109375,
      "eval_mean_token_accuracy": 0.19562271357740604,
      "eval_num_tokens": 10171604.0,
      "eval_runtime": 2.7933,
      "eval_samples_per_second": 1398.7,
      "eval_steps_per_second": 175.061,
      "step": 5500
    },
    {
      "entropy": 4.949925422668457,
      "epoch": 4.606694560669456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033687360179764125,
      "loss": 4.5926,
      "mean_token_accuracy": 0.22750910222530366,
      "num_tokens": 10180659.0,
      "step": 5505
    },
    {
      "entropy": 5.008641195297241,
      "epoch": 4.610878661087866,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003365632158060098,
      "loss": 4.6505,
      "mean_token_accuracy": 0.22003360837697983,
      "num_tokens": 10190226.0,
      "step": 5510
    },
    {
      "entropy": 4.9972716808319095,
      "epoch": 4.615062761506276,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033625270312717805,
      "loss": 4.5518,
      "mean_token_accuracy": 0.22763495445251464,
      "num_tokens": 10198806.0,
      "step": 5515
    },
    {
      "entropy": 4.940847873687744,
      "epoch": 4.619246861924686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003359420644001311,
      "loss": 4.5611,
      "mean_token_accuracy": 0.23088591694831848,
      "num_tokens": 10207927.0,
      "step": 5520
    },
    {
      "entropy": 4.880846691131592,
      "epoch": 4.623430962343097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033563130026411357,
      "loss": 4.491,
      "mean_token_accuracy": 0.23794422745704652,
      "num_tokens": 10216241.0,
      "step": 5525
    },
    {
      "entropy": 4.956955623626709,
      "epoch": 4.627615062761507,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033532041135862816,
      "loss": 4.6959,
      "mean_token_accuracy": 0.2141469269990921,
      "num_tokens": 10225241.0,
      "step": 5530
    },
    {
      "entropy": 4.984134721755981,
      "epoch": 4.631799163179917,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033500939832343407,
      "loss": 4.6577,
      "mean_token_accuracy": 0.21747250556945802,
      "num_tokens": 10233968.0,
      "step": 5535
    },
    {
      "entropy": 5.1195353984832765,
      "epoch": 4.635983263598327,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003346982617985463,
      "loss": 4.7328,
      "mean_token_accuracy": 0.21601124852895737,
      "num_tokens": 10243589.0,
      "step": 5540
    },
    {
      "entropy": 4.9155150890350345,
      "epoch": 4.640167364016737,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003343870024242337,
      "loss": 4.5633,
      "mean_token_accuracy": 0.22466115057468414,
      "num_tokens": 10252416.0,
      "step": 5545
    },
    {
      "entropy": 4.991417646408081,
      "epoch": 4.644351464435147,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033407562084101795,
      "loss": 4.6811,
      "mean_token_accuracy": 0.21211913824081421,
      "num_tokens": 10262213.0,
      "step": 5550
    },
    {
      "entropy": 5.065744113922119,
      "epoch": 4.648535564853557,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033376411768967247,
      "loss": 4.748,
      "mean_token_accuracy": 0.21660719215869903,
      "num_tokens": 10272774.0,
      "step": 5555
    },
    {
      "entropy": 5.003204774856568,
      "epoch": 4.652719665271967,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033345249361122056,
      "loss": 4.5737,
      "mean_token_accuracy": 0.2304500311613083,
      "num_tokens": 10281286.0,
      "step": 5560
    },
    {
      "entropy": 4.920540428161621,
      "epoch": 4.656903765690377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033314074924693446,
      "loss": 4.5992,
      "mean_token_accuracy": 0.22165121883153915,
      "num_tokens": 10290101.0,
      "step": 5565
    },
    {
      "entropy": 4.925906753540039,
      "epoch": 4.661087866108787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033282888523833394,
      "loss": 4.6129,
      "mean_token_accuracy": 0.22973967343568802,
      "num_tokens": 10299580.0,
      "step": 5570
    },
    {
      "entropy": 4.9791007995605465,
      "epoch": 4.665271966527197,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033251690222718504,
      "loss": 4.6203,
      "mean_token_accuracy": 0.23005986958742142,
      "num_tokens": 10307860.0,
      "step": 5575
    },
    {
      "entropy": 5.0008961200714115,
      "epoch": 4.669456066945607,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003322048008554986,
      "loss": 4.6272,
      "mean_token_accuracy": 0.2281336233019829,
      "num_tokens": 10316546.0,
      "step": 5580
    },
    {
      "entropy": 5.057764196395874,
      "epoch": 4.673640167364017,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003318925817655291,
      "loss": 4.7176,
      "mean_token_accuracy": 0.224642813205719,
      "num_tokens": 10326012.0,
      "step": 5585
    },
    {
      "entropy": 5.068313312530518,
      "epoch": 4.677824267782427,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003315802455997732,
      "loss": 4.6591,
      "mean_token_accuracy": 0.2191016674041748,
      "num_tokens": 10334771.0,
      "step": 5590
    },
    {
      "entropy": 5.022720289230347,
      "epoch": 4.682008368200837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033126779300096855,
      "loss": 4.6481,
      "mean_token_accuracy": 0.2229116141796112,
      "num_tokens": 10343416.0,
      "step": 5595
    },
    {
      "entropy": 5.044781160354614,
      "epoch": 4.686192468619247,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003309552246120924,
      "loss": 4.7424,
      "mean_token_accuracy": 0.21645105630159378,
      "num_tokens": 10353539.0,
      "step": 5600
    },
    {
      "entropy": 4.976648902893066,
      "epoch": 4.690376569037657,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003306425410763601,
      "loss": 4.5993,
      "mean_token_accuracy": 0.22718028575181962,
      "num_tokens": 10362020.0,
      "step": 5605
    },
    {
      "entropy": 4.934230995178223,
      "epoch": 4.694560669456067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003303297430372241,
      "loss": 4.5128,
      "mean_token_accuracy": 0.23098493963479996,
      "num_tokens": 10371277.0,
      "step": 5610
    },
    {
      "entropy": 4.920706462860108,
      "epoch": 4.698744769874477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003300168311383727,
      "loss": 4.6243,
      "mean_token_accuracy": 0.22681334167718886,
      "num_tokens": 10380680.0,
      "step": 5615
    },
    {
      "entropy": 4.959516620635986,
      "epoch": 4.702928870292887,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032970380602372805,
      "loss": 4.5551,
      "mean_token_accuracy": 0.23426176905632018,
      "num_tokens": 10389576.0,
      "step": 5620
    },
    {
      "entropy": 4.961257696151733,
      "epoch": 4.707112970711297,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003293906683374456,
      "loss": 4.6616,
      "mean_token_accuracy": 0.22408125102519988,
      "num_tokens": 10399182.0,
      "step": 5625
    },
    {
      "entropy": 5.092632579803467,
      "epoch": 4.711297071129707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003290774187239124,
      "loss": 4.7824,
      "mean_token_accuracy": 0.21829198002815248,
      "num_tokens": 10409595.0,
      "step": 5630
    },
    {
      "entropy": 5.084794235229492,
      "epoch": 4.7154811715481175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032876405782774574,
      "loss": 4.7341,
      "mean_token_accuracy": 0.2125866234302521,
      "num_tokens": 10419057.0,
      "step": 5635
    },
    {
      "entropy": 4.929048109054565,
      "epoch": 4.7196652719665275,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003284505862937919,
      "loss": 4.5254,
      "mean_token_accuracy": 0.23203834891319275,
      "num_tokens": 10428090.0,
      "step": 5640
    },
    {
      "entropy": 5.0079553604125975,
      "epoch": 4.7238493723849375,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000328137004767125,
      "loss": 4.6705,
      "mean_token_accuracy": 0.2221292719244957,
      "num_tokens": 10438088.0,
      "step": 5645
    },
    {
      "entropy": 4.952316427230835,
      "epoch": 4.7280334728033475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003278233138930454,
      "loss": 4.6788,
      "mean_token_accuracy": 0.21837462931871415,
      "num_tokens": 10447934.0,
      "step": 5650
    },
    {
      "entropy": 4.980340957641602,
      "epoch": 4.7322175732217575,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003275095143170784,
      "loss": 4.5284,
      "mean_token_accuracy": 0.2339065134525299,
      "num_tokens": 10456088.0,
      "step": 5655
    },
    {
      "entropy": 4.935347175598144,
      "epoch": 4.7364016736401675,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032719560668497316,
      "loss": 4.5547,
      "mean_token_accuracy": 0.23369274735450746,
      "num_tokens": 10465096.0,
      "step": 5660
    },
    {
      "entropy": 4.874028062820434,
      "epoch": 4.7405857740585775,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032688159164270114,
      "loss": 4.6245,
      "mean_token_accuracy": 0.22505829483270645,
      "num_tokens": 10473516.0,
      "step": 5665
    },
    {
      "entropy": 4.996293830871582,
      "epoch": 4.7447698744769875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032656746983645473,
      "loss": 4.6037,
      "mean_token_accuracy": 0.23210916966199874,
      "num_tokens": 10482651.0,
      "step": 5670
    },
    {
      "entropy": 5.04516568183899,
      "epoch": 4.7489539748953975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003262532419126461,
      "loss": 4.6444,
      "mean_token_accuracy": 0.2159220203757286,
      "num_tokens": 10491831.0,
      "step": 5675
    },
    {
      "entropy": 4.935607957839966,
      "epoch": 4.7531380753138075,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032593890851790593,
      "loss": 4.6451,
      "mean_token_accuracy": 0.22744875103235246,
      "num_tokens": 10502070.0,
      "step": 5680
    },
    {
      "entropy": 4.974724531173706,
      "epoch": 4.7573221757322175,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003256244702990817,
      "loss": 4.621,
      "mean_token_accuracy": 0.21882622241973876,
      "num_tokens": 10510638.0,
      "step": 5685
    },
    {
      "entropy": 4.934746313095093,
      "epoch": 4.7615062761506275,
      "grad_norm": 1.125,
      "learning_rate": 0.00032530992790323664,
      "loss": 4.5543,
      "mean_token_accuracy": 0.2280549019575119,
      "num_tokens": 10519305.0,
      "step": 5690
    },
    {
      "entropy": 4.979501581192016,
      "epoch": 4.7656903765690375,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032499528197764855,
      "loss": 4.604,
      "mean_token_accuracy": 0.22649753987789153,
      "num_tokens": 10528844.0,
      "step": 5695
    },
    {
      "entropy": 5.01939115524292,
      "epoch": 4.7698744769874475,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032468053316980815,
      "loss": 4.7343,
      "mean_token_accuracy": 0.2168871834874153,
      "num_tokens": 10538260.0,
      "step": 5700
    },
    {
      "entropy": 5.004323053359985,
      "epoch": 4.7740585774058575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003243656821274178,
      "loss": 4.656,
      "mean_token_accuracy": 0.22183047384023666,
      "num_tokens": 10547733.0,
      "step": 5705
    },
    {
      "entropy": 4.952843952178955,
      "epoch": 4.7782426778242675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003240507294983905,
      "loss": 4.5979,
      "mean_token_accuracy": 0.23014811426401138,
      "num_tokens": 10558337.0,
      "step": 5710
    },
    {
      "entropy": 4.948026371002197,
      "epoch": 4.7824267782426775,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032373567593084804,
      "loss": 4.6074,
      "mean_token_accuracy": 0.2147175446152687,
      "num_tokens": 10568374.0,
      "step": 5715
    },
    {
      "entropy": 4.9570776462554935,
      "epoch": 4.786610878661088,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032342052207311986,
      "loss": 4.6997,
      "mean_token_accuracy": 0.22087032943964005,
      "num_tokens": 10578448.0,
      "step": 5720
    },
    {
      "entropy": 5.036951065063477,
      "epoch": 4.790794979079498,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032310526857374204,
      "loss": 4.7179,
      "mean_token_accuracy": 0.2130563363432884,
      "num_tokens": 10586803.0,
      "step": 5725
    },
    {
      "entropy": 5.075342178344727,
      "epoch": 4.794979079497908,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003227899160814556,
      "loss": 4.6992,
      "mean_token_accuracy": 0.22343989908695222,
      "num_tokens": 10596596.0,
      "step": 5730
    },
    {
      "entropy": 5.001523494720459,
      "epoch": 4.799163179916318,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003224744652452054,
      "loss": 4.5726,
      "mean_token_accuracy": 0.23056758493185042,
      "num_tokens": 10605462.0,
      "step": 5735
    },
    {
      "entropy": 5.074092531204224,
      "epoch": 4.803347280334728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003221589167141382,
      "loss": 4.7241,
      "mean_token_accuracy": 0.2151061177253723,
      "num_tokens": 10614908.0,
      "step": 5740
    },
    {
      "entropy": 5.038457632064819,
      "epoch": 4.807531380753138,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003218432711376025,
      "loss": 4.6911,
      "mean_token_accuracy": 0.22176578938961028,
      "num_tokens": 10624389.0,
      "step": 5745
    },
    {
      "entropy": 4.992576599121094,
      "epoch": 4.811715481171548,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032152752916514605,
      "loss": 4.6456,
      "mean_token_accuracy": 0.22301490604877472,
      "num_tokens": 10633228.0,
      "step": 5750
    },
    {
      "entropy": 4.9775971412658695,
      "epoch": 4.815899581589958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032121169144651496,
      "loss": 4.6832,
      "mean_token_accuracy": 0.22345893681049347,
      "num_tokens": 10641996.0,
      "step": 5755
    },
    {
      "entropy": 4.972227668762207,
      "epoch": 4.820083682008368,
      "grad_norm": 1.25,
      "learning_rate": 0.0003208957586316525,
      "loss": 4.6007,
      "mean_token_accuracy": 0.2273903489112854,
      "num_tokens": 10651724.0,
      "step": 5760
    },
    {
      "entropy": 5.0259765625,
      "epoch": 4.824267782426778,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003205797313706977,
      "loss": 4.6696,
      "mean_token_accuracy": 0.22497095316648483,
      "num_tokens": 10661150.0,
      "step": 5765
    },
    {
      "entropy": 4.965875244140625,
      "epoch": 4.828451882845188,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032026361031398386,
      "loss": 4.6791,
      "mean_token_accuracy": 0.2171620786190033,
      "num_tokens": 10669890.0,
      "step": 5770
    },
    {
      "entropy": 5.044986629486084,
      "epoch": 4.832635983263598,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031994739611203716,
      "loss": 4.5876,
      "mean_token_accuracy": 0.21953882575035094,
      "num_tokens": 10678939.0,
      "step": 5775
    },
    {
      "entropy": 4.960333061218262,
      "epoch": 4.836820083682008,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031963108941557576,
      "loss": 4.5632,
      "mean_token_accuracy": 0.22089180648326873,
      "num_tokens": 10687465.0,
      "step": 5780
    },
    {
      "entropy": 4.97818603515625,
      "epoch": 4.841004184100418,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003193146908755078,
      "loss": 4.6682,
      "mean_token_accuracy": 0.21882702261209488,
      "num_tokens": 10696930.0,
      "step": 5785
    },
    {
      "entropy": 4.967428350448609,
      "epoch": 4.845188284518828,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031899820114293073,
      "loss": 4.5869,
      "mean_token_accuracy": 0.23010560274124145,
      "num_tokens": 10706820.0,
      "step": 5790
    },
    {
      "entropy": 4.965236711502075,
      "epoch": 4.849372384937238,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031868162086912946,
      "loss": 4.5788,
      "mean_token_accuracy": 0.23480864614248276,
      "num_tokens": 10715528.0,
      "step": 5795
    },
    {
      "entropy": 4.914876461029053,
      "epoch": 4.853556485355648,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003183649507055754,
      "loss": 4.577,
      "mean_token_accuracy": 0.22862247377634048,
      "num_tokens": 10724807.0,
      "step": 5800
    },
    {
      "entropy": 5.027656030654907,
      "epoch": 4.857740585774058,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031804819130392477,
      "loss": 4.6606,
      "mean_token_accuracy": 0.21356358826160432,
      "num_tokens": 10733754.0,
      "step": 5805
    },
    {
      "entropy": 4.967838907241822,
      "epoch": 4.861924686192468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003177313433160175,
      "loss": 4.6662,
      "mean_token_accuracy": 0.22005672007799149,
      "num_tokens": 10742256.0,
      "step": 5810
    },
    {
      "entropy": 4.983591079711914,
      "epoch": 4.866108786610878,
      "grad_norm": 1.25,
      "learning_rate": 0.0003174144073938757,
      "loss": 4.588,
      "mean_token_accuracy": 0.23729257434606552,
      "num_tokens": 10751025.0,
      "step": 5815
    },
    {
      "entropy": 4.9942419052124025,
      "epoch": 4.870292887029288,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031709738418970275,
      "loss": 4.5602,
      "mean_token_accuracy": 0.23407905399799347,
      "num_tokens": 10759970.0,
      "step": 5820
    },
    {
      "entropy": 4.964095640182495,
      "epoch": 4.874476987447698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003167802743558813,
      "loss": 4.61,
      "mean_token_accuracy": 0.2255547896027565,
      "num_tokens": 10768531.0,
      "step": 5825
    },
    {
      "entropy": 5.016986751556397,
      "epoch": 4.878661087866108,
      "grad_norm": 1.125,
      "learning_rate": 0.0003164630785449724,
      "loss": 4.6849,
      "mean_token_accuracy": 0.22082346230745314,
      "num_tokens": 10778802.0,
      "step": 5830
    },
    {
      "entropy": 5.033987522125244,
      "epoch": 4.882845188284519,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031614579740971415,
      "loss": 4.6175,
      "mean_token_accuracy": 0.22324100732803345,
      "num_tokens": 10788121.0,
      "step": 5835
    },
    {
      "entropy": 5.053671789169312,
      "epoch": 4.887029288702929,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031582843160302,
      "loss": 4.7184,
      "mean_token_accuracy": 0.22378790825605394,
      "num_tokens": 10797652.0,
      "step": 5840
    },
    {
      "entropy": 4.966518449783325,
      "epoch": 4.891213389121339,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031551098177797784,
      "loss": 4.6416,
      "mean_token_accuracy": 0.2134111538529396,
      "num_tokens": 10807564.0,
      "step": 5845
    },
    {
      "entropy": 5.000678300857544,
      "epoch": 4.895397489539749,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003151934485878482,
      "loss": 4.6069,
      "mean_token_accuracy": 0.2213506892323494,
      "num_tokens": 10816666.0,
      "step": 5850
    },
    {
      "entropy": 4.945336103439331,
      "epoch": 4.899581589958159,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031487583268606354,
      "loss": 4.6108,
      "mean_token_accuracy": 0.22933099269866944,
      "num_tokens": 10825639.0,
      "step": 5855
    },
    {
      "entropy": 4.922334861755371,
      "epoch": 4.903765690376569,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003145581347262262,
      "loss": 4.6609,
      "mean_token_accuracy": 0.21985768526792526,
      "num_tokens": 10834681.0,
      "step": 5860
    },
    {
      "entropy": 4.978070783615112,
      "epoch": 4.907949790794979,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031424035536210755,
      "loss": 4.6568,
      "mean_token_accuracy": 0.22321512997150422,
      "num_tokens": 10844465.0,
      "step": 5865
    },
    {
      "entropy": 5.006863403320312,
      "epoch": 4.912133891213389,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031392249524764645,
      "loss": 4.6767,
      "mean_token_accuracy": 0.2198706790804863,
      "num_tokens": 10853327.0,
      "step": 5870
    },
    {
      "entropy": 4.977741670608521,
      "epoch": 4.916317991631799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031360455503694787,
      "loss": 4.5866,
      "mean_token_accuracy": 0.22820475846529006,
      "num_tokens": 10861865.0,
      "step": 5875
    },
    {
      "entropy": 5.016619491577148,
      "epoch": 4.920502092050209,
      "grad_norm": 1.25,
      "learning_rate": 0.0003132865353842816,
      "loss": 4.659,
      "mean_token_accuracy": 0.218600495159626,
      "num_tokens": 10870794.0,
      "step": 5880
    },
    {
      "entropy": 5.00802206993103,
      "epoch": 4.924686192468619,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031296843694408107,
      "loss": 4.5799,
      "mean_token_accuracy": 0.22618063986301423,
      "num_tokens": 10879626.0,
      "step": 5885
    },
    {
      "entropy": 4.9072620391845705,
      "epoch": 4.928870292887029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003126502603709417,
      "loss": 4.5278,
      "mean_token_accuracy": 0.2303570032119751,
      "num_tokens": 10888359.0,
      "step": 5890
    },
    {
      "entropy": 4.9940667152404785,
      "epoch": 4.933054393305439,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003123320063196196,
      "loss": 4.6017,
      "mean_token_accuracy": 0.2247793644666672,
      "num_tokens": 10897174.0,
      "step": 5895
    },
    {
      "entropy": 4.988154697418213,
      "epoch": 4.937238493723849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003120136754450306,
      "loss": 4.6611,
      "mean_token_accuracy": 0.21055931001901626,
      "num_tokens": 10906660.0,
      "step": 5900
    },
    {
      "entropy": 4.97977499961853,
      "epoch": 4.941422594142259,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031169526840224836,
      "loss": 4.6836,
      "mean_token_accuracy": 0.22239537239074708,
      "num_tokens": 10915990.0,
      "step": 5905
    },
    {
      "entropy": 5.096631050109863,
      "epoch": 4.945606694560669,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003113767858465034,
      "loss": 4.8115,
      "mean_token_accuracy": 0.19746783524751663,
      "num_tokens": 10926355.0,
      "step": 5910
    },
    {
      "entropy": 4.941648721694946,
      "epoch": 4.949790794979079,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031105822843318173,
      "loss": 4.6445,
      "mean_token_accuracy": 0.22564318776130676,
      "num_tokens": 10935854.0,
      "step": 5915
    },
    {
      "entropy": 4.9659590244293215,
      "epoch": 4.95397489539749,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003107395968178232,
      "loss": 4.5958,
      "mean_token_accuracy": 0.231874543428421,
      "num_tokens": 10944903.0,
      "step": 5920
    },
    {
      "entropy": 4.990147829055786,
      "epoch": 4.9581589958159,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003104208916561203,
      "loss": 4.619,
      "mean_token_accuracy": 0.22236147075891494,
      "num_tokens": 10954009.0,
      "step": 5925
    },
    {
      "entropy": 5.0643370151519775,
      "epoch": 4.96234309623431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031010211360391727,
      "loss": 4.7484,
      "mean_token_accuracy": 0.20919602364301682,
      "num_tokens": 10963174.0,
      "step": 5930
    },
    {
      "entropy": 5.0831780433654785,
      "epoch": 4.96652719665272,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000309783263317208,
      "loss": 4.6819,
      "mean_token_accuracy": 0.21789833903312683,
      "num_tokens": 10972822.0,
      "step": 5935
    },
    {
      "entropy": 4.982867240905762,
      "epoch": 4.97071129707113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003094643414521349,
      "loss": 4.5661,
      "mean_token_accuracy": 0.22501951158046724,
      "num_tokens": 10981758.0,
      "step": 5940
    },
    {
      "entropy": 4.970957851409912,
      "epoch": 4.97489539748954,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003091453486649883,
      "loss": 4.6366,
      "mean_token_accuracy": 0.22454863935709,
      "num_tokens": 10991730.0,
      "step": 5945
    },
    {
      "entropy": 5.0134093284606935,
      "epoch": 4.97907949790795,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030882628561220376,
      "loss": 4.6458,
      "mean_token_accuracy": 0.22744067162275314,
      "num_tokens": 11000827.0,
      "step": 5950
    },
    {
      "entropy": 5.031310129165649,
      "epoch": 4.98326359832636,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030850715295036187,
      "loss": 4.5992,
      "mean_token_accuracy": 0.2287774607539177,
      "num_tokens": 11010544.0,
      "step": 5955
    },
    {
      "entropy": 5.00920467376709,
      "epoch": 4.98744769874477,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003081879513361863,
      "loss": 4.669,
      "mean_token_accuracy": 0.22432855516672134,
      "num_tokens": 11019826.0,
      "step": 5960
    },
    {
      "entropy": 5.065756320953369,
      "epoch": 4.99163179916318,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003078686814265426,
      "loss": 4.6459,
      "mean_token_accuracy": 0.2274928405880928,
      "num_tokens": 11028408.0,
      "step": 5965
    },
    {
      "entropy": 5.041246604919434,
      "epoch": 4.99581589958159,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000307549343878437,
      "loss": 4.6876,
      "mean_token_accuracy": 0.22406591922044755,
      "num_tokens": 11038564.0,
      "step": 5970
    },
    {
      "entropy": 4.965517568588257,
      "epoch": 5.0,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030722993934901487,
      "loss": 4.5551,
      "mean_token_accuracy": 0.23235881328582764,
      "num_tokens": 11046610.0,
      "step": 5975
    },
    {
      "entropy": 4.856838846206665,
      "epoch": 5.00418410041841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030691046849555927,
      "loss": 4.2573,
      "mean_token_accuracy": 0.25432285070419314,
      "num_tokens": 11055929.0,
      "step": 5980
    },
    {
      "entropy": 4.983529758453369,
      "epoch": 5.00836820083682,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030659093197549,
      "loss": 4.3473,
      "mean_token_accuracy": 0.2470561146736145,
      "num_tokens": 11064748.0,
      "step": 5985
    },
    {
      "entropy": 4.859720468521118,
      "epoch": 5.01255230125523,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003062713304463616,
      "loss": 4.3361,
      "mean_token_accuracy": 0.24246127754449845,
      "num_tokens": 11073707.0,
      "step": 5990
    },
    {
      "entropy": 4.939044857025147,
      "epoch": 5.01673640167364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003059516645658629,
      "loss": 4.3558,
      "mean_token_accuracy": 0.2486013278365135,
      "num_tokens": 11082967.0,
      "step": 5995
    },
    {
      "entropy": 4.90383472442627,
      "epoch": 5.02092050209205,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003056319349918146,
      "loss": 4.3594,
      "mean_token_accuracy": 0.24517649561166763,
      "num_tokens": 11092105.0,
      "step": 6000
    },
    {
      "epoch": 5.02092050209205,
      "eval_entropy": 4.784615506423763,
      "eval_loss": 5.362369537353516,
      "eval_mean_token_accuracy": 0.19662759102621938,
      "eval_num_tokens": 11092105.0,
      "eval_runtime": 2.7275,
      "eval_samples_per_second": 1432.429,
      "eval_steps_per_second": 179.283,
      "step": 6000
    },
    {
      "entropy": 4.979133176803589,
      "epoch": 5.02510460251046,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030531214238216906,
      "loss": 4.4361,
      "mean_token_accuracy": 0.23639662712812423,
      "num_tokens": 11100806.0,
      "step": 6005
    },
    {
      "entropy": 4.8986869812011715,
      "epoch": 5.02928870292887,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030499228739500793,
      "loss": 4.2912,
      "mean_token_accuracy": 0.25047903060913085,
      "num_tokens": 11109779.0,
      "step": 6010
    },
    {
      "entropy": 4.9464233875274655,
      "epoch": 5.03347280334728,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030467237068854136,
      "loss": 4.3778,
      "mean_token_accuracy": 0.24178319424390793,
      "num_tokens": 11118838.0,
      "step": 6015
    },
    {
      "entropy": 4.901123142242431,
      "epoch": 5.03765690376569,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003043523929211064,
      "loss": 4.3051,
      "mean_token_accuracy": 0.2513675928115845,
      "num_tokens": 11127029.0,
      "step": 6020
    },
    {
      "entropy": 4.910240507125854,
      "epoch": 5.0418410041841,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000304032354751166,
      "loss": 4.3617,
      "mean_token_accuracy": 0.24584041088819503,
      "num_tokens": 11136073.0,
      "step": 6025
    },
    {
      "entropy": 4.933545780181885,
      "epoch": 5.046025104602511,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003037122568373071,
      "loss": 4.4097,
      "mean_token_accuracy": 0.23784682154655457,
      "num_tokens": 11144807.0,
      "step": 6030
    },
    {
      "entropy": 5.004555511474609,
      "epoch": 5.050209205020921,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003033920998382399,
      "loss": 4.3818,
      "mean_token_accuracy": 0.2407732293009758,
      "num_tokens": 11154308.0,
      "step": 6035
    },
    {
      "entropy": 4.957700777053833,
      "epoch": 5.054393305439331,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003030718844127959,
      "loss": 4.4222,
      "mean_token_accuracy": 0.23837951123714446,
      "num_tokens": 11163752.0,
      "step": 6040
    },
    {
      "entropy": 4.915414524078369,
      "epoch": 5.058577405857741,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003027516112199271,
      "loss": 4.3858,
      "mean_token_accuracy": 0.24238771796226502,
      "num_tokens": 11173763.0,
      "step": 6045
    },
    {
      "entropy": 4.9684919834136965,
      "epoch": 5.062761506276151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030243128091870413,
      "loss": 4.4139,
      "mean_token_accuracy": 0.2301672801375389,
      "num_tokens": 11183577.0,
      "step": 6050
    },
    {
      "entropy": 4.9378808498382565,
      "epoch": 5.066945606694561,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003021108941683152,
      "loss": 4.3821,
      "mean_token_accuracy": 0.2421707585453987,
      "num_tokens": 11192708.0,
      "step": 6055
    },
    {
      "entropy": 4.92610354423523,
      "epoch": 5.071129707112971,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030179045162806494,
      "loss": 4.4434,
      "mean_token_accuracy": 0.23759933263063432,
      "num_tokens": 11201803.0,
      "step": 6060
    },
    {
      "entropy": 4.862002372741699,
      "epoch": 5.075313807531381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030146995395737247,
      "loss": 4.3181,
      "mean_token_accuracy": 0.23811867088079453,
      "num_tokens": 11211208.0,
      "step": 6065
    },
    {
      "entropy": 4.963351678848267,
      "epoch": 5.079497907949791,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003011494018157704,
      "loss": 4.3652,
      "mean_token_accuracy": 0.25035039633512496,
      "num_tokens": 11221091.0,
      "step": 6070
    },
    {
      "entropy": 4.946651458740234,
      "epoch": 5.083682008368201,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003008287958629036,
      "loss": 4.4224,
      "mean_token_accuracy": 0.24136436730623245,
      "num_tokens": 11230084.0,
      "step": 6075
    },
    {
      "entropy": 4.939234590530395,
      "epoch": 5.087866108786611,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030050813675852747,
      "loss": 4.4669,
      "mean_token_accuracy": 0.23908335268497466,
      "num_tokens": 11239310.0,
      "step": 6080
    },
    {
      "entropy": 4.9443622589111325,
      "epoch": 5.092050209205021,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030018742516250697,
      "loss": 4.3405,
      "mean_token_accuracy": 0.2480589672923088,
      "num_tokens": 11248738.0,
      "step": 6085
    },
    {
      "entropy": 4.971943616867065,
      "epoch": 5.096234309623431,
      "grad_norm": 1.3125,
      "learning_rate": 0.000299866661734815,
      "loss": 4.3659,
      "mean_token_accuracy": 0.24639424085617065,
      "num_tokens": 11257813.0,
      "step": 6090
    },
    {
      "entropy": 4.99667797088623,
      "epoch": 5.100418410041841,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029954584713553107,
      "loss": 4.4138,
      "mean_token_accuracy": 0.23487891405820846,
      "num_tokens": 11266475.0,
      "step": 6095
    },
    {
      "entropy": 4.867139053344727,
      "epoch": 5.104602510460251,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029922498202484014,
      "loss": 4.4018,
      "mean_token_accuracy": 0.24182300865650178,
      "num_tokens": 11276011.0,
      "step": 6100
    },
    {
      "entropy": 4.929363489151001,
      "epoch": 5.108786610878661,
      "grad_norm": 1.234375,
      "learning_rate": 0.000298904067063031,
      "loss": 4.4537,
      "mean_token_accuracy": 0.23932157158851625,
      "num_tokens": 11284619.0,
      "step": 6105
    },
    {
      "entropy": 4.935394954681397,
      "epoch": 5.112970711297071,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002985831029104949,
      "loss": 4.3784,
      "mean_token_accuracy": 0.2490593358874321,
      "num_tokens": 11293105.0,
      "step": 6110
    },
    {
      "entropy": 4.927911710739136,
      "epoch": 5.117154811715481,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002982620902277246,
      "loss": 4.3713,
      "mean_token_accuracy": 0.23700384944677352,
      "num_tokens": 11302717.0,
      "step": 6115
    },
    {
      "entropy": 5.027009439468384,
      "epoch": 5.121338912133891,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029794102967531263,
      "loss": 4.4756,
      "mean_token_accuracy": 0.23813671618700027,
      "num_tokens": 11312709.0,
      "step": 6120
    },
    {
      "entropy": 4.909305810928345,
      "epoch": 5.125523012552302,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029761992191394985,
      "loss": 4.373,
      "mean_token_accuracy": 0.24234936833381654,
      "num_tokens": 11322040.0,
      "step": 6125
    },
    {
      "entropy": 4.925977563858032,
      "epoch": 5.129707112970712,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029729876760442456,
      "loss": 4.464,
      "mean_token_accuracy": 0.23859170973300933,
      "num_tokens": 11332029.0,
      "step": 6130
    },
    {
      "entropy": 4.920955419540405,
      "epoch": 5.133891213389122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002969775674076206,
      "loss": 4.4439,
      "mean_token_accuracy": 0.2352142333984375,
      "num_tokens": 11340960.0,
      "step": 6135
    },
    {
      "entropy": 5.0175714015960695,
      "epoch": 5.138075313807532,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002966563219845164,
      "loss": 4.445,
      "mean_token_accuracy": 0.23548276722431183,
      "num_tokens": 11349497.0,
      "step": 6140
    },
    {
      "entropy": 5.00672082901001,
      "epoch": 5.142259414225942,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002963350319961832,
      "loss": 4.394,
      "mean_token_accuracy": 0.24438609033823014,
      "num_tokens": 11358477.0,
      "step": 6145
    },
    {
      "entropy": 4.9383666038513185,
      "epoch": 5.146443514644352,
      "grad_norm": 1.25,
      "learning_rate": 0.0002960136981037844,
      "loss": 4.4057,
      "mean_token_accuracy": 0.23930297046899796,
      "num_tokens": 11367635.0,
      "step": 6150
    },
    {
      "entropy": 4.936536741256714,
      "epoch": 5.150627615062762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002956923209685733,
      "loss": 4.3603,
      "mean_token_accuracy": 0.24261585474014283,
      "num_tokens": 11376737.0,
      "step": 6155
    },
    {
      "entropy": 4.889450454711914,
      "epoch": 5.154811715481172,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002953709012518925,
      "loss": 4.4327,
      "mean_token_accuracy": 0.23794690817594527,
      "num_tokens": 11386756.0,
      "step": 6160
    },
    {
      "entropy": 4.888554430007934,
      "epoch": 5.158995815899582,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029504943961517204,
      "loss": 4.3224,
      "mean_token_accuracy": 0.24895768463611603,
      "num_tokens": 11394889.0,
      "step": 6165
    },
    {
      "entropy": 4.99803581237793,
      "epoch": 5.163179916317992,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029472793671992824,
      "loss": 4.496,
      "mean_token_accuracy": 0.23923881053924562,
      "num_tokens": 11404759.0,
      "step": 6170
    },
    {
      "entropy": 5.017521810531616,
      "epoch": 5.167364016736402,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029440639322776246,
      "loss": 4.4433,
      "mean_token_accuracy": 0.23571338653564453,
      "num_tokens": 11413739.0,
      "step": 6175
    },
    {
      "entropy": 4.9171974658966064,
      "epoch": 5.171548117154812,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029408480980035943,
      "loss": 4.4497,
      "mean_token_accuracy": 0.24457310289144515,
      "num_tokens": 11422835.0,
      "step": 6180
    },
    {
      "entropy": 4.976303148269653,
      "epoch": 5.175732217573222,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002937631870994863,
      "loss": 4.4251,
      "mean_token_accuracy": 0.24418449848890306,
      "num_tokens": 11432613.0,
      "step": 6185
    },
    {
      "entropy": 4.9148296356201175,
      "epoch": 5.179916317991632,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029344152578699074,
      "loss": 4.3942,
      "mean_token_accuracy": 0.23362365514039993,
      "num_tokens": 11441297.0,
      "step": 6190
    },
    {
      "entropy": 4.894047546386719,
      "epoch": 5.184100418410042,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029311982652479993,
      "loss": 4.3544,
      "mean_token_accuracy": 0.23848503082990646,
      "num_tokens": 11450409.0,
      "step": 6195
    },
    {
      "entropy": 4.945587253570556,
      "epoch": 5.188284518828452,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002927980899749195,
      "loss": 4.4653,
      "mean_token_accuracy": 0.23134004175662995,
      "num_tokens": 11460132.0,
      "step": 6200
    },
    {
      "entropy": 4.925260877609253,
      "epoch": 5.192468619246862,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002924763167994312,
      "loss": 4.4431,
      "mean_token_accuracy": 0.24150592237710952,
      "num_tokens": 11468984.0,
      "step": 6205
    },
    {
      "entropy": 5.008875989913941,
      "epoch": 5.196652719665272,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002921545076604928,
      "loss": 4.4783,
      "mean_token_accuracy": 0.2434159517288208,
      "num_tokens": 11478371.0,
      "step": 6210
    },
    {
      "entropy": 4.9509728908538815,
      "epoch": 5.200836820083682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029183266322033565,
      "loss": 4.4472,
      "mean_token_accuracy": 0.23509223461151124,
      "num_tokens": 11487924.0,
      "step": 6215
    },
    {
      "entropy": 4.963478326797485,
      "epoch": 5.205020920502092,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002915107841412638,
      "loss": 4.4281,
      "mean_token_accuracy": 0.23672900646924971,
      "num_tokens": 11497892.0,
      "step": 6220
    },
    {
      "entropy": 4.904073762893677,
      "epoch": 5.209205020920502,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002911888710856528,
      "loss": 4.3114,
      "mean_token_accuracy": 0.24688039422035218,
      "num_tokens": 11507266.0,
      "step": 6225
    },
    {
      "entropy": 4.840623950958252,
      "epoch": 5.2133891213389125,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002908669247159477,
      "loss": 4.305,
      "mean_token_accuracy": 0.2532154396176338,
      "num_tokens": 11516672.0,
      "step": 6230
    },
    {
      "entropy": 4.909411096572876,
      "epoch": 5.2175732217573225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029054494569466266,
      "loss": 4.4421,
      "mean_token_accuracy": 0.2361066684126854,
      "num_tokens": 11525865.0,
      "step": 6235
    },
    {
      "entropy": 4.972150039672852,
      "epoch": 5.2217573221757325,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029022293468437853,
      "loss": 4.4652,
      "mean_token_accuracy": 0.24156434684991837,
      "num_tokens": 11534910.0,
      "step": 6240
    },
    {
      "entropy": 4.858953046798706,
      "epoch": 5.2259414225941425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028990089234774236,
      "loss": 4.3285,
      "mean_token_accuracy": 0.2511567175388336,
      "num_tokens": 11543507.0,
      "step": 6245
    },
    {
      "entropy": 4.902063417434692,
      "epoch": 5.2301255230125525,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028957881934746544,
      "loss": 4.347,
      "mean_token_accuracy": 0.2448756903409958,
      "num_tokens": 11552562.0,
      "step": 6250
    },
    {
      "entropy": 4.980932760238647,
      "epoch": 5.2343096234309625,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028925671634632216,
      "loss": 4.4124,
      "mean_token_accuracy": 0.2383146822452545,
      "num_tokens": 11561931.0,
      "step": 6255
    },
    {
      "entropy": 4.94309778213501,
      "epoch": 5.2384937238493725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028893458400714876,
      "loss": 4.4417,
      "mean_token_accuracy": 0.23999371975660325,
      "num_tokens": 11570639.0,
      "step": 6260
    },
    {
      "entropy": 4.907566404342651,
      "epoch": 5.2426778242677825,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002886124229928418,
      "loss": 4.3692,
      "mean_token_accuracy": 0.24475771337747573,
      "num_tokens": 11579796.0,
      "step": 6265
    },
    {
      "entropy": 4.957763957977295,
      "epoch": 5.2468619246861925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028829023396635683,
      "loss": 4.4623,
      "mean_token_accuracy": 0.24178697168827057,
      "num_tokens": 11589343.0,
      "step": 6270
    },
    {
      "entropy": 5.006877851486206,
      "epoch": 5.2510460251046025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002879680175907071,
      "loss": 4.4552,
      "mean_token_accuracy": 0.24295152723789215,
      "num_tokens": 11598633.0,
      "step": 6275
    },
    {
      "entropy": 4.917856740951538,
      "epoch": 5.2552301255230125,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002876457745289621,
      "loss": 4.3278,
      "mean_token_accuracy": 0.24598244279623033,
      "num_tokens": 11607746.0,
      "step": 6280
    },
    {
      "entropy": 4.965209197998047,
      "epoch": 5.2594142259414225,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002873235054442463,
      "loss": 4.4447,
      "mean_token_accuracy": 0.23381512016057968,
      "num_tokens": 11616574.0,
      "step": 6285
    },
    {
      "entropy": 4.9261744022369385,
      "epoch": 5.2635983263598325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028700121099973765,
      "loss": 4.4564,
      "mean_token_accuracy": 0.24122796356678008,
      "num_tokens": 11625918.0,
      "step": 6290
    },
    {
      "entropy": 4.9476776123046875,
      "epoch": 5.2677824267782425,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002866788918586661,
      "loss": 4.4277,
      "mean_token_accuracy": 0.24722553491592408,
      "num_tokens": 11634255.0,
      "step": 6295
    },
    {
      "entropy": 4.939304780960083,
      "epoch": 5.2719665271966525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028635654868431286,
      "loss": 4.4207,
      "mean_token_accuracy": 0.23400188982486725,
      "num_tokens": 11643532.0,
      "step": 6300
    },
    {
      "entropy": 4.916950511932373,
      "epoch": 5.2761506276150625,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028603418214000827,
      "loss": 4.3956,
      "mean_token_accuracy": 0.23384817093610763,
      "num_tokens": 11653104.0,
      "step": 6305
    },
    {
      "entropy": 4.946879911422729,
      "epoch": 5.2803347280334725,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002857117928891308,
      "loss": 4.4694,
      "mean_token_accuracy": 0.23803068250417708,
      "num_tokens": 11662752.0,
      "step": 6310
    },
    {
      "entropy": 4.860090970993042,
      "epoch": 5.2845188284518825,
      "grad_norm": 1.375,
      "learning_rate": 0.00028538938159510587,
      "loss": 4.3375,
      "mean_token_accuracy": 0.25136750787496565,
      "num_tokens": 11671566.0,
      "step": 6315
    },
    {
      "entropy": 4.92877779006958,
      "epoch": 5.2887029288702925,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028506694892140385,
      "loss": 4.4254,
      "mean_token_accuracy": 0.2452017232775688,
      "num_tokens": 11681770.0,
      "step": 6320
    },
    {
      "entropy": 4.878110027313232,
      "epoch": 5.292887029288703,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028474449553153944,
      "loss": 4.4185,
      "mean_token_accuracy": 0.24156080782413483,
      "num_tokens": 11691177.0,
      "step": 6325
    },
    {
      "entropy": 4.875063228607178,
      "epoch": 5.297071129707113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028442202208906993,
      "loss": 4.3594,
      "mean_token_accuracy": 0.2483922243118286,
      "num_tokens": 11700461.0,
      "step": 6330
    },
    {
      "entropy": 4.910170984268189,
      "epoch": 5.301255230125523,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002840995292575938,
      "loss": 4.3954,
      "mean_token_accuracy": 0.23553913682699204,
      "num_tokens": 11708783.0,
      "step": 6335
    },
    {
      "entropy": 4.948196601867676,
      "epoch": 5.305439330543933,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002837770177007494,
      "loss": 4.4474,
      "mean_token_accuracy": 0.23731833398342134,
      "num_tokens": 11717419.0,
      "step": 6340
    },
    {
      "entropy": 4.963154935836792,
      "epoch": 5.309623430962343,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028345448808221374,
      "loss": 4.4757,
      "mean_token_accuracy": 0.23552718460559846,
      "num_tokens": 11727014.0,
      "step": 6345
    },
    {
      "entropy": 4.939190864562988,
      "epoch": 5.313807531380753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028313194106570095,
      "loss": 4.4322,
      "mean_token_accuracy": 0.23740191757678986,
      "num_tokens": 11736718.0,
      "step": 6350
    },
    {
      "entropy": 5.02469391822815,
      "epoch": 5.317991631799163,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002828093773149607,
      "loss": 4.5077,
      "mean_token_accuracy": 0.23355724811553955,
      "num_tokens": 11747236.0,
      "step": 6355
    },
    {
      "entropy": 4.966774797439575,
      "epoch": 5.322175732217573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028248679749377754,
      "loss": 4.4174,
      "mean_token_accuracy": 0.2411521092057228,
      "num_tokens": 11756912.0,
      "step": 6360
    },
    {
      "entropy": 4.859122943878174,
      "epoch": 5.326359832635983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028216420226596884,
      "loss": 4.3724,
      "mean_token_accuracy": 0.24656179845333098,
      "num_tokens": 11766035.0,
      "step": 6365
    },
    {
      "entropy": 4.836584377288818,
      "epoch": 5.330543933054393,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002818415922953837,
      "loss": 4.3358,
      "mean_token_accuracy": 0.24803601056337357,
      "num_tokens": 11775934.0,
      "step": 6370
    },
    {
      "entropy": 4.938327407836914,
      "epoch": 5.334728033472803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028151896824590156,
      "loss": 4.4189,
      "mean_token_accuracy": 0.24061724096536635,
      "num_tokens": 11784549.0,
      "step": 6375
    },
    {
      "entropy": 4.931270885467529,
      "epoch": 5.338912133891213,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002811963307814309,
      "loss": 4.4678,
      "mean_token_accuracy": 0.23051557093858718,
      "num_tokens": 11794369.0,
      "step": 6380
    },
    {
      "entropy": 4.918437147140503,
      "epoch": 5.343096234309623,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028087368056590767,
      "loss": 4.4354,
      "mean_token_accuracy": 0.237606044113636,
      "num_tokens": 11803221.0,
      "step": 6385
    },
    {
      "entropy": 4.967325210571289,
      "epoch": 5.347280334728033,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002805510182632942,
      "loss": 4.4361,
      "mean_token_accuracy": 0.24399203807115555,
      "num_tokens": 11812811.0,
      "step": 6390
    },
    {
      "entropy": 4.98750262260437,
      "epoch": 5.351464435146443,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002802283445375775,
      "loss": 4.4413,
      "mean_token_accuracy": 0.2419428676366806,
      "num_tokens": 11821522.0,
      "step": 6395
    },
    {
      "entropy": 4.933959913253784,
      "epoch": 5.355648535564853,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027990566005276846,
      "loss": 4.4188,
      "mean_token_accuracy": 0.2418348327279091,
      "num_tokens": 11830021.0,
      "step": 6400
    },
    {
      "entropy": 4.895017290115357,
      "epoch": 5.359832635983263,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027958296547289966,
      "loss": 4.3797,
      "mean_token_accuracy": 0.24695829600095748,
      "num_tokens": 11838612.0,
      "step": 6405
    },
    {
      "entropy": 4.891553163528442,
      "epoch": 5.364016736401673,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027926026146202475,
      "loss": 4.3261,
      "mean_token_accuracy": 0.25527317970991137,
      "num_tokens": 11847648.0,
      "step": 6410
    },
    {
      "entropy": 4.8549212455749515,
      "epoch": 5.368200836820083,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027893754868421665,
      "loss": 4.3631,
      "mean_token_accuracy": 0.2430718496441841,
      "num_tokens": 11857008.0,
      "step": 6415
    },
    {
      "entropy": 4.936502742767334,
      "epoch": 5.372384937238493,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002786148278035665,
      "loss": 4.386,
      "mean_token_accuracy": 0.252345535159111,
      "num_tokens": 11866055.0,
      "step": 6420
    },
    {
      "entropy": 4.957223844528198,
      "epoch": 5.376569037656903,
      "grad_norm": 1.375,
      "learning_rate": 0.0002782920994841819,
      "loss": 4.4627,
      "mean_token_accuracy": 0.23284538984298705,
      "num_tokens": 11875496.0,
      "step": 6425
    },
    {
      "entropy": 4.893859720230102,
      "epoch": 5.380753138075314,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002779693643901858,
      "loss": 4.3848,
      "mean_token_accuracy": 0.24119115769863128,
      "num_tokens": 11884115.0,
      "step": 6430
    },
    {
      "entropy": 4.962661361694336,
      "epoch": 5.384937238493724,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027764662318571526,
      "loss": 4.5426,
      "mean_token_accuracy": 0.23258035033941268,
      "num_tokens": 11893930.0,
      "step": 6435
    },
    {
      "entropy": 4.916618490219117,
      "epoch": 5.389121338912134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002773238765349196,
      "loss": 4.4106,
      "mean_token_accuracy": 0.23833373785018921,
      "num_tokens": 11902702.0,
      "step": 6440
    },
    {
      "entropy": 4.937559509277344,
      "epoch": 5.393305439330544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002770011251019598,
      "loss": 4.449,
      "mean_token_accuracy": 0.23960407823324203,
      "num_tokens": 11911915.0,
      "step": 6445
    },
    {
      "entropy": 4.884865236282349,
      "epoch": 5.397489539748954,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027667836955100615,
      "loss": 4.4027,
      "mean_token_accuracy": 0.2373399019241333,
      "num_tokens": 11920737.0,
      "step": 6450
    },
    {
      "entropy": 4.93814058303833,
      "epoch": 5.401673640167364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002763556105462378,
      "loss": 4.4143,
      "mean_token_accuracy": 0.24265306293964387,
      "num_tokens": 11929548.0,
      "step": 6455
    },
    {
      "entropy": 4.938182592391968,
      "epoch": 5.405857740585774,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027603284875184095,
      "loss": 4.3976,
      "mean_token_accuracy": 0.2405654713511467,
      "num_tokens": 11938459.0,
      "step": 6460
    },
    {
      "entropy": 4.963888120651245,
      "epoch": 5.410041841004184,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027571008483200744,
      "loss": 4.4066,
      "mean_token_accuracy": 0.24538826197385788,
      "num_tokens": 11947282.0,
      "step": 6465
    },
    {
      "entropy": 4.890219831466675,
      "epoch": 5.414225941422594,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027538731945093346,
      "loss": 4.3992,
      "mean_token_accuracy": 0.2380546286702156,
      "num_tokens": 11956209.0,
      "step": 6470
    },
    {
      "entropy": 4.885936737060547,
      "epoch": 5.418410041841004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027506455327281834,
      "loss": 4.4221,
      "mean_token_accuracy": 0.2409121111035347,
      "num_tokens": 11965689.0,
      "step": 6475
    },
    {
      "entropy": 4.9389392852783205,
      "epoch": 5.422594142259414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000274741786961863,
      "loss": 4.4324,
      "mean_token_accuracy": 0.23651644587516785,
      "num_tokens": 11975622.0,
      "step": 6480
    },
    {
      "entropy": 4.930453109741211,
      "epoch": 5.426778242677824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002744190211822685,
      "loss": 4.4919,
      "mean_token_accuracy": 0.23572517037391663,
      "num_tokens": 11984904.0,
      "step": 6485
    },
    {
      "entropy": 4.974074792861939,
      "epoch": 5.430962343096234,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000274096256598235,
      "loss": 4.5214,
      "mean_token_accuracy": 0.23061008155345916,
      "num_tokens": 11995456.0,
      "step": 6490
    },
    {
      "entropy": 4.950842618942261,
      "epoch": 5.435146443514644,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027377349387396017,
      "loss": 4.4048,
      "mean_token_accuracy": 0.23655006736516954,
      "num_tokens": 12004566.0,
      "step": 6495
    },
    {
      "entropy": 4.884503602981567,
      "epoch": 5.439330543933054,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027345073367363775,
      "loss": 4.3459,
      "mean_token_accuracy": 0.24681381583213807,
      "num_tokens": 12013304.0,
      "step": 6500
    },
    {
      "epoch": 5.439330543933054,
      "eval_entropy": 4.753458721749865,
      "eval_loss": 5.356199264526367,
      "eval_mean_token_accuracy": 0.19882751454300188,
      "eval_num_tokens": 12013304.0,
      "eval_runtime": 2.7444,
      "eval_samples_per_second": 1423.621,
      "eval_steps_per_second": 178.18,
      "step": 6500
    },
    {
      "entropy": 4.831784105300903,
      "epoch": 5.443514644351464,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002731279766614563,
      "loss": 4.378,
      "mean_token_accuracy": 0.24216004461050034,
      "num_tokens": 12022007.0,
      "step": 6505
    },
    {
      "entropy": 4.895554161071777,
      "epoch": 5.447698744769874,
      "grad_norm": 1.265625,
      "learning_rate": 0.000272805223501598,
      "loss": 4.4273,
      "mean_token_accuracy": 0.24429392963647842,
      "num_tokens": 12031493.0,
      "step": 6510
    },
    {
      "entropy": 4.901669931411743,
      "epoch": 5.451882845188284,
      "grad_norm": 1.25,
      "learning_rate": 0.0002724824748582369,
      "loss": 4.4059,
      "mean_token_accuracy": 0.24256129562854767,
      "num_tokens": 12040143.0,
      "step": 6515
    },
    {
      "entropy": 5.027035140991211,
      "epoch": 5.456066945606695,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002721597313955376,
      "loss": 4.5741,
      "mean_token_accuracy": 0.2293592020869255,
      "num_tokens": 12050136.0,
      "step": 6520
    },
    {
      "entropy": 4.943906545639038,
      "epoch": 5.460251046025105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002718369937776546,
      "loss": 4.442,
      "mean_token_accuracy": 0.23771191239356995,
      "num_tokens": 12060065.0,
      "step": 6525
    },
    {
      "entropy": 5.00283727645874,
      "epoch": 5.464435146443515,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002715142626687298,
      "loss": 4.4816,
      "mean_token_accuracy": 0.225738126039505,
      "num_tokens": 12069175.0,
      "step": 6530
    },
    {
      "entropy": 4.924898672103882,
      "epoch": 5.468619246861925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000271191538732892,
      "loss": 4.3945,
      "mean_token_accuracy": 0.248943030834198,
      "num_tokens": 12077308.0,
      "step": 6535
    },
    {
      "entropy": 4.868170642852784,
      "epoch": 5.472803347280335,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027086882263425527,
      "loss": 4.3469,
      "mean_token_accuracy": 0.24434713125228882,
      "num_tokens": 12085678.0,
      "step": 6540
    },
    {
      "entropy": 4.9272254467010494,
      "epoch": 5.476987447698745,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027054611503691736,
      "loss": 4.4727,
      "mean_token_accuracy": 0.235773766040802,
      "num_tokens": 12095163.0,
      "step": 6545
    },
    {
      "entropy": 4.9810584545135494,
      "epoch": 5.481171548117155,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027022341660495853,
      "loss": 4.4615,
      "mean_token_accuracy": 0.2322252631187439,
      "num_tokens": 12104185.0,
      "step": 6550
    },
    {
      "entropy": 4.943159389495849,
      "epoch": 5.485355648535565,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026990072800244045,
      "loss": 4.4853,
      "mean_token_accuracy": 0.22796425521373748,
      "num_tokens": 12114298.0,
      "step": 6555
    },
    {
      "entropy": 4.905063915252685,
      "epoch": 5.489539748953975,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026957804989340427,
      "loss": 4.4329,
      "mean_token_accuracy": 0.2471110925078392,
      "num_tokens": 12122512.0,
      "step": 6560
    },
    {
      "entropy": 4.883237552642822,
      "epoch": 5.493723849372385,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002692553829418697,
      "loss": 4.3661,
      "mean_token_accuracy": 0.24588553458452225,
      "num_tokens": 12132133.0,
      "step": 6565
    },
    {
      "entropy": 4.908546018600464,
      "epoch": 5.497907949790795,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002689327278118334,
      "loss": 4.4243,
      "mean_token_accuracy": 0.24072849601507187,
      "num_tokens": 12141299.0,
      "step": 6570
    },
    {
      "entropy": 5.047840929031372,
      "epoch": 5.502092050209205,
      "grad_norm": 1.3125,
      "learning_rate": 0.000268610085167268,
      "loss": 4.5139,
      "mean_token_accuracy": 0.23675930202007295,
      "num_tokens": 12150412.0,
      "step": 6575
    },
    {
      "entropy": 4.94555778503418,
      "epoch": 5.506276150627615,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002682874556721198,
      "loss": 4.4462,
      "mean_token_accuracy": 0.23309631645679474,
      "num_tokens": 12159754.0,
      "step": 6580
    },
    {
      "entropy": 4.94200325012207,
      "epoch": 5.510460251046025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002679648399903087,
      "loss": 4.4183,
      "mean_token_accuracy": 0.24340333193540573,
      "num_tokens": 12169098.0,
      "step": 6585
    },
    {
      "entropy": 4.929845666885376,
      "epoch": 5.514644351464435,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002676422387857258,
      "loss": 4.3956,
      "mean_token_accuracy": 0.2484296292066574,
      "num_tokens": 12178304.0,
      "step": 6590
    },
    {
      "entropy": 4.8538493633270265,
      "epoch": 5.518828451882845,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026731965272223246,
      "loss": 4.3881,
      "mean_token_accuracy": 0.24849053621292114,
      "num_tokens": 12188029.0,
      "step": 6595
    },
    {
      "entropy": 4.904757881164551,
      "epoch": 5.523012552301255,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002669970824636591,
      "loss": 4.3708,
      "mean_token_accuracy": 0.2378548949956894,
      "num_tokens": 12196663.0,
      "step": 6600
    },
    {
      "entropy": 4.92256350517273,
      "epoch": 5.527196652719665,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002666745286738033,
      "loss": 4.4949,
      "mean_token_accuracy": 0.23056450188159944,
      "num_tokens": 12206496.0,
      "step": 6605
    },
    {
      "entropy": 4.980793046951294,
      "epoch": 5.531380753138075,
      "grad_norm": 1.125,
      "learning_rate": 0.00026635199201642914,
      "loss": 4.4696,
      "mean_token_accuracy": 0.23839080929756165,
      "num_tokens": 12216597.0,
      "step": 6610
    },
    {
      "entropy": 4.890468978881836,
      "epoch": 5.535564853556485,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002660294731552649,
      "loss": 4.4172,
      "mean_token_accuracy": 0.24145790934562683,
      "num_tokens": 12225125.0,
      "step": 6615
    },
    {
      "entropy": 4.867322492599487,
      "epoch": 5.539748953974895,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026570697275400267,
      "loss": 4.3809,
      "mean_token_accuracy": 0.24067502319812775,
      "num_tokens": 12234402.0,
      "step": 6620
    },
    {
      "entropy": 4.860488080978394,
      "epoch": 5.543933054393305,
      "grad_norm": 1.25,
      "learning_rate": 0.0002653844914762964,
      "loss": 4.3524,
      "mean_token_accuracy": 0.23982334434986113,
      "num_tokens": 12244136.0,
      "step": 6625
    },
    {
      "entropy": 4.967105150222778,
      "epoch": 5.548117154811716,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002650620299857607,
      "loss": 4.4064,
      "mean_token_accuracy": 0.2431383788585663,
      "num_tokens": 12253039.0,
      "step": 6630
    },
    {
      "entropy": 4.974341297149659,
      "epoch": 5.552301255230126,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026473958894596963,
      "loss": 4.4342,
      "mean_token_accuracy": 0.24317444115877151,
      "num_tokens": 12263172.0,
      "step": 6635
    },
    {
      "entropy": 4.91245846748352,
      "epoch": 5.556485355648536,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000264417169020455,
      "loss": 4.4678,
      "mean_token_accuracy": 0.2331171378493309,
      "num_tokens": 12273324.0,
      "step": 6640
    },
    {
      "entropy": 4.966329908370971,
      "epoch": 5.560669456066946,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002640947708727049,
      "loss": 4.523,
      "mean_token_accuracy": 0.22865411788225173,
      "num_tokens": 12283719.0,
      "step": 6645
    },
    {
      "entropy": 4.904250144958496,
      "epoch": 5.564853556485356,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002637723951661631,
      "loss": 4.4571,
      "mean_token_accuracy": 0.23165811896324157,
      "num_tokens": 12293451.0,
      "step": 6650
    },
    {
      "entropy": 4.997565841674804,
      "epoch": 5.569037656903766,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002634500425642268,
      "loss": 4.5312,
      "mean_token_accuracy": 0.2290041148662567,
      "num_tokens": 12302560.0,
      "step": 6655
    },
    {
      "entropy": 5.003148984909058,
      "epoch": 5.573221757322176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002631277137302461,
      "loss": 4.4809,
      "mean_token_accuracy": 0.22768278867006303,
      "num_tokens": 12312115.0,
      "step": 6660
    },
    {
      "entropy": 4.8941388607025145,
      "epoch": 5.577405857740586,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026280540932752176,
      "loss": 4.3446,
      "mean_token_accuracy": 0.25246768444776535,
      "num_tokens": 12320689.0,
      "step": 6665
    },
    {
      "entropy": 4.85246844291687,
      "epoch": 5.581589958158996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026248313001930445,
      "loss": 4.3422,
      "mean_token_accuracy": 0.24498029947280883,
      "num_tokens": 12329643.0,
      "step": 6670
    },
    {
      "entropy": 4.919214391708374,
      "epoch": 5.585774058577406,
      "grad_norm": 1.25,
      "learning_rate": 0.00026216087646879326,
      "loss": 4.4587,
      "mean_token_accuracy": 0.24117285162210464,
      "num_tokens": 12339267.0,
      "step": 6675
    },
    {
      "entropy": 5.00288987159729,
      "epoch": 5.589958158995816,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002618386493391339,
      "loss": 4.5303,
      "mean_token_accuracy": 0.23370158076286315,
      "num_tokens": 12348515.0,
      "step": 6680
    },
    {
      "entropy": 4.94783124923706,
      "epoch": 5.594142259414226,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026151644929341833,
      "loss": 4.4899,
      "mean_token_accuracy": 0.23416378647089003,
      "num_tokens": 12358023.0,
      "step": 6685
    },
    {
      "entropy": 4.901445722579956,
      "epoch": 5.598326359832636,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026119427699468226,
      "loss": 4.4017,
      "mean_token_accuracy": 0.2440904498100281,
      "num_tokens": 12367129.0,
      "step": 6690
    },
    {
      "entropy": 4.88800950050354,
      "epoch": 5.602510460251046,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026087213310590455,
      "loss": 4.4118,
      "mean_token_accuracy": 0.24863233268260956,
      "num_tokens": 12375933.0,
      "step": 6695
    },
    {
      "entropy": 4.864757108688354,
      "epoch": 5.606694560669456,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026055001829000543,
      "loss": 4.4078,
      "mean_token_accuracy": 0.24556056559085845,
      "num_tokens": 12384784.0,
      "step": 6700
    },
    {
      "entropy": 4.911721515655517,
      "epoch": 5.610878661087866,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026022793320984553,
      "loss": 4.4415,
      "mean_token_accuracy": 0.24079318791627885,
      "num_tokens": 12393933.0,
      "step": 6705
    },
    {
      "entropy": 4.94800033569336,
      "epoch": 5.615062761506276,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002599058785282241,
      "loss": 4.4525,
      "mean_token_accuracy": 0.23182492554187775,
      "num_tokens": 12403442.0,
      "step": 6710
    },
    {
      "entropy": 4.898030710220337,
      "epoch": 5.619246861924686,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025958385490787797,
      "loss": 4.3698,
      "mean_token_accuracy": 0.25023130923509596,
      "num_tokens": 12412317.0,
      "step": 6715
    },
    {
      "entropy": 4.880168390274048,
      "epoch": 5.623430962343097,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002592618630114798,
      "loss": 4.3852,
      "mean_token_accuracy": 0.23724651485681533,
      "num_tokens": 12421459.0,
      "step": 6720
    },
    {
      "entropy": 4.919149541854859,
      "epoch": 5.627615062761507,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025893990350163724,
      "loss": 4.4495,
      "mean_token_accuracy": 0.2384878635406494,
      "num_tokens": 12431224.0,
      "step": 6725
    },
    {
      "entropy": 4.869623470306396,
      "epoch": 5.631799163179917,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025861797704089133,
      "loss": 4.357,
      "mean_token_accuracy": 0.23839060366153716,
      "num_tokens": 12439891.0,
      "step": 6730
    },
    {
      "entropy": 4.850959825515747,
      "epoch": 5.635983263598327,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002582960842917148,
      "loss": 4.3424,
      "mean_token_accuracy": 0.24046635031700134,
      "num_tokens": 12449852.0,
      "step": 6735
    },
    {
      "entropy": 4.962769842147827,
      "epoch": 5.640167364016737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002579742259165113,
      "loss": 4.4443,
      "mean_token_accuracy": 0.24487461149692535,
      "num_tokens": 12459058.0,
      "step": 6740
    },
    {
      "entropy": 4.972282981872558,
      "epoch": 5.644351464435147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002576524025776135,
      "loss": 4.4743,
      "mean_token_accuracy": 0.23543714731931686,
      "num_tokens": 12468684.0,
      "step": 6745
    },
    {
      "entropy": 4.898228549957276,
      "epoch": 5.648535564853557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002573306149372821,
      "loss": 4.4392,
      "mean_token_accuracy": 0.2478930115699768,
      "num_tokens": 12478102.0,
      "step": 6750
    },
    {
      "entropy": 4.935813903808594,
      "epoch": 5.652719665271967,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002570088636577044,
      "loss": 4.46,
      "mean_token_accuracy": 0.2370460882782936,
      "num_tokens": 12488036.0,
      "step": 6755
    },
    {
      "entropy": 4.991780567169189,
      "epoch": 5.656903765690377,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002566871494009928,
      "loss": 4.4323,
      "mean_token_accuracy": 0.23562087565660478,
      "num_tokens": 12496593.0,
      "step": 6760
    },
    {
      "entropy": 4.995029306411743,
      "epoch": 5.661087866108787,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002563654728291835,
      "loss": 4.519,
      "mean_token_accuracy": 0.22567374557256697,
      "num_tokens": 12505752.0,
      "step": 6765
    },
    {
      "entropy": 4.895358467102051,
      "epoch": 5.665271966527197,
      "grad_norm": 1.375,
      "learning_rate": 0.0002560438346042351,
      "loss": 4.4086,
      "mean_token_accuracy": 0.24609856754541398,
      "num_tokens": 12515192.0,
      "step": 6770
    },
    {
      "entropy": 4.971926975250244,
      "epoch": 5.669456066945607,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002557222353880275,
      "loss": 4.4851,
      "mean_token_accuracy": 0.23513922095298767,
      "num_tokens": 12525613.0,
      "step": 6775
    },
    {
      "entropy": 4.958098316192627,
      "epoch": 5.673640167364017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025540067584235997,
      "loss": 4.4558,
      "mean_token_accuracy": 0.23611305505037308,
      "num_tokens": 12535076.0,
      "step": 6780
    },
    {
      "entropy": 4.922168302536011,
      "epoch": 5.677824267782427,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025507915662895055,
      "loss": 4.4549,
      "mean_token_accuracy": 0.23815622925758362,
      "num_tokens": 12544333.0,
      "step": 6785
    },
    {
      "entropy": 4.9443594455719,
      "epoch": 5.682008368200837,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025475767840943403,
      "loss": 4.4137,
      "mean_token_accuracy": 0.24219993203878404,
      "num_tokens": 12554292.0,
      "step": 6790
    },
    {
      "entropy": 4.938510942459106,
      "epoch": 5.686192468619247,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002544362418453608,
      "loss": 4.4934,
      "mean_token_accuracy": 0.23289280533790588,
      "num_tokens": 12564240.0,
      "step": 6795
    },
    {
      "entropy": 4.909607458114624,
      "epoch": 5.690376569037657,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025411484759819584,
      "loss": 4.4118,
      "mean_token_accuracy": 0.24779236912727357,
      "num_tokens": 12573280.0,
      "step": 6800
    },
    {
      "entropy": 4.952965545654297,
      "epoch": 5.694560669456067,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025379349632931666,
      "loss": 4.5195,
      "mean_token_accuracy": 0.23496149778366088,
      "num_tokens": 12583133.0,
      "step": 6805
    },
    {
      "entropy": 4.9455070972442625,
      "epoch": 5.698744769874477,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002534721887000127,
      "loss": 4.4605,
      "mean_token_accuracy": 0.23753785640001296,
      "num_tokens": 12591709.0,
      "step": 6810
    },
    {
      "entropy": 4.951023578643799,
      "epoch": 5.702928870292887,
      "grad_norm": 1.375,
      "learning_rate": 0.0002531509253714831,
      "loss": 4.3943,
      "mean_token_accuracy": 0.24239160418510436,
      "num_tokens": 12600187.0,
      "step": 6815
    },
    {
      "entropy": 4.840760707855225,
      "epoch": 5.707112970711297,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002528297070048364,
      "loss": 4.3747,
      "mean_token_accuracy": 0.24387603849172593,
      "num_tokens": 12608910.0,
      "step": 6820
    },
    {
      "entropy": 4.85672082901001,
      "epoch": 5.711297071129707,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025250853426108814,
      "loss": 4.3554,
      "mean_token_accuracy": 0.2479291006922722,
      "num_tokens": 12617370.0,
      "step": 6825
    },
    {
      "entropy": 4.987340211868286,
      "epoch": 5.7154811715481175,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025218740780116037,
      "loss": 4.4442,
      "mean_token_accuracy": 0.24387548863887787,
      "num_tokens": 12626146.0,
      "step": 6830
    },
    {
      "entropy": 4.956174898147583,
      "epoch": 5.7196652719665275,
      "grad_norm": 1.375,
      "learning_rate": 0.00025186632828587977,
      "loss": 4.3957,
      "mean_token_accuracy": 0.24063418060541153,
      "num_tokens": 12635565.0,
      "step": 6835
    },
    {
      "entropy": 4.925436925888062,
      "epoch": 5.7238493723849375,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002515452963759761,
      "loss": 4.4865,
      "mean_token_accuracy": 0.23178475201129914,
      "num_tokens": 12644869.0,
      "step": 6840
    },
    {
      "entropy": 4.906055498123169,
      "epoch": 5.7280334728033475,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002512243127320817,
      "loss": 4.3838,
      "mean_token_accuracy": 0.24545985609292983,
      "num_tokens": 12655215.0,
      "step": 6845
    },
    {
      "entropy": 4.929975938796997,
      "epoch": 5.7322175732217575,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002509033780147291,
      "loss": 4.3626,
      "mean_token_accuracy": 0.2428971379995346,
      "num_tokens": 12663684.0,
      "step": 6850
    },
    {
      "entropy": 4.966853046417237,
      "epoch": 5.7364016736401675,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025058249288435037,
      "loss": 4.4571,
      "mean_token_accuracy": 0.2336862415075302,
      "num_tokens": 12673367.0,
      "step": 6855
    },
    {
      "entropy": 4.895157670974731,
      "epoch": 5.7405857740585775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002502616580012756,
      "loss": 4.4078,
      "mean_token_accuracy": 0.24551099836826323,
      "num_tokens": 12682472.0,
      "step": 6860
    },
    {
      "entropy": 4.876219844818115,
      "epoch": 5.7447698744769875,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024994087402573133,
      "loss": 4.3889,
      "mean_token_accuracy": 0.24304187297821045,
      "num_tokens": 12691215.0,
      "step": 6865
    },
    {
      "entropy": 4.8414239406585695,
      "epoch": 5.7489539748953975,
      "grad_norm": 1.25,
      "learning_rate": 0.00024962014161783946,
      "loss": 4.351,
      "mean_token_accuracy": 0.23853762000799178,
      "num_tokens": 12700332.0,
      "step": 6870
    },
    {
      "entropy": 4.879138422012329,
      "epoch": 5.7531380753138075,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024929946143761585,
      "loss": 4.409,
      "mean_token_accuracy": 0.2397585391998291,
      "num_tokens": 12708886.0,
      "step": 6875
    },
    {
      "entropy": 4.936747646331787,
      "epoch": 5.7573221757322175,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024897883414496844,
      "loss": 4.3957,
      "mean_token_accuracy": 0.2442208617925644,
      "num_tokens": 12717476.0,
      "step": 6880
    },
    {
      "entropy": 5.097469472885132,
      "epoch": 5.7615062761506275,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024865826039969683,
      "loss": 4.534,
      "mean_token_accuracy": 0.23407083004713058,
      "num_tokens": 12726737.0,
      "step": 6885
    },
    {
      "entropy": 4.916766023635864,
      "epoch": 5.7656903765690375,
      "grad_norm": 1.375,
      "learning_rate": 0.0002483377408614903,
      "loss": 4.4223,
      "mean_token_accuracy": 0.24064549505710603,
      "num_tokens": 12735750.0,
      "step": 6890
    },
    {
      "entropy": 4.842455339431763,
      "epoch": 5.7698744769874475,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024801727618992635,
      "loss": 4.3942,
      "mean_token_accuracy": 0.2363695979118347,
      "num_tokens": 12745143.0,
      "step": 6895
    },
    {
      "entropy": 4.919575643539429,
      "epoch": 5.7740585774058575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024769686704446996,
      "loss": 4.4218,
      "mean_token_accuracy": 0.23829784542322158,
      "num_tokens": 12754623.0,
      "step": 6900
    },
    {
      "entropy": 5.015362215042114,
      "epoch": 5.7782426778242675,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024737651408447155,
      "loss": 4.4444,
      "mean_token_accuracy": 0.2395350381731987,
      "num_tokens": 12762921.0,
      "step": 6905
    },
    {
      "entropy": 4.957454252243042,
      "epoch": 5.7824267782426775,
      "grad_norm": 1.234375,
      "learning_rate": 0.000247056217969166,
      "loss": 4.4958,
      "mean_token_accuracy": 0.23810326904058457,
      "num_tokens": 12772159.0,
      "step": 6910
    },
    {
      "entropy": 4.89158992767334,
      "epoch": 5.786610878661088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024673597935767123,
      "loss": 4.4636,
      "mean_token_accuracy": 0.23871254473924636,
      "num_tokens": 12781916.0,
      "step": 6915
    },
    {
      "entropy": 4.91577296257019,
      "epoch": 5.790794979079498,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002464157989089867,
      "loss": 4.4503,
      "mean_token_accuracy": 0.23965023905038835,
      "num_tokens": 12791275.0,
      "step": 6920
    },
    {
      "entropy": 4.927522468566894,
      "epoch": 5.794979079497908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002460956772819925,
      "loss": 4.421,
      "mean_token_accuracy": 0.24392324686050415,
      "num_tokens": 12800667.0,
      "step": 6925
    },
    {
      "entropy": 4.910789060592651,
      "epoch": 5.799163179916318,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024577561513544724,
      "loss": 4.4156,
      "mean_token_accuracy": 0.2409498244524002,
      "num_tokens": 12808986.0,
      "step": 6930
    },
    {
      "entropy": 4.927042722702026,
      "epoch": 5.803347280334728,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024545561312798756,
      "loss": 4.4519,
      "mean_token_accuracy": 0.2430231660604477,
      "num_tokens": 12817658.0,
      "step": 6935
    },
    {
      "entropy": 4.951921129226685,
      "epoch": 5.807531380753138,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024513567191812607,
      "loss": 4.4546,
      "mean_token_accuracy": 0.24040859043598176,
      "num_tokens": 12826301.0,
      "step": 6940
    },
    {
      "entropy": 5.034375286102295,
      "epoch": 5.811715481171548,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002448157921642505,
      "loss": 4.4852,
      "mean_token_accuracy": 0.23482562601566315,
      "num_tokens": 12836295.0,
      "step": 6945
    },
    {
      "entropy": 4.907037544250488,
      "epoch": 5.815899581589958,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024449597452462165,
      "loss": 4.4398,
      "mean_token_accuracy": 0.2379249557852745,
      "num_tokens": 12847128.0,
      "step": 6950
    },
    {
      "entropy": 4.879659795761109,
      "epoch": 5.820083682008368,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002441762196573729,
      "loss": 4.4654,
      "mean_token_accuracy": 0.23565457314252852,
      "num_tokens": 12856198.0,
      "step": 6955
    },
    {
      "entropy": 4.914762115478515,
      "epoch": 5.824267782426778,
      "grad_norm": 1.203125,
      "learning_rate": 0.00024385652822050846,
      "loss": 4.378,
      "mean_token_accuracy": 0.24151598513126374,
      "num_tokens": 12865513.0,
      "step": 6960
    },
    {
      "entropy": 4.960512447357178,
      "epoch": 5.828451882845188,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024353690087190188,
      "loss": 4.436,
      "mean_token_accuracy": 0.23041707277297974,
      "num_tokens": 12874086.0,
      "step": 6965
    },
    {
      "entropy": 4.888470029830932,
      "epoch": 5.832635983263598,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024321733826929487,
      "loss": 4.4331,
      "mean_token_accuracy": 0.241251377761364,
      "num_tokens": 12883382.0,
      "step": 6970
    },
    {
      "entropy": 5.003223991394043,
      "epoch": 5.836820083682008,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024289784107029588,
      "loss": 4.4711,
      "mean_token_accuracy": 0.23910648226737977,
      "num_tokens": 12893510.0,
      "step": 6975
    },
    {
      "entropy": 5.003660821914673,
      "epoch": 5.841004184100418,
      "grad_norm": 1.25,
      "learning_rate": 0.00024257840993237862,
      "loss": 4.4973,
      "mean_token_accuracy": 0.23477276861667634,
      "num_tokens": 12902719.0,
      "step": 6980
    },
    {
      "entropy": 4.942154407501221,
      "epoch": 5.845188284518828,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024225904551288124,
      "loss": 4.514,
      "mean_token_accuracy": 0.2354858323931694,
      "num_tokens": 12912889.0,
      "step": 6985
    },
    {
      "entropy": 4.918697023391724,
      "epoch": 5.849372384937238,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024193974846900424,
      "loss": 4.3677,
      "mean_token_accuracy": 0.2442517638206482,
      "num_tokens": 12922454.0,
      "step": 6990
    },
    {
      "entropy": 4.917197513580322,
      "epoch": 5.853556485355648,
      "grad_norm": 1.375,
      "learning_rate": 0.00024162051945780967,
      "loss": 4.4043,
      "mean_token_accuracy": 0.24790869504213334,
      "num_tokens": 12931038.0,
      "step": 6995
    },
    {
      "entropy": 4.877459526062012,
      "epoch": 5.857740585774058,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002413013591362195,
      "loss": 4.4912,
      "mean_token_accuracy": 0.2366922229528427,
      "num_tokens": 12940830.0,
      "step": 7000
    },
    {
      "epoch": 5.857740585774058,
      "eval_entropy": 4.773090738947894,
      "eval_loss": 5.32881498336792,
      "eval_mean_token_accuracy": 0.19971691661634328,
      "eval_num_tokens": 12940830.0,
      "eval_runtime": 2.9364,
      "eval_samples_per_second": 1330.523,
      "eval_steps_per_second": 166.528,
      "step": 7000
    },
    {
      "entropy": 4.963376903533936,
      "epoch": 5.861924686192468,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024098226816101427,
      "loss": 4.4906,
      "mean_token_accuracy": 0.22440392076969146,
      "num_tokens": 12949159.0,
      "step": 7005
    },
    {
      "entropy": 5.03302903175354,
      "epoch": 5.866108786610878,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024066324718883205,
      "loss": 4.5078,
      "mean_token_accuracy": 0.23481764346361161,
      "num_tokens": 12957818.0,
      "step": 7010
    },
    {
      "entropy": 4.964997816085815,
      "epoch": 5.870292887029288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002403442968761666,
      "loss": 4.5263,
      "mean_token_accuracy": 0.2324890449643135,
      "num_tokens": 12966667.0,
      "step": 7015
    },
    {
      "entropy": 4.942241764068603,
      "epoch": 5.874476987447698,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024002541787936644,
      "loss": 4.3848,
      "mean_token_accuracy": 0.24017480611801148,
      "num_tokens": 12976081.0,
      "step": 7020
    },
    {
      "entropy": 4.917860078811645,
      "epoch": 5.878661087866108,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002397066108546332,
      "loss": 4.4965,
      "mean_token_accuracy": 0.23347352892160417,
      "num_tokens": 12985006.0,
      "step": 7025
    },
    {
      "entropy": 4.9558271884918215,
      "epoch": 5.882845188284519,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023938787645802056,
      "loss": 4.4278,
      "mean_token_accuracy": 0.24209991246461868,
      "num_tokens": 12993862.0,
      "step": 7030
    },
    {
      "entropy": 4.890793800354004,
      "epoch": 5.887029288702929,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023906921534543269,
      "loss": 4.4372,
      "mean_token_accuracy": 0.2417416214942932,
      "num_tokens": 13003637.0,
      "step": 7035
    },
    {
      "entropy": 4.9392311573028564,
      "epoch": 5.891213389121339,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023875062817262284,
      "loss": 4.4734,
      "mean_token_accuracy": 0.241826231777668,
      "num_tokens": 13013364.0,
      "step": 7040
    },
    {
      "entropy": 4.888514804840088,
      "epoch": 5.895397489539749,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00023843211559519228,
      "loss": 4.4373,
      "mean_token_accuracy": 0.23778630942106246,
      "num_tokens": 13023432.0,
      "step": 7045
    },
    {
      "entropy": 4.957732820510865,
      "epoch": 5.899581589958159,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023811367826858854,
      "loss": 4.453,
      "mean_token_accuracy": 0.23861077576875686,
      "num_tokens": 13034293.0,
      "step": 7050
    },
    {
      "entropy": 4.977032661437988,
      "epoch": 5.903765690376569,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023779531684810462,
      "loss": 4.4555,
      "mean_token_accuracy": 0.2363516718149185,
      "num_tokens": 13043985.0,
      "step": 7055
    },
    {
      "entropy": 4.858273363113403,
      "epoch": 5.907949790794979,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023747703198887698,
      "loss": 4.3016,
      "mean_token_accuracy": 0.2529794916510582,
      "num_tokens": 13052339.0,
      "step": 7060
    },
    {
      "entropy": 4.937607908248902,
      "epoch": 5.912133891213389,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023715882434588483,
      "loss": 4.3956,
      "mean_token_accuracy": 0.24051055014133454,
      "num_tokens": 13060858.0,
      "step": 7065
    },
    {
      "entropy": 4.923560953140258,
      "epoch": 5.916317991631799,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023684069457394826,
      "loss": 4.3672,
      "mean_token_accuracy": 0.2541592210531235,
      "num_tokens": 13069542.0,
      "step": 7070
    },
    {
      "entropy": 4.830236959457397,
      "epoch": 5.920502092050209,
      "grad_norm": 1.25,
      "learning_rate": 0.00023652264332772721,
      "loss": 4.3323,
      "mean_token_accuracy": 0.25167457163333895,
      "num_tokens": 13078868.0,
      "step": 7075
    },
    {
      "entropy": 4.962153005599975,
      "epoch": 5.924686192468619,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023620467126172001,
      "loss": 4.4966,
      "mean_token_accuracy": 0.23023190200328827,
      "num_tokens": 13088573.0,
      "step": 7080
    },
    {
      "entropy": 4.925450372695923,
      "epoch": 5.928870292887029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023588677903026202,
      "loss": 4.4084,
      "mean_token_accuracy": 0.23884457796812059,
      "num_tokens": 13097723.0,
      "step": 7085
    },
    {
      "entropy": 4.965341997146607,
      "epoch": 5.933054393305439,
      "grad_norm": 1.25,
      "learning_rate": 0.00023556896728752435,
      "loss": 4.3907,
      "mean_token_accuracy": 0.24108972400426865,
      "num_tokens": 13106071.0,
      "step": 7090
    },
    {
      "entropy": 4.879955053329468,
      "epoch": 5.937238493723849,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002352512366875125,
      "loss": 4.4096,
      "mean_token_accuracy": 0.24718322455883027,
      "num_tokens": 13115150.0,
      "step": 7095
    },
    {
      "entropy": 4.900448369979858,
      "epoch": 5.941422594142259,
      "grad_norm": 1.21875,
      "learning_rate": 0.000234933587884065,
      "loss": 4.478,
      "mean_token_accuracy": 0.23781720101833342,
      "num_tokens": 13124833.0,
      "step": 7100
    },
    {
      "entropy": 4.956347846984864,
      "epoch": 5.945606694560669,
      "grad_norm": 1.203125,
      "learning_rate": 0.000234616021530852,
      "loss": 4.5047,
      "mean_token_accuracy": 0.23291781693696975,
      "num_tokens": 13135751.0,
      "step": 7105
    },
    {
      "entropy": 4.895522975921631,
      "epoch": 5.949790794979079,
      "grad_norm": 1.375,
      "learning_rate": 0.00023429853828137393,
      "loss": 4.3437,
      "mean_token_accuracy": 0.2515662878751755,
      "num_tokens": 13144285.0,
      "step": 7110
    },
    {
      "entropy": 4.911907958984375,
      "epoch": 5.95397489539749,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023398113878896033,
      "loss": 4.4267,
      "mean_token_accuracy": 0.24330713748931884,
      "num_tokens": 13152856.0,
      "step": 7115
    },
    {
      "entropy": 4.971693134307861,
      "epoch": 5.9581589958159,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023366382370676824,
      "loss": 4.4786,
      "mean_token_accuracy": 0.23745597153902054,
      "num_tokens": 13162665.0,
      "step": 7120
    },
    {
      "entropy": 4.922029066085815,
      "epoch": 5.96234309623431,
      "grad_norm": 1.375,
      "learning_rate": 0.00023334659368778117,
      "loss": 4.3769,
      "mean_token_accuracy": 0.24553806632757186,
      "num_tokens": 13171110.0,
      "step": 7125
    },
    {
      "entropy": 4.9070416450500485,
      "epoch": 5.96652719665272,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023302944938480752,
      "loss": 4.4717,
      "mean_token_accuracy": 0.23965099602937698,
      "num_tokens": 13179948.0,
      "step": 7130
    },
    {
      "entropy": 4.989357614517212,
      "epoch": 5.97071129707113,
      "grad_norm": 1.21875,
      "learning_rate": 0.00023271239145047917,
      "loss": 4.4766,
      "mean_token_accuracy": 0.237283755838871,
      "num_tokens": 13190214.0,
      "step": 7135
    },
    {
      "entropy": 4.836311626434326,
      "epoch": 5.97489539748954,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023239542053725044,
      "loss": 4.3186,
      "mean_token_accuracy": 0.2568186715245247,
      "num_tokens": 13199048.0,
      "step": 7140
    },
    {
      "entropy": 4.899762487411499,
      "epoch": 5.97907949790795,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023207853729739663,
      "loss": 4.4156,
      "mean_token_accuracy": 0.23999714255332946,
      "num_tokens": 13208553.0,
      "step": 7145
    },
    {
      "entropy": 4.994880819320679,
      "epoch": 5.98326359832636,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023176174238301228,
      "loss": 4.5057,
      "mean_token_accuracy": 0.234708271920681,
      "num_tokens": 13218151.0,
      "step": 7150
    },
    {
      "entropy": 4.912987899780274,
      "epoch": 5.98744769874477,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002314450364460105,
      "loss": 4.4073,
      "mean_token_accuracy": 0.24769434183835984,
      "num_tokens": 13227730.0,
      "step": 7155
    },
    {
      "entropy": 4.914330959320068,
      "epoch": 5.99163179916318,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002311284201381212,
      "loss": 4.4975,
      "mean_token_accuracy": 0.23541647642850877,
      "num_tokens": 13237449.0,
      "step": 7160
    },
    {
      "entropy": 4.922647523880005,
      "epoch": 5.99581589958159,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023081189411088972,
      "loss": 4.4065,
      "mean_token_accuracy": 0.23772554397583007,
      "num_tokens": 13246662.0,
      "step": 7165
    },
    {
      "entropy": 4.976375961303711,
      "epoch": 6.0,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023049545901567586,
      "loss": 4.4593,
      "mean_token_accuracy": 0.23511881083250047,
      "num_tokens": 13255932.0,
      "step": 7170
    },
    {
      "entropy": 4.841125631332398,
      "epoch": 6.00418410041841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023017911550365233,
      "loss": 4.2093,
      "mean_token_accuracy": 0.26788610965013504,
      "num_tokens": 13264600.0,
      "step": 7175
    },
    {
      "entropy": 4.971485614776611,
      "epoch": 6.00836820083682,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022986286422580288,
      "loss": 4.3173,
      "mean_token_accuracy": 0.24957340806722642,
      "num_tokens": 13274261.0,
      "step": 7180
    },
    {
      "entropy": 4.955532264709473,
      "epoch": 6.01255230125523,
      "grad_norm": 1.21875,
      "learning_rate": 0.00022954670583292196,
      "loss": 4.251,
      "mean_token_accuracy": 0.2583720698952675,
      "num_tokens": 13284150.0,
      "step": 7185
    },
    {
      "entropy": 4.958441162109375,
      "epoch": 6.01673640167364,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00022923064097561265,
      "loss": 4.2352,
      "mean_token_accuracy": 0.24291809052228927,
      "num_tokens": 13293114.0,
      "step": 7190
    },
    {
      "entropy": 4.878188514709473,
      "epoch": 6.02092050209205,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002289146703042857,
      "loss": 4.2789,
      "mean_token_accuracy": 0.25481448173522947,
      "num_tokens": 13303391.0,
      "step": 7195
    },
    {
      "entropy": 4.952470397949218,
      "epoch": 6.02510460251046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022859879446915788,
      "loss": 4.3545,
      "mean_token_accuracy": 0.24760647416114806,
      "num_tokens": 13312372.0,
      "step": 7200
    },
    {
      "entropy": 5.003921747207642,
      "epoch": 6.02928870292887,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022828301412025094,
      "loss": 4.2906,
      "mean_token_accuracy": 0.25405008643865584,
      "num_tokens": 13321543.0,
      "step": 7205
    },
    {
      "entropy": 4.9234734058380125,
      "epoch": 6.03347280334728,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002279673299073901,
      "loss": 4.2376,
      "mean_token_accuracy": 0.2610643908381462,
      "num_tokens": 13330179.0,
      "step": 7210
    },
    {
      "entropy": 4.902516174316406,
      "epoch": 6.03765690376569,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022765174248020258,
      "loss": 4.2175,
      "mean_token_accuracy": 0.254955480992794,
      "num_tokens": 13339116.0,
      "step": 7215
    },
    {
      "entropy": 4.8294282913208,
      "epoch": 6.0418410041841,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022733625248811656,
      "loss": 4.2285,
      "mean_token_accuracy": 0.2605817899107933,
      "num_tokens": 13347908.0,
      "step": 7220
    },
    {
      "entropy": 4.823246097564697,
      "epoch": 6.046025104602511,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022702086058035985,
      "loss": 4.1923,
      "mean_token_accuracy": 0.26449614465236665,
      "num_tokens": 13356840.0,
      "step": 7225
    },
    {
      "entropy": 4.88328857421875,
      "epoch": 6.050209205020921,
      "grad_norm": 1.234375,
      "learning_rate": 0.00022670556740595827,
      "loss": 4.233,
      "mean_token_accuracy": 0.256473907828331,
      "num_tokens": 13365288.0,
      "step": 7230
    },
    {
      "entropy": 4.875168085098267,
      "epoch": 6.054393305439331,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022639037361373444,
      "loss": 4.2039,
      "mean_token_accuracy": 0.26240345388650893,
      "num_tokens": 13374612.0,
      "step": 7235
    },
    {
      "entropy": 4.890217065811157,
      "epoch": 6.058577405857741,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022607527985230648,
      "loss": 4.2422,
      "mean_token_accuracy": 0.25657849907875063,
      "num_tokens": 13384307.0,
      "step": 7240
    },
    {
      "entropy": 4.891270017623901,
      "epoch": 6.062761506276151,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022576028677008688,
      "loss": 4.2527,
      "mean_token_accuracy": 0.2515742689371109,
      "num_tokens": 13392626.0,
      "step": 7245
    },
    {
      "entropy": 4.917079925537109,
      "epoch": 6.066945606694561,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002254453950152804,
      "loss": 4.2483,
      "mean_token_accuracy": 0.25923923403024673,
      "num_tokens": 13401694.0,
      "step": 7250
    },
    {
      "entropy": 4.920544767379761,
      "epoch": 6.071129707112971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002251306052358839,
      "loss": 4.3033,
      "mean_token_accuracy": 0.2575088158249855,
      "num_tokens": 13410330.0,
      "step": 7255
    },
    {
      "entropy": 4.876627683639526,
      "epoch": 6.075313807531381,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022481591807968417,
      "loss": 4.1982,
      "mean_token_accuracy": 0.25978527814149854,
      "num_tokens": 13418988.0,
      "step": 7260
    },
    {
      "entropy": 4.9080901622772215,
      "epoch": 6.079497907949791,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002245013341942566,
      "loss": 4.3093,
      "mean_token_accuracy": 0.25117188692092896,
      "num_tokens": 13427487.0,
      "step": 7265
    },
    {
      "entropy": 4.921813106536865,
      "epoch": 6.083682008368201,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022418685422696444,
      "loss": 4.2509,
      "mean_token_accuracy": 0.2594044506549835,
      "num_tokens": 13436450.0,
      "step": 7270
    },
    {
      "entropy": 4.86575837135315,
      "epoch": 6.087866108786611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002238724788249568,
      "loss": 4.2176,
      "mean_token_accuracy": 0.2594246670603752,
      "num_tokens": 13445363.0,
      "step": 7275
    },
    {
      "entropy": 4.859816694259644,
      "epoch": 6.092050209205021,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022355820863516774,
      "loss": 4.1633,
      "mean_token_accuracy": 0.26392319798469543,
      "num_tokens": 13453960.0,
      "step": 7280
    },
    {
      "entropy": 4.928954696655273,
      "epoch": 6.096234309623431,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002232440443043147,
      "loss": 4.3169,
      "mean_token_accuracy": 0.2626212194561958,
      "num_tokens": 13463619.0,
      "step": 7285
    },
    {
      "entropy": 4.931155061721801,
      "epoch": 6.100418410041841,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002229299864788975,
      "loss": 4.2436,
      "mean_token_accuracy": 0.25904297530651094,
      "num_tokens": 13473566.0,
      "step": 7290
    },
    {
      "entropy": 4.864071559906006,
      "epoch": 6.104602510460251,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022261603580519656,
      "loss": 4.1923,
      "mean_token_accuracy": 0.2600919231772423,
      "num_tokens": 13481848.0,
      "step": 7295
    },
    {
      "entropy": 4.978577756881714,
      "epoch": 6.108786610878661,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022230219292927188,
      "loss": 4.3576,
      "mean_token_accuracy": 0.247738616168499,
      "num_tokens": 13491600.0,
      "step": 7300
    },
    {
      "entropy": 4.937346029281616,
      "epoch": 6.112970711297071,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022198845849696175,
      "loss": 4.3085,
      "mean_token_accuracy": 0.25170052498579026,
      "num_tokens": 13500888.0,
      "step": 7305
    },
    {
      "entropy": 4.900428056716919,
      "epoch": 6.117154811715481,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00022167483315388108,
      "loss": 4.2742,
      "mean_token_accuracy": 0.2601487383246422,
      "num_tokens": 13511031.0,
      "step": 7310
    },
    {
      "entropy": 4.874665927886963,
      "epoch": 6.121338912133891,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022136131754542048,
      "loss": 4.3026,
      "mean_token_accuracy": 0.2416843965649605,
      "num_tokens": 13520704.0,
      "step": 7315
    },
    {
      "entropy": 4.860784721374512,
      "epoch": 6.125523012552302,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022104791231674465,
      "loss": 4.1879,
      "mean_token_accuracy": 0.26054607331752777,
      "num_tokens": 13530297.0,
      "step": 7320
    },
    {
      "entropy": 4.918175554275512,
      "epoch": 6.129707112970712,
      "grad_norm": 1.375,
      "learning_rate": 0.00022073461811279106,
      "loss": 4.2219,
      "mean_token_accuracy": 0.2538872003555298,
      "num_tokens": 13538796.0,
      "step": 7325
    },
    {
      "entropy": 4.799028921127319,
      "epoch": 6.133891213389122,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022042143557826892,
      "loss": 4.1491,
      "mean_token_accuracy": 0.2669206440448761,
      "num_tokens": 13548145.0,
      "step": 7330
    },
    {
      "entropy": 4.783310842514038,
      "epoch": 6.138075313807532,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022010836535765749,
      "loss": 4.1777,
      "mean_token_accuracy": 0.25984172374010084,
      "num_tokens": 13557025.0,
      "step": 7335
    },
    {
      "entropy": 4.822976350784302,
      "epoch": 6.142259414225942,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021979540809520504,
      "loss": 4.2602,
      "mean_token_accuracy": 0.2523474544286728,
      "num_tokens": 13566485.0,
      "step": 7340
    },
    {
      "entropy": 4.9044952392578125,
      "epoch": 6.146443514644352,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002194825644349272,
      "loss": 4.271,
      "mean_token_accuracy": 0.2578278034925461,
      "num_tokens": 13575167.0,
      "step": 7345
    },
    {
      "entropy": 4.958434581756592,
      "epoch": 6.150627615062762,
      "grad_norm": 1.375,
      "learning_rate": 0.00021916983502060595,
      "loss": 4.2641,
      "mean_token_accuracy": 0.2530984923243523,
      "num_tokens": 13585757.0,
      "step": 7350
    },
    {
      "entropy": 4.889122486114502,
      "epoch": 6.154811715481172,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021885722049578817,
      "loss": 4.3111,
      "mean_token_accuracy": 0.24904566556215285,
      "num_tokens": 13595862.0,
      "step": 7355
    },
    {
      "entropy": 4.937966299057007,
      "epoch": 6.158995815899582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002185447215037842,
      "loss": 4.2829,
      "mean_token_accuracy": 0.25213314294815065,
      "num_tokens": 13605106.0,
      "step": 7360
    },
    {
      "entropy": 4.878959226608276,
      "epoch": 6.163179916317992,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021823233868766684,
      "loss": 4.2326,
      "mean_token_accuracy": 0.2576837927103043,
      "num_tokens": 13614317.0,
      "step": 7365
    },
    {
      "entropy": 4.847636461257935,
      "epoch": 6.167364016736402,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002179200726902697,
      "loss": 4.192,
      "mean_token_accuracy": 0.2635843068361282,
      "num_tokens": 13623196.0,
      "step": 7370
    },
    {
      "entropy": 4.824855041503906,
      "epoch": 6.171548117154812,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002176079241541859,
      "loss": 4.2286,
      "mean_token_accuracy": 0.25100663006305696,
      "num_tokens": 13631872.0,
      "step": 7375
    },
    {
      "entropy": 4.91606593132019,
      "epoch": 6.175732217573222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000217295893721767,
      "loss": 4.239,
      "mean_token_accuracy": 0.2546003073453903,
      "num_tokens": 13641240.0,
      "step": 7380
    },
    {
      "entropy": 4.846160268783569,
      "epoch": 6.179916317991632,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021698398203512148,
      "loss": 4.2139,
      "mean_token_accuracy": 0.25525360107421874,
      "num_tokens": 13650495.0,
      "step": 7385
    },
    {
      "entropy": 4.805061292648316,
      "epoch": 6.184100418410042,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021667218973611346,
      "loss": 4.1765,
      "mean_token_accuracy": 0.2681700840592384,
      "num_tokens": 13660234.0,
      "step": 7390
    },
    {
      "entropy": 4.9019921779632565,
      "epoch": 6.188284518828452,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00021636051746636126,
      "loss": 4.3262,
      "mean_token_accuracy": 0.2471445009112358,
      "num_tokens": 13670060.0,
      "step": 7395
    },
    {
      "entropy": 4.94047179222107,
      "epoch": 6.192468619246862,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021604896586723644,
      "loss": 4.2854,
      "mean_token_accuracy": 0.2581521555781364,
      "num_tokens": 13679551.0,
      "step": 7400
    },
    {
      "entropy": 4.941265392303467,
      "epoch": 6.196652719665272,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021573753557986198,
      "loss": 4.2844,
      "mean_token_accuracy": 0.2565193593502045,
      "num_tokens": 13688772.0,
      "step": 7405
    },
    {
      "entropy": 4.969020795822144,
      "epoch": 6.200836820083682,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021542622724511146,
      "loss": 4.3226,
      "mean_token_accuracy": 0.2486939400434494,
      "num_tokens": 13698110.0,
      "step": 7410
    },
    {
      "entropy": 4.972086954116821,
      "epoch": 6.205020920502092,
      "grad_norm": 1.375,
      "learning_rate": 0.00021511504150360724,
      "loss": 4.2489,
      "mean_token_accuracy": 0.2544397562742233,
      "num_tokens": 13707459.0,
      "step": 7415
    },
    {
      "entropy": 4.9269537925720215,
      "epoch": 6.209205020920502,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00021480397899571963,
      "loss": 4.3276,
      "mean_token_accuracy": 0.25087572634220123,
      "num_tokens": 13717596.0,
      "step": 7420
    },
    {
      "entropy": 4.819339656829834,
      "epoch": 6.2133891213389125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021449304036156525,
      "loss": 4.2228,
      "mean_token_accuracy": 0.2598550021648407,
      "num_tokens": 13726871.0,
      "step": 7425
    },
    {
      "entropy": 4.879504632949829,
      "epoch": 6.2175732217573225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021418222624100587,
      "loss": 4.2221,
      "mean_token_accuracy": 0.25467288941144944,
      "num_tokens": 13734960.0,
      "step": 7430
    },
    {
      "entropy": 4.924690389633179,
      "epoch": 6.2217573221757325,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021387153727364677,
      "loss": 4.2539,
      "mean_token_accuracy": 0.2617220789194107,
      "num_tokens": 13743771.0,
      "step": 7435
    },
    {
      "entropy": 4.876286411285401,
      "epoch": 6.2259414225941425,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002135609740988362,
      "loss": 4.2074,
      "mean_token_accuracy": 0.26186573356389997,
      "num_tokens": 13753080.0,
      "step": 7440
    },
    {
      "entropy": 4.91796760559082,
      "epoch": 6.2301255230125525,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002132505373556632,
      "loss": 4.321,
      "mean_token_accuracy": 0.25148753076791763,
      "num_tokens": 13762698.0,
      "step": 7445
    },
    {
      "entropy": 4.886522006988526,
      "epoch": 6.2343096234309625,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002129402276829564,
      "loss": 4.2226,
      "mean_token_accuracy": 0.2568728342652321,
      "num_tokens": 13771693.0,
      "step": 7450
    },
    {
      "entropy": 4.946479082107544,
      "epoch": 6.2384937238493725,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021263004571928345,
      "loss": 4.3331,
      "mean_token_accuracy": 0.25601724684238436,
      "num_tokens": 13780939.0,
      "step": 7455
    },
    {
      "entropy": 4.850085353851318,
      "epoch": 6.2426778242677825,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021231999210294882,
      "loss": 4.1897,
      "mean_token_accuracy": 0.2645399823784828,
      "num_tokens": 13790001.0,
      "step": 7460
    },
    {
      "entropy": 4.965839576721192,
      "epoch": 6.2468619246861925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021201006747199313,
      "loss": 4.2783,
      "mean_token_accuracy": 0.26187181621789934,
      "num_tokens": 13799360.0,
      "step": 7465
    },
    {
      "entropy": 4.867313385009766,
      "epoch": 6.2510460251046025,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021170027246419137,
      "loss": 4.2492,
      "mean_token_accuracy": 0.25460500568151473,
      "num_tokens": 13808518.0,
      "step": 7470
    },
    {
      "entropy": 4.9593085765838625,
      "epoch": 6.2552301255230125,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021139060771705205,
      "loss": 4.3373,
      "mean_token_accuracy": 0.24999587684869767,
      "num_tokens": 13818144.0,
      "step": 7475
    },
    {
      "entropy": 4.900734806060791,
      "epoch": 6.2594142259414225,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021108107386781517,
      "loss": 4.2427,
      "mean_token_accuracy": 0.2624722018837929,
      "num_tokens": 13828063.0,
      "step": 7480
    },
    {
      "entropy": 4.8111542701721195,
      "epoch": 6.2635983263598325,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021077167155345166,
      "loss": 4.2117,
      "mean_token_accuracy": 0.2537585452198982,
      "num_tokens": 13837014.0,
      "step": 7485
    },
    {
      "entropy": 4.856605434417725,
      "epoch": 6.2677824267782425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002104624014106618,
      "loss": 4.2986,
      "mean_token_accuracy": 0.25745722055435183,
      "num_tokens": 13845518.0,
      "step": 7490
    },
    {
      "entropy": 4.916968917846679,
      "epoch": 6.2719665271966525,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021015326407587389,
      "loss": 4.2986,
      "mean_token_accuracy": 0.25087255537509917,
      "num_tokens": 13854154.0,
      "step": 7495
    },
    {
      "entropy": 4.8995709896087645,
      "epoch": 6.2761506276150625,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020984426018524278,
      "loss": 4.284,
      "mean_token_accuracy": 0.2628625437617302,
      "num_tokens": 13863404.0,
      "step": 7500
    },
    {
      "epoch": 6.2761506276150625,
      "eval_entropy": 4.737074861741017,
      "eval_loss": 5.3355712890625,
      "eval_mean_token_accuracy": 0.2007908179198062,
      "eval_num_tokens": 13863404.0,
      "eval_runtime": 2.7354,
      "eval_samples_per_second": 1428.297,
      "eval_steps_per_second": 178.766,
      "step": 7500
    },
    {
      "entropy": 4.792803335189819,
      "epoch": 6.2803347280334725,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020953539037464882,
      "loss": 4.1516,
      "mean_token_accuracy": 0.26716977953910825,
      "num_tokens": 13872554.0,
      "step": 7505
    },
    {
      "entropy": 4.8783057689666744,
      "epoch": 6.2845188284518825,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020922665527969646,
      "loss": 4.2571,
      "mean_token_accuracy": 0.259717258810997,
      "num_tokens": 13881740.0,
      "step": 7510
    },
    {
      "entropy": 4.961845588684082,
      "epoch": 6.2887029288702925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002089180555357128,
      "loss": 4.3072,
      "mean_token_accuracy": 0.2604274481534958,
      "num_tokens": 13890114.0,
      "step": 7515
    },
    {
      "entropy": 4.939940404891968,
      "epoch": 6.292887029288703,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020860959177774652,
      "loss": 4.3876,
      "mean_token_accuracy": 0.23963457643985747,
      "num_tokens": 13899744.0,
      "step": 7520
    },
    {
      "entropy": 4.974252510070801,
      "epoch": 6.297071129707113,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020830126464056647,
      "loss": 4.2706,
      "mean_token_accuracy": 0.25403507202863695,
      "num_tokens": 13909011.0,
      "step": 7525
    },
    {
      "entropy": 4.925878190994263,
      "epoch": 6.301255230125523,
      "grad_norm": 1.375,
      "learning_rate": 0.00020799307475866035,
      "loss": 4.2696,
      "mean_token_accuracy": 0.25529755353927613,
      "num_tokens": 13918435.0,
      "step": 7530
    },
    {
      "entropy": 4.890402555465698,
      "epoch": 6.305439330543933,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020768502276623336,
      "loss": 4.2888,
      "mean_token_accuracy": 0.2550456926226616,
      "num_tokens": 13927164.0,
      "step": 7535
    },
    {
      "entropy": 4.86189866065979,
      "epoch": 6.309623430962343,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020737710929720694,
      "loss": 4.181,
      "mean_token_accuracy": 0.2679505005478859,
      "num_tokens": 13935694.0,
      "step": 7540
    },
    {
      "entropy": 4.963479471206665,
      "epoch": 6.313807531380753,
      "grad_norm": 1.25,
      "learning_rate": 0.00020706933498521767,
      "loss": 4.3601,
      "mean_token_accuracy": 0.24827947467565536,
      "num_tokens": 13946068.0,
      "step": 7545
    },
    {
      "entropy": 4.963890075683594,
      "epoch": 6.317991631799163,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002067617004636153,
      "loss": 4.3331,
      "mean_token_accuracy": 0.2518668591976166,
      "num_tokens": 13955238.0,
      "step": 7550
    },
    {
      "entropy": 4.889068603515625,
      "epoch": 6.322175732217573,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020645420636546246,
      "loss": 4.2255,
      "mean_token_accuracy": 0.26399461925029755,
      "num_tokens": 13964251.0,
      "step": 7555
    },
    {
      "entropy": 4.95390133857727,
      "epoch": 6.326359832635983,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020614685332353246,
      "loss": 4.3849,
      "mean_token_accuracy": 0.2399700850248337,
      "num_tokens": 13974111.0,
      "step": 7560
    },
    {
      "entropy": 4.899079370498657,
      "epoch": 6.330543933054393,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020583964197030852,
      "loss": 4.2271,
      "mean_token_accuracy": 0.25263774394989014,
      "num_tokens": 13982842.0,
      "step": 7565
    },
    {
      "entropy": 4.804543161392212,
      "epoch": 6.334728033472803,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020553257293798216,
      "loss": 4.1947,
      "mean_token_accuracy": 0.2614595413208008,
      "num_tokens": 13992010.0,
      "step": 7570
    },
    {
      "entropy": 4.869327354431152,
      "epoch": 6.338912133891213,
      "grad_norm": 1.375,
      "learning_rate": 0.00020522564685845207,
      "loss": 4.1658,
      "mean_token_accuracy": 0.2684245839715004,
      "num_tokens": 14000746.0,
      "step": 7575
    },
    {
      "entropy": 4.907274484634399,
      "epoch": 6.343096234309623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020491886436332287,
      "loss": 4.2491,
      "mean_token_accuracy": 0.2591560587286949,
      "num_tokens": 14010664.0,
      "step": 7580
    },
    {
      "entropy": 4.918039560317993,
      "epoch": 6.347280334728033,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020461222608390346,
      "loss": 4.2875,
      "mean_token_accuracy": 0.24419411420822143,
      "num_tokens": 14019473.0,
      "step": 7585
    },
    {
      "entropy": 4.874756765365601,
      "epoch": 6.351464435146443,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020430573265120623,
      "loss": 4.2786,
      "mean_token_accuracy": 0.25727693140506747,
      "num_tokens": 14027971.0,
      "step": 7590
    },
    {
      "entropy": 4.795595502853393,
      "epoch": 6.355648535564853,
      "grad_norm": 1.375,
      "learning_rate": 0.0002039993846959453,
      "loss": 4.1095,
      "mean_token_accuracy": 0.26757019609212873,
      "num_tokens": 14036972.0,
      "step": 7595
    },
    {
      "entropy": 4.8994526863098145,
      "epoch": 6.359832635983263,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020369318284853562,
      "loss": 4.2163,
      "mean_token_accuracy": 0.24928725957870485,
      "num_tokens": 14046486.0,
      "step": 7600
    },
    {
      "entropy": 4.846242523193359,
      "epoch": 6.364016736401673,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020338712773909125,
      "loss": 4.2373,
      "mean_token_accuracy": 0.2593563675880432,
      "num_tokens": 14056398.0,
      "step": 7605
    },
    {
      "entropy": 5.013672161102295,
      "epoch": 6.368200836820083,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020308121999742458,
      "loss": 4.3775,
      "mean_token_accuracy": 0.24053625911474227,
      "num_tokens": 14065517.0,
      "step": 7610
    },
    {
      "entropy": 4.9137287616729735,
      "epoch": 6.372384937238493,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020277546025304432,
      "loss": 4.267,
      "mean_token_accuracy": 0.2573018029332161,
      "num_tokens": 14074563.0,
      "step": 7615
    },
    {
      "entropy": 4.882937002182007,
      "epoch": 6.376569037656903,
      "grad_norm": 1.25,
      "learning_rate": 0.00020246984913515503,
      "loss": 4.3021,
      "mean_token_accuracy": 0.257138654589653,
      "num_tokens": 14083125.0,
      "step": 7620
    },
    {
      "entropy": 4.851533699035644,
      "epoch": 6.380753138075314,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020216438727265517,
      "loss": 4.2393,
      "mean_token_accuracy": 0.257677298784256,
      "num_tokens": 14091965.0,
      "step": 7625
    },
    {
      "entropy": 4.824789428710938,
      "epoch": 6.384937238493724,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020185907529413622,
      "loss": 4.2229,
      "mean_token_accuracy": 0.25763061493635175,
      "num_tokens": 14100872.0,
      "step": 7630
    },
    {
      "entropy": 4.926395320892334,
      "epoch": 6.389121338912134,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020155391382788108,
      "loss": 4.2844,
      "mean_token_accuracy": 0.25538465678691863,
      "num_tokens": 14109860.0,
      "step": 7635
    },
    {
      "entropy": 4.95124773979187,
      "epoch": 6.393305439330544,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020124890350186304,
      "loss": 4.3157,
      "mean_token_accuracy": 0.25554013848304746,
      "num_tokens": 14119256.0,
      "step": 7640
    },
    {
      "entropy": 5.003087949752808,
      "epoch": 6.397489539748954,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002009440449437443,
      "loss": 4.4052,
      "mean_token_accuracy": 0.2463753789663315,
      "num_tokens": 14128160.0,
      "step": 7645
    },
    {
      "entropy": 4.947859477996826,
      "epoch": 6.401673640167364,
      "grad_norm": 1.375,
      "learning_rate": 0.0002006393387808747,
      "loss": 4.3393,
      "mean_token_accuracy": 0.2513251855969429,
      "num_tokens": 14137810.0,
      "step": 7650
    },
    {
      "entropy": 4.9075024127960205,
      "epoch": 6.405857740585774,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020033478564029063,
      "loss": 4.1628,
      "mean_token_accuracy": 0.26644574850797653,
      "num_tokens": 14145914.0,
      "step": 7655
    },
    {
      "entropy": 4.98628306388855,
      "epoch": 6.410041841004184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002000303861487134,
      "loss": 4.3537,
      "mean_token_accuracy": 0.2442624494433403,
      "num_tokens": 14156348.0,
      "step": 7660
    },
    {
      "entropy": 5.018626356124878,
      "epoch": 6.414225941422594,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019972614093254822,
      "loss": 4.4068,
      "mean_token_accuracy": 0.24657621681690217,
      "num_tokens": 14166079.0,
      "step": 7665
    },
    {
      "entropy": 4.873981618881226,
      "epoch": 6.418410041841004,
      "grad_norm": 1.21875,
      "learning_rate": 0.00019942205061788292,
      "loss": 4.2567,
      "mean_token_accuracy": 0.25728431195020673,
      "num_tokens": 14175732.0,
      "step": 7670
    },
    {
      "entropy": 4.914537715911865,
      "epoch": 6.422594142259414,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019911811583048645,
      "loss": 4.2908,
      "mean_token_accuracy": 0.25591263473033904,
      "num_tokens": 14184935.0,
      "step": 7675
    },
    {
      "entropy": 4.926764917373657,
      "epoch": 6.426778242677824,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019881433719580766,
      "loss": 4.3522,
      "mean_token_accuracy": 0.24484024345874786,
      "num_tokens": 14193693.0,
      "step": 7680
    },
    {
      "entropy": 4.854112815856934,
      "epoch": 6.430962343096234,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019851071533897425,
      "loss": 4.2679,
      "mean_token_accuracy": 0.2602280467748642,
      "num_tokens": 14203323.0,
      "step": 7685
    },
    {
      "entropy": 4.886469888687134,
      "epoch": 6.435146443514644,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019820725088479107,
      "loss": 4.2127,
      "mean_token_accuracy": 0.25545543134212495,
      "num_tokens": 14211992.0,
      "step": 7690
    },
    {
      "entropy": 4.888636732101441,
      "epoch": 6.439330543933054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001979039444577393,
      "loss": 4.256,
      "mean_token_accuracy": 0.2579985737800598,
      "num_tokens": 14221578.0,
      "step": 7695
    },
    {
      "entropy": 4.915421199798584,
      "epoch": 6.443514644351464,
      "grad_norm": 1.375,
      "learning_rate": 0.00019760079668197474,
      "loss": 4.291,
      "mean_token_accuracy": 0.2550632029771805,
      "num_tokens": 14231411.0,
      "step": 7700
    },
    {
      "entropy": 4.885698413848877,
      "epoch": 6.447698744769874,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019729780818132664,
      "loss": 4.2631,
      "mean_token_accuracy": 0.25803789496421814,
      "num_tokens": 14241317.0,
      "step": 7705
    },
    {
      "entropy": 4.925375127792359,
      "epoch": 6.451882845188284,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019699497957929702,
      "loss": 4.3083,
      "mean_token_accuracy": 0.25833956748247144,
      "num_tokens": 14251248.0,
      "step": 7710
    },
    {
      "entropy": 4.881535530090332,
      "epoch": 6.456066945606695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019669231149905804,
      "loss": 4.2464,
      "mean_token_accuracy": 0.2501637414097786,
      "num_tokens": 14261487.0,
      "step": 7715
    },
    {
      "entropy": 4.890447282791138,
      "epoch": 6.460251046025105,
      "grad_norm": 1.375,
      "learning_rate": 0.00019638980456345207,
      "loss": 4.3151,
      "mean_token_accuracy": 0.24139856547117233,
      "num_tokens": 14271438.0,
      "step": 7720
    },
    {
      "entropy": 4.909246158599854,
      "epoch": 6.464435146443515,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019608745939498983,
      "loss": 4.2906,
      "mean_token_accuracy": 0.250335656106472,
      "num_tokens": 14280683.0,
      "step": 7725
    },
    {
      "entropy": 4.840122413635254,
      "epoch": 6.468619246861925,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019578527661584898,
      "loss": 4.3316,
      "mean_token_accuracy": 0.2528389900922775,
      "num_tokens": 14290041.0,
      "step": 7730
    },
    {
      "entropy": 4.911505270004272,
      "epoch": 6.472803347280335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019548325684787311,
      "loss": 4.3166,
      "mean_token_accuracy": 0.243887822329998,
      "num_tokens": 14298914.0,
      "step": 7735
    },
    {
      "entropy": 4.897787427902221,
      "epoch": 6.476987447698745,
      "grad_norm": 1.375,
      "learning_rate": 0.0001951814007125704,
      "loss": 4.2675,
      "mean_token_accuracy": 0.24346271604299546,
      "num_tokens": 14308347.0,
      "step": 7740
    },
    {
      "entropy": 4.849288415908814,
      "epoch": 6.481171548117155,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019487970883111234,
      "loss": 4.2626,
      "mean_token_accuracy": 0.2525020852684975,
      "num_tokens": 14317839.0,
      "step": 7745
    },
    {
      "entropy": 4.852408123016358,
      "epoch": 6.485355648535565,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019457818182433206,
      "loss": 4.2921,
      "mean_token_accuracy": 0.24868250787258148,
      "num_tokens": 14327037.0,
      "step": 7750
    },
    {
      "entropy": 4.8763762474060055,
      "epoch": 6.489539748953975,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019427682031272382,
      "loss": 4.2608,
      "mean_token_accuracy": 0.2586748331785202,
      "num_tokens": 14337417.0,
      "step": 7755
    },
    {
      "entropy": 5.033905982971191,
      "epoch": 6.493723849372385,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019397562491644112,
      "loss": 4.3779,
      "mean_token_accuracy": 0.24872520714998245,
      "num_tokens": 14347037.0,
      "step": 7760
    },
    {
      "entropy": 4.923737668991089,
      "epoch": 6.497907949790795,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019367459625529564,
      "loss": 4.3366,
      "mean_token_accuracy": 0.24386311024427415,
      "num_tokens": 14356509.0,
      "step": 7765
    },
    {
      "entropy": 4.89388861656189,
      "epoch": 6.502092050209205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019337373494875604,
      "loss": 4.2053,
      "mean_token_accuracy": 0.2583172768354416,
      "num_tokens": 14365276.0,
      "step": 7770
    },
    {
      "entropy": 4.9482402324676515,
      "epoch": 6.506276150627615,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019307304161594657,
      "loss": 4.3033,
      "mean_token_accuracy": 0.25110249519348143,
      "num_tokens": 14373546.0,
      "step": 7775
    },
    {
      "entropy": 4.894652605056763,
      "epoch": 6.510460251046025,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019277251687564557,
      "loss": 4.2655,
      "mean_token_accuracy": 0.2555797412991524,
      "num_tokens": 14382767.0,
      "step": 7780
    },
    {
      "entropy": 4.837567901611328,
      "epoch": 6.514644351464435,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019247216134628476,
      "loss": 4.2601,
      "mean_token_accuracy": 0.25367254614830015,
      "num_tokens": 14392602.0,
      "step": 7785
    },
    {
      "entropy": 4.864850854873657,
      "epoch": 6.518828451882845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019217197564594746,
      "loss": 4.3501,
      "mean_token_accuracy": 0.24254341721534728,
      "num_tokens": 14402191.0,
      "step": 7790
    },
    {
      "entropy": 4.944798946380615,
      "epoch": 6.523012552301255,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019187196039236772,
      "loss": 4.2833,
      "mean_token_accuracy": 0.25113302022218703,
      "num_tokens": 14411116.0,
      "step": 7795
    },
    {
      "entropy": 4.824346351623535,
      "epoch": 6.527196652719665,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019157211620292853,
      "loss": 4.2163,
      "mean_token_accuracy": 0.2582664847373962,
      "num_tokens": 14419725.0,
      "step": 7800
    },
    {
      "entropy": 4.922324323654175,
      "epoch": 6.531380753138075,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019127244369466118,
      "loss": 4.3606,
      "mean_token_accuracy": 0.2509873121976852,
      "num_tokens": 14429904.0,
      "step": 7805
    },
    {
      "entropy": 4.87157678604126,
      "epoch": 6.535564853556485,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019097294348424348,
      "loss": 4.2379,
      "mean_token_accuracy": 0.25241530537605283,
      "num_tokens": 14439384.0,
      "step": 7810
    },
    {
      "entropy": 4.905696678161621,
      "epoch": 6.539748953974895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019067361618799856,
      "loss": 4.2207,
      "mean_token_accuracy": 0.2625962272286415,
      "num_tokens": 14448002.0,
      "step": 7815
    },
    {
      "entropy": 4.891553163528442,
      "epoch": 6.543933054393305,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019037446242189398,
      "loss": 4.2611,
      "mean_token_accuracy": 0.2507487043738365,
      "num_tokens": 14456887.0,
      "step": 7820
    },
    {
      "entropy": 4.9406877040863035,
      "epoch": 6.548117154811716,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019007548280154007,
      "loss": 4.4125,
      "mean_token_accuracy": 0.24211733043193817,
      "num_tokens": 14466615.0,
      "step": 7825
    },
    {
      "entropy": 4.875655651092529,
      "epoch": 6.552301255230126,
      "grad_norm": 1.21875,
      "learning_rate": 0.00018977667794218894,
      "loss": 4.2222,
      "mean_token_accuracy": 0.25692217200994494,
      "num_tokens": 14476684.0,
      "step": 7830
    },
    {
      "entropy": 4.8763219833374025,
      "epoch": 6.556485355648536,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001894780484587328,
      "loss": 4.233,
      "mean_token_accuracy": 0.2553003191947937,
      "num_tokens": 14485364.0,
      "step": 7835
    },
    {
      "entropy": 4.86587347984314,
      "epoch": 6.560669456066946,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018917959496570318,
      "loss": 4.3151,
      "mean_token_accuracy": 0.2534299612045288,
      "num_tokens": 14495099.0,
      "step": 7840
    },
    {
      "entropy": 4.916420793533325,
      "epoch": 6.564853556485356,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001888813180772694,
      "loss": 4.2664,
      "mean_token_accuracy": 0.2574158802628517,
      "num_tokens": 14504051.0,
      "step": 7845
    },
    {
      "entropy": 4.911519622802734,
      "epoch": 6.569037656903766,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018858321840723724,
      "loss": 4.2507,
      "mean_token_accuracy": 0.24954316020011902,
      "num_tokens": 14513517.0,
      "step": 7850
    },
    {
      "entropy": 4.812881946563721,
      "epoch": 6.573221757322176,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018828529656904802,
      "loss": 4.3125,
      "mean_token_accuracy": 0.25231822580099106,
      "num_tokens": 14522504.0,
      "step": 7855
    },
    {
      "entropy": 4.925061845779419,
      "epoch": 6.577405857740586,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018798755317577691,
      "loss": 4.3048,
      "mean_token_accuracy": 0.25611866265535355,
      "num_tokens": 14532215.0,
      "step": 7860
    },
    {
      "entropy": 4.849268913269043,
      "epoch": 6.581589958158996,
      "grad_norm": 1.375,
      "learning_rate": 0.000187689988840132,
      "loss": 4.1521,
      "mean_token_accuracy": 0.272016441822052,
      "num_tokens": 14540489.0,
      "step": 7865
    },
    {
      "entropy": 4.87277626991272,
      "epoch": 6.585774058577406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001873926041744528,
      "loss": 4.2628,
      "mean_token_accuracy": 0.25840317010879515,
      "num_tokens": 14550476.0,
      "step": 7870
    },
    {
      "entropy": 4.932324552536011,
      "epoch": 6.589958158995816,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018709539979070922,
      "loss": 4.2212,
      "mean_token_accuracy": 0.26221890896558764,
      "num_tokens": 14559349.0,
      "step": 7875
    },
    {
      "entropy": 4.873563814163208,
      "epoch": 6.594142259414226,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00018679837630049994,
      "loss": 4.2729,
      "mean_token_accuracy": 0.25966800898313525,
      "num_tokens": 14568480.0,
      "step": 7880
    },
    {
      "entropy": 4.8944169044494625,
      "epoch": 6.598326359832636,
      "grad_norm": 1.265625,
      "learning_rate": 0.00018650153431505167,
      "loss": 4.2559,
      "mean_token_accuracy": 0.26460086554288864,
      "num_tokens": 14578071.0,
      "step": 7885
    },
    {
      "entropy": 4.888390922546387,
      "epoch": 6.602510460251046,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001862048744452175,
      "loss": 4.2635,
      "mean_token_accuracy": 0.2594033434987068,
      "num_tokens": 14586211.0,
      "step": 7890
    },
    {
      "entropy": 4.8708961486816404,
      "epoch": 6.606694560669456,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001859083973014757,
      "loss": 4.2862,
      "mean_token_accuracy": 0.2583386659622192,
      "num_tokens": 14595206.0,
      "step": 7895
    },
    {
      "entropy": 4.890885591506958,
      "epoch": 6.610878661087866,
      "grad_norm": 1.4375,
      "learning_rate": 0.00018561210349392855,
      "loss": 4.2627,
      "mean_token_accuracy": 0.2522640347480774,
      "num_tokens": 14604569.0,
      "step": 7900
    },
    {
      "entropy": 4.91824893951416,
      "epoch": 6.615062761506276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001853159936323011,
      "loss": 4.2801,
      "mean_token_accuracy": 0.2558894708752632,
      "num_tokens": 14614040.0,
      "step": 7905
    },
    {
      "entropy": 4.8941497802734375,
      "epoch": 6.619246861924686,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018502006832593992,
      "loss": 4.2718,
      "mean_token_accuracy": 0.25785543769598007,
      "num_tokens": 14624461.0,
      "step": 7910
    },
    {
      "entropy": 4.876320171356201,
      "epoch": 6.623430962343097,
      "grad_norm": 1.296875,
      "learning_rate": 0.00018472432818381158,
      "loss": 4.3122,
      "mean_token_accuracy": 0.2537187024950981,
      "num_tokens": 14633417.0,
      "step": 7915
    },
    {
      "entropy": 4.875927829742432,
      "epoch": 6.627615062761507,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018442877381450175,
      "loss": 4.2025,
      "mean_token_accuracy": 0.2627200722694397,
      "num_tokens": 14642387.0,
      "step": 7920
    },
    {
      "entropy": 4.857219886779785,
      "epoch": 6.631799163179917,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001841334058262139,
      "loss": 4.2071,
      "mean_token_accuracy": 0.2628923267126083,
      "num_tokens": 14650613.0,
      "step": 7925
    },
    {
      "entropy": 4.836830806732178,
      "epoch": 6.635983263598327,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00018383822482676777,
      "loss": 4.3205,
      "mean_token_accuracy": 0.2607922062277794,
      "num_tokens": 14659411.0,
      "step": 7930
    },
    {
      "entropy": 4.817849159240723,
      "epoch": 6.640167364016737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00018354323142359848,
      "loss": 4.2439,
      "mean_token_accuracy": 0.2582366704940796,
      "num_tokens": 14668375.0,
      "step": 7935
    },
    {
      "entropy": 4.873711824417114,
      "epoch": 6.644351464435147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018324842622375497,
      "loss": 4.2572,
      "mean_token_accuracy": 0.2655599147081375,
      "num_tokens": 14677608.0,
      "step": 7940
    },
    {
      "entropy": 4.928869104385376,
      "epoch": 6.648535564853557,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018295380983389905,
      "loss": 4.296,
      "mean_token_accuracy": 0.2580977395176888,
      "num_tokens": 14686646.0,
      "step": 7945
    },
    {
      "entropy": 4.810101652145386,
      "epoch": 6.652719665271967,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001826593828603037,
      "loss": 4.2473,
      "mean_token_accuracy": 0.25163774341344836,
      "num_tokens": 14695703.0,
      "step": 7950
    },
    {
      "entropy": 4.863131141662597,
      "epoch": 6.656903765690377,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001823651459088524,
      "loss": 4.2403,
      "mean_token_accuracy": 0.25747397989034654,
      "num_tokens": 14704493.0,
      "step": 7955
    },
    {
      "entropy": 4.946060705184936,
      "epoch": 6.661087866108787,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00018207109958503747,
      "loss": 4.3099,
      "mean_token_accuracy": 0.25769668221473696,
      "num_tokens": 14713119.0,
      "step": 7960
    },
    {
      "entropy": 4.9195647716522215,
      "epoch": 6.665271966527197,
      "grad_norm": 1.390625,
      "learning_rate": 0.000181777244493959,
      "loss": 4.2991,
      "mean_token_accuracy": 0.2529951870441437,
      "num_tokens": 14722360.0,
      "step": 7965
    },
    {
      "entropy": 4.911259269714355,
      "epoch": 6.669456066945607,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001814835812403235,
      "loss": 4.2857,
      "mean_token_accuracy": 0.2531762793660164,
      "num_tokens": 14732066.0,
      "step": 7970
    },
    {
      "entropy": 4.845459127426148,
      "epoch": 6.673640167364017,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00018119011042844286,
      "loss": 4.2415,
      "mean_token_accuracy": 0.24859005212783813,
      "num_tokens": 14740968.0,
      "step": 7975
    },
    {
      "entropy": 4.863522624969482,
      "epoch": 6.677824267782427,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001808968326622325,
      "loss": 4.2723,
      "mean_token_accuracy": 0.2472422704100609,
      "num_tokens": 14750574.0,
      "step": 7980
    },
    {
      "entropy": 4.910039472579956,
      "epoch": 6.682008368200837,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018060374854521116,
      "loss": 4.2465,
      "mean_token_accuracy": 0.2479236379265785,
      "num_tokens": 14760064.0,
      "step": 7985
    },
    {
      "entropy": 4.865352010726928,
      "epoch": 6.686192468619247,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018031085868049884,
      "loss": 4.3096,
      "mean_token_accuracy": 0.2554003268480301,
      "num_tokens": 14769721.0,
      "step": 7990
    },
    {
      "entropy": 4.908555698394776,
      "epoch": 6.690376569037657,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001800181636708157,
      "loss": 4.3174,
      "mean_token_accuracy": 0.25341242700815203,
      "num_tokens": 14778895.0,
      "step": 7995
    },
    {
      "entropy": 4.894577598571777,
      "epoch": 6.694560669456067,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017972566411848112,
      "loss": 4.2822,
      "mean_token_accuracy": 0.2502142354846001,
      "num_tokens": 14787541.0,
      "step": 8000
    },
    {
      "epoch": 6.694560669456067,
      "eval_entropy": 4.78593703964249,
      "eval_loss": 5.320886611938477,
      "eval_mean_token_accuracy": 0.20199173917433966,
      "eval_num_tokens": 14787541.0,
      "eval_runtime": 2.7325,
      "eval_samples_per_second": 1429.821,
      "eval_steps_per_second": 178.956,
      "step": 8000
    },
    {
      "entropy": 4.921368503570557,
      "epoch": 6.698744769874477,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000179433360625412,
      "loss": 4.2627,
      "mean_token_accuracy": 0.26033646911382674,
      "num_tokens": 14795669.0,
      "step": 8005
    },
    {
      "entropy": 4.900566053390503,
      "epoch": 6.702928870292887,
      "grad_norm": 1.25,
      "learning_rate": 0.00017914125379312208,
      "loss": 4.243,
      "mean_token_accuracy": 0.25711684226989745,
      "num_tokens": 14804587.0,
      "step": 8010
    },
    {
      "entropy": 4.872041273117065,
      "epoch": 6.707112970711297,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017884934422272026,
      "loss": 4.24,
      "mean_token_accuracy": 0.25765829533338547,
      "num_tokens": 14813746.0,
      "step": 8015
    },
    {
      "entropy": 4.879439067840576,
      "epoch": 6.711297071129707,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017855763251490924,
      "loss": 4.3132,
      "mean_token_accuracy": 0.24586979448795318,
      "num_tokens": 14822906.0,
      "step": 8020
    },
    {
      "entropy": 4.892772102355957,
      "epoch": 6.7154811715481175,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017826611926998514,
      "loss": 4.2807,
      "mean_token_accuracy": 0.25640886574983596,
      "num_tokens": 14832212.0,
      "step": 8025
    },
    {
      "entropy": 4.911296272277832,
      "epoch": 6.7196652719665275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001779748050878352,
      "loss": 4.2683,
      "mean_token_accuracy": 0.25317064076662066,
      "num_tokens": 14841381.0,
      "step": 8030
    },
    {
      "entropy": 4.775091361999512,
      "epoch": 6.7238493723849375,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017768369056793727,
      "loss": 4.1561,
      "mean_token_accuracy": 0.2628245249390602,
      "num_tokens": 14850090.0,
      "step": 8035
    },
    {
      "entropy": 4.85731782913208,
      "epoch": 6.7280334728033475,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000177392776309358,
      "loss": 4.2649,
      "mean_token_accuracy": 0.25863724648952485,
      "num_tokens": 14858885.0,
      "step": 8040
    },
    {
      "entropy": 4.8620813369750975,
      "epoch": 6.7322175732217575,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017710206291075248,
      "loss": 4.2541,
      "mean_token_accuracy": 0.2523441717028618,
      "num_tokens": 14868173.0,
      "step": 8045
    },
    {
      "entropy": 4.875956726074219,
      "epoch": 6.7364016736401675,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017681155097036194,
      "loss": 4.2597,
      "mean_token_accuracy": 0.2522418096661568,
      "num_tokens": 14877164.0,
      "step": 8050
    },
    {
      "entropy": 4.898303508758545,
      "epoch": 6.7405857740585775,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001765212410860132,
      "loss": 4.2928,
      "mean_token_accuracy": 0.25798328071832655,
      "num_tokens": 14886285.0,
      "step": 8055
    },
    {
      "entropy": 4.792060613632202,
      "epoch": 6.7447698744769875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00017623113385511753,
      "loss": 4.2568,
      "mean_token_accuracy": 0.24958813637495042,
      "num_tokens": 14895079.0,
      "step": 8060
    },
    {
      "entropy": 4.869246101379394,
      "epoch": 6.7489539748953975,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001759412298746687,
      "loss": 4.2647,
      "mean_token_accuracy": 0.2500820249319077,
      "num_tokens": 14904418.0,
      "step": 8065
    },
    {
      "entropy": 4.867325735092163,
      "epoch": 6.7531380753138075,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017565152974124276,
      "loss": 4.3103,
      "mean_token_accuracy": 0.2455492228269577,
      "num_tokens": 14913621.0,
      "step": 8070
    },
    {
      "entropy": 4.83552508354187,
      "epoch": 6.7573221757322175,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001753620340509959,
      "loss": 4.2457,
      "mean_token_accuracy": 0.2618093118071556,
      "num_tokens": 14921621.0,
      "step": 8075
    },
    {
      "entropy": 4.809543466567993,
      "epoch": 6.7615062761506275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017507274339966368,
      "loss": 4.2076,
      "mean_token_accuracy": 0.26108432859182357,
      "num_tokens": 14930732.0,
      "step": 8080
    },
    {
      "entropy": 4.849083042144775,
      "epoch": 6.7656903765690375,
      "grad_norm": 1.375,
      "learning_rate": 0.00017478365838255978,
      "loss": 4.3037,
      "mean_token_accuracy": 0.254167939722538,
      "num_tokens": 14939626.0,
      "step": 8085
    },
    {
      "entropy": 4.9347635269165036,
      "epoch": 6.7698744769874475,
      "grad_norm": 1.25,
      "learning_rate": 0.0001744947795945746,
      "loss": 4.2765,
      "mean_token_accuracy": 0.2591494545340538,
      "num_tokens": 14948718.0,
      "step": 8090
    },
    {
      "entropy": 4.817011451721191,
      "epoch": 6.7740585774058575,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001742061076301744,
      "loss": 4.1271,
      "mean_token_accuracy": 0.2658863469958305,
      "num_tokens": 14957703.0,
      "step": 8095
    },
    {
      "entropy": 4.815194177627563,
      "epoch": 6.7782426778242675,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017391764308339957,
      "loss": 4.2225,
      "mean_token_accuracy": 0.2585373193025589,
      "num_tokens": 14966693.0,
      "step": 8100
    },
    {
      "entropy": 4.867482280731201,
      "epoch": 6.7824267782426775,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017362938654786383,
      "loss": 4.3291,
      "mean_token_accuracy": 0.2513656660914421,
      "num_tokens": 14977267.0,
      "step": 8105
    },
    {
      "entropy": 4.958212184906006,
      "epoch": 6.786610878661088,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017334133861675278,
      "loss": 4.3119,
      "mean_token_accuracy": 0.24928641468286514,
      "num_tokens": 14985648.0,
      "step": 8110
    },
    {
      "entropy": 4.868992853164673,
      "epoch": 6.790794979079498,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001730534998828226,
      "loss": 4.2615,
      "mean_token_accuracy": 0.25616350024938583,
      "num_tokens": 14994915.0,
      "step": 8115
    },
    {
      "entropy": 4.848013782501221,
      "epoch": 6.794979079497908,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001727658709383993,
      "loss": 4.2405,
      "mean_token_accuracy": 0.2578317865729332,
      "num_tokens": 15004288.0,
      "step": 8120
    },
    {
      "entropy": 4.9150121212005615,
      "epoch": 6.799163179916318,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017247845237537693,
      "loss": 4.3116,
      "mean_token_accuracy": 0.2574670657515526,
      "num_tokens": 15012970.0,
      "step": 8125
    },
    {
      "entropy": 4.869945001602173,
      "epoch": 6.803347280334728,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017219124478521676,
      "loss": 4.3291,
      "mean_token_accuracy": 0.24649845361709594,
      "num_tokens": 15022989.0,
      "step": 8130
    },
    {
      "entropy": 4.964499139785767,
      "epoch": 6.807531380753138,
      "grad_norm": 1.25,
      "learning_rate": 0.00017190424875894567,
      "loss": 4.295,
      "mean_token_accuracy": 0.26333503127098085,
      "num_tokens": 15032665.0,
      "step": 8135
    },
    {
      "entropy": 4.8997962474823,
      "epoch": 6.811715481171548,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00017161746488715546,
      "loss": 4.2586,
      "mean_token_accuracy": 0.25248582661151886,
      "num_tokens": 15041109.0,
      "step": 8140
    },
    {
      "entropy": 4.806291913986206,
      "epoch": 6.815899581589958,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017133089376000142,
      "loss": 4.2153,
      "mean_token_accuracy": 0.2676341384649277,
      "num_tokens": 15050080.0,
      "step": 8145
    },
    {
      "entropy": 4.838059616088867,
      "epoch": 6.820083682008368,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001710445359672005,
      "loss": 4.2591,
      "mean_token_accuracy": 0.257693475484848,
      "num_tokens": 15058429.0,
      "step": 8150
    },
    {
      "entropy": 4.852031469345093,
      "epoch": 6.824267782426778,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017075839209803117,
      "loss": 4.2213,
      "mean_token_accuracy": 0.24955253303050995,
      "num_tokens": 15068652.0,
      "step": 8155
    },
    {
      "entropy": 4.884077262878418,
      "epoch": 6.828451882845188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017047246274133144,
      "loss": 4.2213,
      "mean_token_accuracy": 0.25899398177862165,
      "num_tokens": 15077643.0,
      "step": 8160
    },
    {
      "entropy": 4.878689575195312,
      "epoch": 6.832635983263598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001701867484854981,
      "loss": 4.2787,
      "mean_token_accuracy": 0.2549693092703819,
      "num_tokens": 15087471.0,
      "step": 8165
    },
    {
      "entropy": 4.971179580688476,
      "epoch": 6.836820083682008,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016990124991848497,
      "loss": 4.3879,
      "mean_token_accuracy": 0.25190470218658445,
      "num_tokens": 15096357.0,
      "step": 8170
    },
    {
      "entropy": 4.863401269912719,
      "epoch": 6.841004184100418,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001696159676278023,
      "loss": 4.2411,
      "mean_token_accuracy": 0.25683325082063674,
      "num_tokens": 15105424.0,
      "step": 8175
    },
    {
      "entropy": 4.904805660247803,
      "epoch": 6.845188284518828,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001693309022005153,
      "loss": 4.2808,
      "mean_token_accuracy": 0.2604569837450981,
      "num_tokens": 15115641.0,
      "step": 8180
    },
    {
      "entropy": 4.9058074951171875,
      "epoch": 6.849372384937238,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016904605422324255,
      "loss": 4.2352,
      "mean_token_accuracy": 0.25957991629838945,
      "num_tokens": 15124765.0,
      "step": 8185
    },
    {
      "entropy": 4.806936359405517,
      "epoch": 6.853556485355648,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016876142428215552,
      "loss": 4.2209,
      "mean_token_accuracy": 0.2615069255232811,
      "num_tokens": 15133591.0,
      "step": 8190
    },
    {
      "entropy": 4.8744291305542,
      "epoch": 6.857740585774058,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001684770129629768,
      "loss": 4.2937,
      "mean_token_accuracy": 0.2483627751469612,
      "num_tokens": 15143291.0,
      "step": 8195
    },
    {
      "entropy": 4.898120880126953,
      "epoch": 6.861924686192468,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016819282085097918,
      "loss": 4.3485,
      "mean_token_accuracy": 0.25307171493768693,
      "num_tokens": 15152806.0,
      "step": 8200
    },
    {
      "entropy": 4.911009693145752,
      "epoch": 6.866108786610878,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00016790884853098441,
      "loss": 4.266,
      "mean_token_accuracy": 0.25389271080493925,
      "num_tokens": 15163116.0,
      "step": 8205
    },
    {
      "entropy": 4.849863719940186,
      "epoch": 6.870292887029288,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001676250965873619,
      "loss": 4.285,
      "mean_token_accuracy": 0.25495756417512894,
      "num_tokens": 15172916.0,
      "step": 8210
    },
    {
      "entropy": 4.887068605422973,
      "epoch": 6.874476987447698,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016734156560402742,
      "loss": 4.261,
      "mean_token_accuracy": 0.2565524235367775,
      "num_tokens": 15181715.0,
      "step": 8215
    },
    {
      "entropy": 4.839970970153809,
      "epoch": 6.878661087866108,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001670582561644421,
      "loss": 4.2283,
      "mean_token_accuracy": 0.2580070897936821,
      "num_tokens": 15191290.0,
      "step": 8220
    },
    {
      "entropy": 4.868176460266113,
      "epoch": 6.882845188284519,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016677516885161127,
      "loss": 4.2515,
      "mean_token_accuracy": 0.26074880361557007,
      "num_tokens": 15200614.0,
      "step": 8225
    },
    {
      "entropy": 4.99262490272522,
      "epoch": 6.887029288702929,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001664923042480832,
      "loss": 4.3713,
      "mean_token_accuracy": 0.2388916164636612,
      "num_tokens": 15210793.0,
      "step": 8230
    },
    {
      "entropy": 4.848638534545898,
      "epoch": 6.891213389121339,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016620966293594763,
      "loss": 4.1971,
      "mean_token_accuracy": 0.2602669090032578,
      "num_tokens": 15219968.0,
      "step": 8235
    },
    {
      "entropy": 4.870445251464844,
      "epoch": 6.895397489539749,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016592724549683514,
      "loss": 4.2564,
      "mean_token_accuracy": 0.2572238802909851,
      "num_tokens": 15229434.0,
      "step": 8240
    },
    {
      "entropy": 4.8222362995147705,
      "epoch": 6.899581589958159,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001656450525119152,
      "loss": 4.1778,
      "mean_token_accuracy": 0.2640391528606415,
      "num_tokens": 15238784.0,
      "step": 8245
    },
    {
      "entropy": 4.891118001937866,
      "epoch": 6.903765690376569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016536308456189568,
      "loss": 4.319,
      "mean_token_accuracy": 0.24879578799009322,
      "num_tokens": 15247926.0,
      "step": 8250
    },
    {
      "entropy": 4.946751356124878,
      "epoch": 6.907949790794979,
      "grad_norm": 1.390625,
      "learning_rate": 0.00016508134222702133,
      "loss": 4.3512,
      "mean_token_accuracy": 0.24704710245132447,
      "num_tokens": 15257158.0,
      "step": 8255
    },
    {
      "entropy": 4.903260087966919,
      "epoch": 6.912133891213389,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016479982608707245,
      "loss": 4.3082,
      "mean_token_accuracy": 0.25689251720905304,
      "num_tokens": 15266639.0,
      "step": 8260
    },
    {
      "entropy": 4.862286710739136,
      "epoch": 6.916317991631799,
      "grad_norm": 1.375,
      "learning_rate": 0.00016451853672136418,
      "loss": 4.2633,
      "mean_token_accuracy": 0.2554127112030983,
      "num_tokens": 15275446.0,
      "step": 8265
    },
    {
      "entropy": 4.856708002090454,
      "epoch": 6.920502092050209,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001642374747087446,
      "loss": 4.3296,
      "mean_token_accuracy": 0.24469215869903566,
      "num_tokens": 15285011.0,
      "step": 8270
    },
    {
      "entropy": 4.903177738189697,
      "epoch": 6.924686192468619,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00016395664062759442,
      "loss": 4.4398,
      "mean_token_accuracy": 0.24572516083717347,
      "num_tokens": 15295066.0,
      "step": 8275
    },
    {
      "entropy": 4.961024904251099,
      "epoch": 6.928870292887029,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016367603505582483,
      "loss": 4.2717,
      "mean_token_accuracy": 0.25419566333293914,
      "num_tokens": 15304660.0,
      "step": 8280
    },
    {
      "entropy": 4.875931072235107,
      "epoch": 6.933054393305439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001633956585708769,
      "loss": 4.2567,
      "mean_token_accuracy": 0.2549095168709755,
      "num_tokens": 15314376.0,
      "step": 8285
    },
    {
      "entropy": 4.785683441162109,
      "epoch": 6.937238493723849,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016311551174972062,
      "loss": 4.1824,
      "mean_token_accuracy": 0.26544595807790755,
      "num_tokens": 15323414.0,
      "step": 8290
    },
    {
      "entropy": 4.85221905708313,
      "epoch": 6.941422594142259,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00016283559516885285,
      "loss": 4.2783,
      "mean_token_accuracy": 0.2469296023249626,
      "num_tokens": 15332266.0,
      "step": 8295
    },
    {
      "entropy": 4.8977837562561035,
      "epoch": 6.945606694560669,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016255590940429717,
      "loss": 4.2629,
      "mean_token_accuracy": 0.2544433206319809,
      "num_tokens": 15342238.0,
      "step": 8300
    },
    {
      "entropy": 4.957839584350586,
      "epoch": 6.949790794979079,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001622764550316016,
      "loss": 4.3501,
      "mean_token_accuracy": 0.24854763597249985,
      "num_tokens": 15351565.0,
      "step": 8305
    },
    {
      "entropy": 4.9110376834869385,
      "epoch": 6.95397489539749,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001619972326258386,
      "loss": 4.2821,
      "mean_token_accuracy": 0.25371793657541275,
      "num_tokens": 15361064.0,
      "step": 8310
    },
    {
      "entropy": 4.830585908889771,
      "epoch": 6.9581589958159,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016171824276160288,
      "loss": 4.2908,
      "mean_token_accuracy": 0.25340026468038557,
      "num_tokens": 15371152.0,
      "step": 8315
    },
    {
      "entropy": 4.915218210220337,
      "epoch": 6.96234309623431,
      "grad_norm": 1.34375,
      "learning_rate": 0.00016143948601301064,
      "loss": 4.2995,
      "mean_token_accuracy": 0.24950313270092012,
      "num_tokens": 15381546.0,
      "step": 8320
    },
    {
      "entropy": 4.864971685409546,
      "epoch": 6.96652719665272,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001611609629536986,
      "loss": 4.2383,
      "mean_token_accuracy": 0.2612897098064423,
      "num_tokens": 15390581.0,
      "step": 8325
    },
    {
      "entropy": 4.869308042526245,
      "epoch": 6.97071129707113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001608826741568223,
      "loss": 4.2573,
      "mean_token_accuracy": 0.2655170798301697,
      "num_tokens": 15400035.0,
      "step": 8330
    },
    {
      "entropy": 4.886803531646729,
      "epoch": 6.97489539748954,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001606046201950555,
      "loss": 4.3119,
      "mean_token_accuracy": 0.24939847439527513,
      "num_tokens": 15409199.0,
      "step": 8335
    },
    {
      "entropy": 4.832603073120117,
      "epoch": 6.97907949790795,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016032680164058832,
      "loss": 4.2068,
      "mean_token_accuracy": 0.2640422940254211,
      "num_tokens": 15417851.0,
      "step": 8340
    },
    {
      "entropy": 4.8886829853057865,
      "epoch": 6.98326359832636,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016004921906512699,
      "loss": 4.2144,
      "mean_token_accuracy": 0.2642232820391655,
      "num_tokens": 15427072.0,
      "step": 8345
    },
    {
      "entropy": 4.903195381164551,
      "epoch": 6.98744769874477,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001597718730398916,
      "loss": 4.3049,
      "mean_token_accuracy": 0.25140344351530075,
      "num_tokens": 15437486.0,
      "step": 8350
    },
    {
      "entropy": 4.877507734298706,
      "epoch": 6.99163179916318,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015949476413561564,
      "loss": 4.2639,
      "mean_token_accuracy": 0.25702105462551117,
      "num_tokens": 15447650.0,
      "step": 8355
    },
    {
      "entropy": 4.822701501846313,
      "epoch": 6.99581589958159,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00015921789292254484,
      "loss": 4.2023,
      "mean_token_accuracy": 0.2630909562110901,
      "num_tokens": 15456650.0,
      "step": 8360
    },
    {
      "entropy": 4.832033109664917,
      "epoch": 7.0,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015894125997043551,
      "loss": 4.2397,
      "mean_token_accuracy": 0.26185877323150636,
      "num_tokens": 15465254.0,
      "step": 8365
    },
    {
      "entropy": 4.897889757156372,
      "epoch": 7.00418410041841,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015866486584855397,
      "loss": 4.1261,
      "mean_token_accuracy": 0.27315237522125246,
      "num_tokens": 15474018.0,
      "step": 8370
    },
    {
      "entropy": 4.918653202056885,
      "epoch": 7.00836820083682,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015838871112567465,
      "loss": 4.1861,
      "mean_token_accuracy": 0.2603330373764038,
      "num_tokens": 15483871.0,
      "step": 8375
    },
    {
      "entropy": 4.852699136734008,
      "epoch": 7.01255230125523,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001581127963700798,
      "loss": 4.1127,
      "mean_token_accuracy": 0.2689139887690544,
      "num_tokens": 15492996.0,
      "step": 8380
    },
    {
      "entropy": 4.930728721618652,
      "epoch": 7.01673640167364,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001578371221495575,
      "loss": 4.2046,
      "mean_token_accuracy": 0.2617241308093071,
      "num_tokens": 15501703.0,
      "step": 8385
    },
    {
      "entropy": 4.8944072246551515,
      "epoch": 7.02092050209205,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001575616890314009,
      "loss": 4.159,
      "mean_token_accuracy": 0.26958867311477663,
      "num_tokens": 15510110.0,
      "step": 8390
    },
    {
      "entropy": 4.788698387145996,
      "epoch": 7.02510460251046,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015728649758240727,
      "loss": 4.0849,
      "mean_token_accuracy": 0.27451213002204894,
      "num_tokens": 15518352.0,
      "step": 8395
    },
    {
      "entropy": 4.834490585327148,
      "epoch": 7.02928870292887,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001570115483688762,
      "loss": 4.0808,
      "mean_token_accuracy": 0.2790457621216774,
      "num_tokens": 15527133.0,
      "step": 8400
    },
    {
      "entropy": 4.8102654933929445,
      "epoch": 7.03347280334728,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001567368419566091,
      "loss": 4.0718,
      "mean_token_accuracy": 0.2743163824081421,
      "num_tokens": 15536326.0,
      "step": 8405
    },
    {
      "entropy": 4.92043662071228,
      "epoch": 7.03765690376569,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015646237891090749,
      "loss": 4.1944,
      "mean_token_accuracy": 0.25921525061130524,
      "num_tokens": 15546408.0,
      "step": 8410
    },
    {
      "entropy": 4.874996614456177,
      "epoch": 7.0418410041841,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001561881597965722,
      "loss": 4.1973,
      "mean_token_accuracy": 0.2626881882548332,
      "num_tokens": 15554988.0,
      "step": 8415
    },
    {
      "entropy": 4.861639070510864,
      "epoch": 7.046025104602511,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015591418517790217,
      "loss": 4.0497,
      "mean_token_accuracy": 0.2832050368189812,
      "num_tokens": 15564127.0,
      "step": 8420
    },
    {
      "entropy": 4.879622554779052,
      "epoch": 7.050209205020921,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000155640455618693,
      "loss": 4.1818,
      "mean_token_accuracy": 0.2661409631371498,
      "num_tokens": 15573658.0,
      "step": 8425
    },
    {
      "entropy": 4.913450241088867,
      "epoch": 7.054393305439331,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00015536697168223626,
      "loss": 4.1995,
      "mean_token_accuracy": 0.26531809717416766,
      "num_tokens": 15583728.0,
      "step": 8430
    },
    {
      "entropy": 4.769423007965088,
      "epoch": 7.058577405857741,
      "grad_norm": 1.375,
      "learning_rate": 0.0001550937339313177,
      "loss": 4.0953,
      "mean_token_accuracy": 0.27650831937789916,
      "num_tokens": 15593335.0,
      "step": 8435
    },
    {
      "entropy": 4.795948076248169,
      "epoch": 7.062761506276151,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015482074292821687,
      "loss": 4.0275,
      "mean_token_accuracy": 0.27450029700994494,
      "num_tokens": 15603133.0,
      "step": 8440
    },
    {
      "entropy": 4.869350481033325,
      "epoch": 7.066945606694561,
      "grad_norm": 1.375,
      "learning_rate": 0.0001545479992347055,
      "loss": 4.1563,
      "mean_token_accuracy": 0.2627036228775978,
      "num_tokens": 15612509.0,
      "step": 8445
    },
    {
      "entropy": 4.850227117538452,
      "epoch": 7.071129707112971,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015427550341204585,
      "loss": 4.1241,
      "mean_token_accuracy": 0.2757692188024521,
      "num_tokens": 15622160.0,
      "step": 8450
    },
    {
      "entropy": 4.97234148979187,
      "epoch": 7.075313807531381,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015400325602099086,
      "loss": 4.2526,
      "mean_token_accuracy": 0.25731676667928693,
      "num_tokens": 15631268.0,
      "step": 8455
    },
    {
      "entropy": 4.95090970993042,
      "epoch": 7.079497907949791,
      "grad_norm": 1.484375,
      "learning_rate": 0.00015373125762178168,
      "loss": 4.1966,
      "mean_token_accuracy": 0.2606123358011246,
      "num_tokens": 15640153.0,
      "step": 8460
    },
    {
      "entropy": 4.849340152740479,
      "epoch": 7.083682008368201,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00015345950877414732,
      "loss": 4.0477,
      "mean_token_accuracy": 0.275472329556942,
      "num_tokens": 15647752.0,
      "step": 8465
    },
    {
      "entropy": 4.915920400619507,
      "epoch": 7.087866108786611,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015318801003730344,
      "loss": 4.196,
      "mean_token_accuracy": 0.26615143567323685,
      "num_tokens": 15656822.0,
      "step": 8470
    },
    {
      "entropy": 4.806095170974731,
      "epoch": 7.092050209205021,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015291676196995047,
      "loss": 4.0543,
      "mean_token_accuracy": 0.2740225300192833,
      "num_tokens": 15665045.0,
      "step": 8475
    },
    {
      "entropy": 4.874582719802857,
      "epoch": 7.096234309623431,
      "grad_norm": 1.25,
      "learning_rate": 0.00015264576513027367,
      "loss": 4.1458,
      "mean_token_accuracy": 0.26340978145599364,
      "num_tokens": 15674206.0,
      "step": 8480
    },
    {
      "entropy": 4.951577091217041,
      "epoch": 7.100418410041841,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001523750200759406,
      "loss": 4.2221,
      "mean_token_accuracy": 0.25723854154348375,
      "num_tokens": 15683075.0,
      "step": 8485
    },
    {
      "entropy": 4.83560962677002,
      "epoch": 7.104602510460251,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001521045273641013,
      "loss": 4.0508,
      "mean_token_accuracy": 0.27632636576890945,
      "num_tokens": 15691698.0,
      "step": 8490
    },
    {
      "entropy": 4.898476505279541,
      "epoch": 7.108786610878661,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00015183428755138613,
      "loss": 4.2158,
      "mean_token_accuracy": 0.2609706729650497,
      "num_tokens": 15701927.0,
      "step": 8495
    },
    {
      "entropy": 4.8751959800720215,
      "epoch": 7.112970711297071,
      "grad_norm": 1.453125,
      "learning_rate": 0.00015156430119390524,
      "loss": 4.1822,
      "mean_token_accuracy": 0.2627851277589798,
      "num_tokens": 15710813.0,
      "step": 8500
    },
    {
      "epoch": 7.112970711297071,
      "eval_entropy": 4.770936147323171,
      "eval_loss": 5.324575901031494,
      "eval_mean_token_accuracy": 0.20208134381805218,
      "eval_num_tokens": 15710813.0,
      "eval_runtime": 2.7498,
      "eval_samples_per_second": 1420.839,
      "eval_steps_per_second": 177.832,
      "step": 8500
    },
    {
      "entropy": 4.9473268508911135,
      "epoch": 7.117154811715481,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015129456884724723,
      "loss": 4.1581,
      "mean_token_accuracy": 0.26040602177381517,
      "num_tokens": 15720559.0,
      "step": 8505
    },
    {
      "entropy": 4.8496805191040036,
      "epoch": 7.121338912133891,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015102509106647782,
      "loss": 4.1423,
      "mean_token_accuracy": 0.267295403778553,
      "num_tokens": 15730649.0,
      "step": 8510
    },
    {
      "entropy": 4.846027612686157,
      "epoch": 7.125523012552302,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015075586840613902,
      "loss": 4.1051,
      "mean_token_accuracy": 0.276242783665657,
      "num_tokens": 15739619.0,
      "step": 8515
    },
    {
      "entropy": 4.859666538238526,
      "epoch": 7.129707112970712,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00015048690142024763,
      "loss": 4.1837,
      "mean_token_accuracy": 0.26712564527988436,
      "num_tokens": 15748457.0,
      "step": 8520
    },
    {
      "entropy": 4.83686671257019,
      "epoch": 7.133891213389122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015021819066229452,
      "loss": 4.0964,
      "mean_token_accuracy": 0.264189612865448,
      "num_tokens": 15758527.0,
      "step": 8525
    },
    {
      "entropy": 4.88442096710205,
      "epoch": 7.138075313807532,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014994973668524343,
      "loss": 4.2322,
      "mean_token_accuracy": 0.2572506874799728,
      "num_tokens": 15767980.0,
      "step": 8530
    },
    {
      "entropy": 4.801035308837891,
      "epoch": 7.142259414225942,
      "grad_norm": 1.375,
      "learning_rate": 0.00014968154004152922,
      "loss": 4.1105,
      "mean_token_accuracy": 0.27370408922433853,
      "num_tokens": 15777323.0,
      "step": 8535
    },
    {
      "entropy": 4.816228199005127,
      "epoch": 7.146443514644352,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014941360128305765,
      "loss": 4.1366,
      "mean_token_accuracy": 0.2776454210281372,
      "num_tokens": 15787589.0,
      "step": 8540
    },
    {
      "entropy": 4.900264072418213,
      "epoch": 7.150627615062762,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014914592096120348,
      "loss": 4.1612,
      "mean_token_accuracy": 0.2704088449478149,
      "num_tokens": 15797248.0,
      "step": 8545
    },
    {
      "entropy": 4.902388048171997,
      "epoch": 7.154811715481172,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014887849962680978,
      "loss": 4.2006,
      "mean_token_accuracy": 0.25468218624591826,
      "num_tokens": 15806389.0,
      "step": 8550
    },
    {
      "entropy": 4.907098579406738,
      "epoch": 7.158995815899582,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014861133783018678,
      "loss": 4.1515,
      "mean_token_accuracy": 0.27172635197639466,
      "num_tokens": 15814719.0,
      "step": 8555
    },
    {
      "entropy": 4.899928426742553,
      "epoch": 7.163179916317992,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014834443612111035,
      "loss": 4.2166,
      "mean_token_accuracy": 0.25961185693740846,
      "num_tokens": 15823751.0,
      "step": 8560
    },
    {
      "entropy": 4.931470108032227,
      "epoch": 7.167364016736402,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014807779504882147,
      "loss": 4.1997,
      "mean_token_accuracy": 0.263980995118618,
      "num_tokens": 15832717.0,
      "step": 8565
    },
    {
      "entropy": 4.837965965270996,
      "epoch": 7.171548117154812,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014781141516202439,
      "loss": 4.1245,
      "mean_token_accuracy": 0.2676696628332138,
      "num_tokens": 15841494.0,
      "step": 8570
    },
    {
      "entropy": 4.845661640167236,
      "epoch": 7.175732217573222,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001475452970088863,
      "loss": 4.1909,
      "mean_token_accuracy": 0.2627507135272026,
      "num_tokens": 15851619.0,
      "step": 8575
    },
    {
      "entropy": 4.888016414642334,
      "epoch": 7.179916317991632,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001472794411370355,
      "loss": 4.1953,
      "mean_token_accuracy": 0.26400979459285734,
      "num_tokens": 15861563.0,
      "step": 8580
    },
    {
      "entropy": 4.908272218704224,
      "epoch": 7.184100418410042,
      "grad_norm": 1.375,
      "learning_rate": 0.00014701384809356053,
      "loss": 4.1686,
      "mean_token_accuracy": 0.266612234711647,
      "num_tokens": 15870554.0,
      "step": 8585
    },
    {
      "entropy": 4.871564674377441,
      "epoch": 7.188284518828452,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014674851842500931,
      "loss": 4.1408,
      "mean_token_accuracy": 0.2718515515327454,
      "num_tokens": 15879679.0,
      "step": 8590
    },
    {
      "entropy": 4.812321043014526,
      "epoch": 7.192468619246862,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001464834526773875,
      "loss": 4.118,
      "mean_token_accuracy": 0.2648796260356903,
      "num_tokens": 15888533.0,
      "step": 8595
    },
    {
      "entropy": 4.872211170196533,
      "epoch": 7.196652719665272,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014621865139615795,
      "loss": 4.1275,
      "mean_token_accuracy": 0.27128332257270815,
      "num_tokens": 15896888.0,
      "step": 8600
    },
    {
      "entropy": 4.911849069595337,
      "epoch": 7.200836820083682,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014595411512623895,
      "loss": 4.1948,
      "mean_token_accuracy": 0.2646011829376221,
      "num_tokens": 15906843.0,
      "step": 8605
    },
    {
      "entropy": 4.788438940048218,
      "epoch": 7.205020920502092,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014568984441200384,
      "loss": 4.0956,
      "mean_token_accuracy": 0.2687289580702782,
      "num_tokens": 15915333.0,
      "step": 8610
    },
    {
      "entropy": 4.928157615661621,
      "epoch": 7.209205020920502,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014542583979727908,
      "loss": 4.1985,
      "mean_token_accuracy": 0.2629579335451126,
      "num_tokens": 15924187.0,
      "step": 8615
    },
    {
      "entropy": 4.903908014297485,
      "epoch": 7.2133891213389125,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014516210182534373,
      "loss": 4.1656,
      "mean_token_accuracy": 0.2729055091738701,
      "num_tokens": 15932763.0,
      "step": 8620
    },
    {
      "entropy": 4.872292470932007,
      "epoch": 7.2175732217573225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014489863103892825,
      "loss": 4.1156,
      "mean_token_accuracy": 0.2670705676078796,
      "num_tokens": 15942272.0,
      "step": 8625
    },
    {
      "entropy": 4.847052717208863,
      "epoch": 7.2217573221757325,
      "grad_norm": 1.359375,
      "learning_rate": 0.000144635427980213,
      "loss": 4.2116,
      "mean_token_accuracy": 0.2572847858071327,
      "num_tokens": 15952099.0,
      "step": 8630
    },
    {
      "entropy": 4.834543609619141,
      "epoch": 7.2259414225941425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014437249319082773,
      "loss": 4.1252,
      "mean_token_accuracy": 0.26500718146562574,
      "num_tokens": 15961206.0,
      "step": 8635
    },
    {
      "entropy": 4.847073125839233,
      "epoch": 7.2301255230125525,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00014410982721184984,
      "loss": 4.125,
      "mean_token_accuracy": 0.27491633147001265,
      "num_tokens": 15970694.0,
      "step": 8640
    },
    {
      "entropy": 4.891429138183594,
      "epoch": 7.2343096234309625,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014384743058380373,
      "loss": 4.1293,
      "mean_token_accuracy": 0.26978846192359923,
      "num_tokens": 15979271.0,
      "step": 8645
    },
    {
      "entropy": 4.860075378417969,
      "epoch": 7.2384937238493725,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014358530384665953,
      "loss": 4.1603,
      "mean_token_accuracy": 0.26602810323238374,
      "num_tokens": 15988124.0,
      "step": 8650
    },
    {
      "entropy": 4.887761354446411,
      "epoch": 7.2426778242677825,
      "grad_norm": 1.390625,
      "learning_rate": 0.00014332344753983174,
      "loss": 4.1881,
      "mean_token_accuracy": 0.26642207354307174,
      "num_tokens": 15998912.0,
      "step": 8655
    },
    {
      "entropy": 4.8735105991363525,
      "epoch": 7.2468619246861925,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014306186220217873,
      "loss": 4.1501,
      "mean_token_accuracy": 0.2633297950029373,
      "num_tokens": 16008295.0,
      "step": 8660
    },
    {
      "entropy": 4.890497016906738,
      "epoch": 7.2510460251046025,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014280054837200094,
      "loss": 4.1284,
      "mean_token_accuracy": 0.27586863189935684,
      "num_tokens": 16017120.0,
      "step": 8665
    },
    {
      "entropy": 4.92688307762146,
      "epoch": 7.2552301255230125,
      "grad_norm": 1.453125,
      "learning_rate": 0.00014253950658704034,
      "loss": 4.2407,
      "mean_token_accuracy": 0.25845147520303724,
      "num_tokens": 16026630.0,
      "step": 8670
    },
    {
      "entropy": 4.8994443893432615,
      "epoch": 7.2594142259414225,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014227873738447883,
      "loss": 4.1742,
      "mean_token_accuracy": 0.2635588049888611,
      "num_tokens": 16034897.0,
      "step": 8675
    },
    {
      "entropy": 4.802002191543579,
      "epoch": 7.2635983263598325,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00014201824130093764,
      "loss": 4.1102,
      "mean_token_accuracy": 0.26977455914020537,
      "num_tokens": 16044387.0,
      "step": 8680
    },
    {
      "entropy": 4.816623020172119,
      "epoch": 7.2677824267782425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014175801887247576,
      "loss": 4.1253,
      "mean_token_accuracy": 0.2662367284297943,
      "num_tokens": 16053659.0,
      "step": 8685
    },
    {
      "entropy": 4.857808589935303,
      "epoch": 7.2719665271966525,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014149807063458896,
      "loss": 4.1357,
      "mean_token_accuracy": 0.27447866797447207,
      "num_tokens": 16062997.0,
      "step": 8690
    },
    {
      "entropy": 4.897714233398437,
      "epoch": 7.2761506276150625,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014123839712220921,
      "loss": 4.1595,
      "mean_token_accuracy": 0.2599164515733719,
      "num_tokens": 16071491.0,
      "step": 8695
    },
    {
      "entropy": 4.923414993286133,
      "epoch": 7.2803347280334725,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001409789988697026,
      "loss": 4.2252,
      "mean_token_accuracy": 0.2675188437104225,
      "num_tokens": 16081008.0,
      "step": 8700
    },
    {
      "entropy": 4.85807204246521,
      "epoch": 7.2845188284518825,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014071987641086926,
      "loss": 4.1133,
      "mean_token_accuracy": 0.26677933186292646,
      "num_tokens": 16090516.0,
      "step": 8705
    },
    {
      "entropy": 4.8439164638519285,
      "epoch": 7.2887029288702925,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001404610302789414,
      "loss": 4.2051,
      "mean_token_accuracy": 0.26577525287866594,
      "num_tokens": 16100952.0,
      "step": 8710
    },
    {
      "entropy": 4.86191577911377,
      "epoch": 7.292887029288703,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014020246100658268,
      "loss": 4.176,
      "mean_token_accuracy": 0.2614373505115509,
      "num_tokens": 16110493.0,
      "step": 8715
    },
    {
      "entropy": 4.864150285720825,
      "epoch": 7.297071129707113,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013994416912588726,
      "loss": 4.1376,
      "mean_token_accuracy": 0.2737789660692215,
      "num_tokens": 16119227.0,
      "step": 8720
    },
    {
      "entropy": 4.861856985092163,
      "epoch": 7.301255230125523,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001396861551683781,
      "loss": 4.2478,
      "mean_token_accuracy": 0.26253372579813006,
      "num_tokens": 16129268.0,
      "step": 8725
    },
    {
      "entropy": 4.899822759628296,
      "epoch": 7.305439330543933,
      "grad_norm": 1.375,
      "learning_rate": 0.0001394284196650067,
      "loss": 4.1901,
      "mean_token_accuracy": 0.26335217505693437,
      "num_tokens": 16137523.0,
      "step": 8730
    },
    {
      "entropy": 4.899425315856933,
      "epoch": 7.309623430962343,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001391709631461511,
      "loss": 4.224,
      "mean_token_accuracy": 0.26522876918315885,
      "num_tokens": 16146535.0,
      "step": 8735
    },
    {
      "entropy": 4.869970655441284,
      "epoch": 7.313807531380753,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013891378614161538,
      "loss": 4.164,
      "mean_token_accuracy": 0.2626946985721588,
      "num_tokens": 16155892.0,
      "step": 8740
    },
    {
      "entropy": 4.844182348251342,
      "epoch": 7.317991631799163,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013865688918062875,
      "loss": 4.1312,
      "mean_token_accuracy": 0.262988843023777,
      "num_tokens": 16164697.0,
      "step": 8745
    },
    {
      "entropy": 4.877269315719604,
      "epoch": 7.322175732217573,
      "grad_norm": 1.515625,
      "learning_rate": 0.00013840027279184353,
      "loss": 4.1633,
      "mean_token_accuracy": 0.2623536616563797,
      "num_tokens": 16173387.0,
      "step": 8750
    },
    {
      "entropy": 4.853242635726929,
      "epoch": 7.326359832635983,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013814393750333524,
      "loss": 4.0984,
      "mean_token_accuracy": 0.27460795640945435,
      "num_tokens": 16182438.0,
      "step": 8755
    },
    {
      "entropy": 4.841980791091919,
      "epoch": 7.330543933054393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013788788384260043,
      "loss": 4.0894,
      "mean_token_accuracy": 0.27134654521942136,
      "num_tokens": 16193149.0,
      "step": 8760
    },
    {
      "entropy": 4.8949638366699215,
      "epoch": 7.334728033472803,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013763211233655656,
      "loss": 4.2287,
      "mean_token_accuracy": 0.2604277104139328,
      "num_tokens": 16202354.0,
      "step": 8765
    },
    {
      "entropy": 4.86085147857666,
      "epoch": 7.338912133891213,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013737662351154028,
      "loss": 4.177,
      "mean_token_accuracy": 0.25628658682107924,
      "num_tokens": 16210608.0,
      "step": 8770
    },
    {
      "entropy": 4.904495573043823,
      "epoch": 7.343096234309623,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013712141789330642,
      "loss": 4.2325,
      "mean_token_accuracy": 0.2641896948218346,
      "num_tokens": 16219544.0,
      "step": 8775
    },
    {
      "entropy": 4.85392541885376,
      "epoch": 7.347280334728033,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013686649600702728,
      "loss": 4.1217,
      "mean_token_accuracy": 0.2634648308157921,
      "num_tokens": 16228820.0,
      "step": 8780
    },
    {
      "entropy": 4.794688034057617,
      "epoch": 7.351464435146443,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013661185837729085,
      "loss": 4.028,
      "mean_token_accuracy": 0.2752455174922943,
      "num_tokens": 16237751.0,
      "step": 8785
    },
    {
      "entropy": 4.87891993522644,
      "epoch": 7.355648535564853,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013635750552810065,
      "loss": 4.1524,
      "mean_token_accuracy": 0.27512101233005526,
      "num_tokens": 16246944.0,
      "step": 8790
    },
    {
      "entropy": 4.910146903991699,
      "epoch": 7.359832635983263,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013610343798287376,
      "loss": 4.246,
      "mean_token_accuracy": 0.26094659566879275,
      "num_tokens": 16256565.0,
      "step": 8795
    },
    {
      "entropy": 4.806262969970703,
      "epoch": 7.364016736401673,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013584965626444045,
      "loss": 4.0779,
      "mean_token_accuracy": 0.2748170971870422,
      "num_tokens": 16264886.0,
      "step": 8800
    },
    {
      "entropy": 4.946498107910156,
      "epoch": 7.368200836820083,
      "grad_norm": 1.5,
      "learning_rate": 0.00013559616089504285,
      "loss": 4.2632,
      "mean_token_accuracy": 0.25725755393505095,
      "num_tokens": 16274039.0,
      "step": 8805
    },
    {
      "entropy": 4.842801761627197,
      "epoch": 7.372384937238493,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013534295239633342,
      "loss": 4.1537,
      "mean_token_accuracy": 0.26939127296209336,
      "num_tokens": 16283274.0,
      "step": 8810
    },
    {
      "entropy": 4.849980640411377,
      "epoch": 7.376569037656903,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001350900312893747,
      "loss": 4.1163,
      "mean_token_accuracy": 0.2758858188986778,
      "num_tokens": 16292048.0,
      "step": 8815
    },
    {
      "entropy": 4.855568218231201,
      "epoch": 7.380753138075314,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013483739809463753,
      "loss": 4.121,
      "mean_token_accuracy": 0.2653131425380707,
      "num_tokens": 16301910.0,
      "step": 8820
    },
    {
      "entropy": 4.867168617248535,
      "epoch": 7.384937238493724,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013458505333200044,
      "loss": 4.1487,
      "mean_token_accuracy": 0.2660368666052818,
      "num_tokens": 16311221.0,
      "step": 8825
    },
    {
      "entropy": 4.791717386245727,
      "epoch": 7.389121338912134,
      "grad_norm": 1.46875,
      "learning_rate": 0.00013433299752074857,
      "loss": 4.0964,
      "mean_token_accuracy": 0.2761238291859627,
      "num_tokens": 16319530.0,
      "step": 8830
    },
    {
      "entropy": 4.792592525482178,
      "epoch": 7.393305439330544,
      "grad_norm": 1.484375,
      "learning_rate": 0.000134081231179572,
      "loss": 4.0649,
      "mean_token_accuracy": 0.2744856745004654,
      "num_tokens": 16327927.0,
      "step": 8835
    },
    {
      "entropy": 4.9177062034606935,
      "epoch": 7.397489539748954,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013382975482656556,
      "loss": 4.1919,
      "mean_token_accuracy": 0.25646031349897386,
      "num_tokens": 16337551.0,
      "step": 8840
    },
    {
      "entropy": 4.856431293487549,
      "epoch": 7.401673640167364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001335785689792271,
      "loss": 4.1412,
      "mean_token_accuracy": 0.2678917944431305,
      "num_tokens": 16345937.0,
      "step": 8845
    },
    {
      "entropy": 4.895708465576172,
      "epoch": 7.405857740585774,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001333276741544566,
      "loss": 4.2087,
      "mean_token_accuracy": 0.2597071647644043,
      "num_tokens": 16355797.0,
      "step": 8850
    },
    {
      "entropy": 4.87229585647583,
      "epoch": 7.410041841004184,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013307707086855546,
      "loss": 4.1612,
      "mean_token_accuracy": 0.2638948380947113,
      "num_tokens": 16364547.0,
      "step": 8855
    },
    {
      "entropy": 4.802583456039429,
      "epoch": 7.414225941422594,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001328267596372248,
      "loss": 4.0452,
      "mean_token_accuracy": 0.2873002916574478,
      "num_tokens": 16373506.0,
      "step": 8860
    },
    {
      "entropy": 4.863208770751953,
      "epoch": 7.418410041841004,
      "grad_norm": 1.25,
      "learning_rate": 0.000132576740975565,
      "loss": 4.1089,
      "mean_token_accuracy": 0.2774243250489235,
      "num_tokens": 16383006.0,
      "step": 8865
    },
    {
      "entropy": 4.877791881561279,
      "epoch": 7.422594142259414,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013232701539807415,
      "loss": 4.1948,
      "mean_token_accuracy": 0.265048511326313,
      "num_tokens": 16392721.0,
      "step": 8870
    },
    {
      "entropy": 4.8189537048339846,
      "epoch": 7.426778242677824,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001320775834186475,
      "loss": 4.1377,
      "mean_token_accuracy": 0.2663384735584259,
      "num_tokens": 16402404.0,
      "step": 8875
    },
    {
      "entropy": 4.913763999938965,
      "epoch": 7.430962343096234,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013182844555057583,
      "loss": 4.1932,
      "mean_token_accuracy": 0.26417246013879775,
      "num_tokens": 16411386.0,
      "step": 8880
    },
    {
      "entropy": 4.817476415634156,
      "epoch": 7.435146443514644,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001315796023065448,
      "loss": 4.1221,
      "mean_token_accuracy": 0.268984979391098,
      "num_tokens": 16420407.0,
      "step": 8885
    },
    {
      "entropy": 4.801479291915894,
      "epoch": 7.439330543933054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013133105419863394,
      "loss": 4.0247,
      "mean_token_accuracy": 0.2923597127199173,
      "num_tokens": 16429197.0,
      "step": 8890
    },
    {
      "entropy": 4.873983764648438,
      "epoch": 7.443514644351464,
      "grad_norm": 1.375,
      "learning_rate": 0.00013108280173831517,
      "loss": 4.1829,
      "mean_token_accuracy": 0.2626629889011383,
      "num_tokens": 16438144.0,
      "step": 8895
    },
    {
      "entropy": 4.822049713134765,
      "epoch": 7.447698744769874,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001308348454364522,
      "loss": 4.1907,
      "mean_token_accuracy": 0.26922242194414137,
      "num_tokens": 16447861.0,
      "step": 8900
    },
    {
      "entropy": 4.824089813232422,
      "epoch": 7.451882845188284,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013058718580329917,
      "loss": 4.0322,
      "mean_token_accuracy": 0.28743570744991304,
      "num_tokens": 16456937.0,
      "step": 8905
    },
    {
      "entropy": 4.863957166671753,
      "epoch": 7.456066945606695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001303398233484999,
      "loss": 4.1787,
      "mean_token_accuracy": 0.26807293593883513,
      "num_tokens": 16466714.0,
      "step": 8910
    },
    {
      "entropy": 4.859753131866455,
      "epoch": 7.460251046025105,
      "grad_norm": 1.5625,
      "learning_rate": 0.00013009275858108648,
      "loss": 4.1087,
      "mean_token_accuracy": 0.2712841719388962,
      "num_tokens": 16475219.0,
      "step": 8915
    },
    {
      "entropy": 4.803961181640625,
      "epoch": 7.464435146443515,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001298459920094784,
      "loss": 4.065,
      "mean_token_accuracy": 0.270451195538044,
      "num_tokens": 16484169.0,
      "step": 8920
    },
    {
      "entropy": 4.847228288650513,
      "epoch": 7.468619246861925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001295995241414818,
      "loss": 4.1554,
      "mean_token_accuracy": 0.26732047349214555,
      "num_tokens": 16493413.0,
      "step": 8925
    },
    {
      "entropy": 4.829778671264648,
      "epoch": 7.472803347280335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012935335548428767,
      "loss": 4.125,
      "mean_token_accuracy": 0.26589876413345337,
      "num_tokens": 16503195.0,
      "step": 8930
    },
    {
      "entropy": 4.825352621078491,
      "epoch": 7.476987447698745,
      "grad_norm": 1.375,
      "learning_rate": 0.0001291074865444718,
      "loss": 4.0465,
      "mean_token_accuracy": 0.26883878707885744,
      "num_tokens": 16511168.0,
      "step": 8935
    },
    {
      "entropy": 4.847069501876831,
      "epoch": 7.481171548117155,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012886191782799267,
      "loss": 4.141,
      "mean_token_accuracy": 0.2733569532632828,
      "num_tokens": 16520125.0,
      "step": 8940
    },
    {
      "entropy": 4.796758556365967,
      "epoch": 7.485355648535565,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012861664984019144,
      "loss": 4.0585,
      "mean_token_accuracy": 0.2742341339588165,
      "num_tokens": 16528121.0,
      "step": 8945
    },
    {
      "entropy": 4.7933584690094,
      "epoch": 7.489539748953975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012837168308579012,
      "loss": 4.0844,
      "mean_token_accuracy": 0.27577037215232847,
      "num_tokens": 16537741.0,
      "step": 8950
    },
    {
      "entropy": 4.855648708343506,
      "epoch": 7.493723849372385,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012812701806889077,
      "loss": 4.1707,
      "mean_token_accuracy": 0.26674157232046125,
      "num_tokens": 16548091.0,
      "step": 8955
    },
    {
      "entropy": 4.881253051757812,
      "epoch": 7.497907949790795,
      "grad_norm": 1.25,
      "learning_rate": 0.0001278826552929749,
      "loss": 4.1582,
      "mean_token_accuracy": 0.266350582242012,
      "num_tokens": 16557011.0,
      "step": 8960
    },
    {
      "entropy": 4.902585840225219,
      "epoch": 7.502092050209205,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001276385952609016,
      "loss": 4.1863,
      "mean_token_accuracy": 0.2587046086788177,
      "num_tokens": 16565868.0,
      "step": 8965
    },
    {
      "entropy": 4.882581329345703,
      "epoch": 7.506276150627615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012739483847490734,
      "loss": 4.1721,
      "mean_token_accuracy": 0.26881828904151917,
      "num_tokens": 16575131.0,
      "step": 8970
    },
    {
      "entropy": 4.879101371765136,
      "epoch": 7.510460251046025,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012715138543660425,
      "loss": 4.187,
      "mean_token_accuracy": 0.26573489159345626,
      "num_tokens": 16584789.0,
      "step": 8975
    },
    {
      "entropy": 4.898157119750977,
      "epoch": 7.514644351464435,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001269082366469797,
      "loss": 4.1764,
      "mean_token_accuracy": 0.26596491038799286,
      "num_tokens": 16594488.0,
      "step": 8980
    },
    {
      "entropy": 4.802444314956665,
      "epoch": 7.518828451882845,
      "grad_norm": 1.5625,
      "learning_rate": 0.00012666539260639475,
      "loss": 4.0625,
      "mean_token_accuracy": 0.2733553409576416,
      "num_tokens": 16602691.0,
      "step": 8985
    },
    {
      "entropy": 4.925993156433106,
      "epoch": 7.523012552301255,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001264228538145833,
      "loss": 4.1882,
      "mean_token_accuracy": 0.26833981871604917,
      "num_tokens": 16611843.0,
      "step": 8990
    },
    {
      "entropy": 4.914250755310059,
      "epoch": 7.527196652719665,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012618062077065126,
      "loss": 4.1873,
      "mean_token_accuracy": 0.25436250418424605,
      "num_tokens": 16620587.0,
      "step": 8995
    },
    {
      "entropy": 4.901272296905518,
      "epoch": 7.531380753138075,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012593869397307527,
      "loss": 4.2602,
      "mean_token_accuracy": 0.25316916704177855,
      "num_tokens": 16631113.0,
      "step": 9000
    },
    {
      "epoch": 7.531380753138075,
      "eval_entropy": 4.753826975091103,
      "eval_loss": 5.321648120880127,
      "eval_mean_token_accuracy": 0.20316442187517456,
      "eval_num_tokens": 16631113.0,
      "eval_runtime": 2.9397,
      "eval_samples_per_second": 1329.064,
      "eval_steps_per_second": 166.346,
      "step": 9000
    },
    {
      "entropy": 4.905578470230102,
      "epoch": 7.535564853556485,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012569707391970187,
      "loss": 4.0882,
      "mean_token_accuracy": 0.27134772539138796,
      "num_tokens": 16639747.0,
      "step": 9005
    },
    {
      "entropy": 4.885472154617309,
      "epoch": 7.539748953974895,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012545576110774617,
      "loss": 4.1374,
      "mean_token_accuracy": 0.26891689747571945,
      "num_tokens": 16648619.0,
      "step": 9010
    },
    {
      "entropy": 4.869478178024292,
      "epoch": 7.543933054393305,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012521475603379117,
      "loss": 4.226,
      "mean_token_accuracy": 0.26458795219659803,
      "num_tokens": 16656860.0,
      "step": 9015
    },
    {
      "entropy": 4.934126853942871,
      "epoch": 7.548117154811716,
      "grad_norm": 1.3125,
      "learning_rate": 0.00012497405919378662,
      "loss": 4.2617,
      "mean_token_accuracy": 0.26156105399131774,
      "num_tokens": 16666121.0,
      "step": 9020
    },
    {
      "entropy": 4.875466775894165,
      "epoch": 7.552301255230126,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001247336710830478,
      "loss": 4.1267,
      "mean_token_accuracy": 0.2695460766553879,
      "num_tokens": 16675296.0,
      "step": 9025
    },
    {
      "entropy": 4.842286539077759,
      "epoch": 7.556485355648536,
      "grad_norm": 1.53125,
      "learning_rate": 0.00012449359219625494,
      "loss": 4.0914,
      "mean_token_accuracy": 0.28073961436748507,
      "num_tokens": 16683554.0,
      "step": 9030
    },
    {
      "entropy": 4.92423267364502,
      "epoch": 7.560669456066946,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012425382302745167,
      "loss": 4.1897,
      "mean_token_accuracy": 0.26741879135370256,
      "num_tokens": 16692081.0,
      "step": 9035
    },
    {
      "entropy": 4.92785792350769,
      "epoch": 7.564853556485356,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012401436407004444,
      "loss": 4.221,
      "mean_token_accuracy": 0.25792713910341264,
      "num_tokens": 16701543.0,
      "step": 9040
    },
    {
      "entropy": 4.870514011383056,
      "epoch": 7.569037656903766,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012377521581680146,
      "loss": 4.1629,
      "mean_token_accuracy": 0.26579001247882844,
      "num_tokens": 16710044.0,
      "step": 9045
    },
    {
      "entropy": 4.889408111572266,
      "epoch": 7.573221757322176,
      "grad_norm": 1.375,
      "learning_rate": 0.000123536378759851,
      "loss": 4.1556,
      "mean_token_accuracy": 0.27675602585077286,
      "num_tokens": 16718765.0,
      "step": 9050
    },
    {
      "entropy": 4.802488660812378,
      "epoch": 7.577405857740586,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012329785339068163,
      "loss": 4.0677,
      "mean_token_accuracy": 0.26978994011878965,
      "num_tokens": 16727470.0,
      "step": 9055
    },
    {
      "entropy": 4.911803388595581,
      "epoch": 7.581589958158996,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001230596402001401,
      "loss": 4.2112,
      "mean_token_accuracy": 0.2603839635848999,
      "num_tokens": 16736557.0,
      "step": 9060
    },
    {
      "entropy": 4.876327896118164,
      "epoch": 7.585774058577406,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00012282173967843085,
      "loss": 4.1454,
      "mean_token_accuracy": 0.272234608232975,
      "num_tokens": 16745601.0,
      "step": 9065
    },
    {
      "entropy": 4.90815691947937,
      "epoch": 7.589958158995816,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012258415231511497,
      "loss": 4.1814,
      "mean_token_accuracy": 0.2694088816642761,
      "num_tokens": 16756141.0,
      "step": 9070
    },
    {
      "entropy": 4.8483093738555905,
      "epoch": 7.594142259414226,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012234687859910895,
      "loss": 4.202,
      "mean_token_accuracy": 0.2640699580311775,
      "num_tokens": 16766080.0,
      "step": 9075
    },
    {
      "entropy": 4.8648974895477295,
      "epoch": 7.598326359832636,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001221099190186842,
      "loss": 4.1417,
      "mean_token_accuracy": 0.2622421011328697,
      "num_tokens": 16775826.0,
      "step": 9080
    },
    {
      "entropy": 4.825019311904907,
      "epoch": 7.602510460251046,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001218732740614651,
      "loss": 4.1857,
      "mean_token_accuracy": 0.2669943183660507,
      "num_tokens": 16785139.0,
      "step": 9085
    },
    {
      "entropy": 4.816349267959595,
      "epoch": 7.606694560669456,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012163694421442905,
      "loss": 4.1484,
      "mean_token_accuracy": 0.2705557107925415,
      "num_tokens": 16793935.0,
      "step": 9090
    },
    {
      "entropy": 4.866408395767212,
      "epoch": 7.610878661087866,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012140092996390502,
      "loss": 4.1148,
      "mean_token_accuracy": 0.2685813635587692,
      "num_tokens": 16803451.0,
      "step": 9095
    },
    {
      "entropy": 4.9276025772094725,
      "epoch": 7.615062761506276,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012116523179557221,
      "loss": 4.1441,
      "mean_token_accuracy": 0.2729563295841217,
      "num_tokens": 16813229.0,
      "step": 9100
    },
    {
      "entropy": 4.867234802246093,
      "epoch": 7.619246861924686,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012092985019445968,
      "loss": 4.1305,
      "mean_token_accuracy": 0.26671237498521805,
      "num_tokens": 16822088.0,
      "step": 9105
    },
    {
      "entropy": 4.887724494934082,
      "epoch": 7.623430962343097,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001206947856449448,
      "loss": 4.1642,
      "mean_token_accuracy": 0.2650635838508606,
      "num_tokens": 16831172.0,
      "step": 9110
    },
    {
      "entropy": 4.889850425720215,
      "epoch": 7.627615062761507,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012046003863075285,
      "loss": 4.2146,
      "mean_token_accuracy": 0.2609300330281258,
      "num_tokens": 16840482.0,
      "step": 9115
    },
    {
      "entropy": 4.877443933486939,
      "epoch": 7.631799163179917,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012022560963495516,
      "loss": 4.1606,
      "mean_token_accuracy": 0.25309752076864245,
      "num_tokens": 16849205.0,
      "step": 9120
    },
    {
      "entropy": 4.903221225738525,
      "epoch": 7.635983263598327,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011999149913996902,
      "loss": 4.1811,
      "mean_token_accuracy": 0.26468147337436676,
      "num_tokens": 16857500.0,
      "step": 9125
    },
    {
      "entropy": 4.8319501876831055,
      "epoch": 7.640167364016737,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011975770762755627,
      "loss": 4.0569,
      "mean_token_accuracy": 0.2773285284638405,
      "num_tokens": 16866329.0,
      "step": 9130
    },
    {
      "entropy": 4.880252313613892,
      "epoch": 7.644351464435147,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001195242355788221,
      "loss": 4.1563,
      "mean_token_accuracy": 0.2660258561372757,
      "num_tokens": 16875408.0,
      "step": 9135
    },
    {
      "entropy": 4.883033180236817,
      "epoch": 7.648535564853557,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011929108347421468,
      "loss": 4.2003,
      "mean_token_accuracy": 0.25708654522895813,
      "num_tokens": 16885272.0,
      "step": 9140
    },
    {
      "entropy": 4.825729751586914,
      "epoch": 7.652719665271967,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011905825179352337,
      "loss": 4.1426,
      "mean_token_accuracy": 0.26986626386642454,
      "num_tokens": 16894375.0,
      "step": 9145
    },
    {
      "entropy": 4.784857749938965,
      "epoch": 7.656903765690377,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011882574101587834,
      "loss": 4.1047,
      "mean_token_accuracy": 0.2733860224485397,
      "num_tokens": 16903790.0,
      "step": 9150
    },
    {
      "entropy": 4.831704139709473,
      "epoch": 7.661087866108787,
      "grad_norm": 1.5,
      "learning_rate": 0.00011859355161974948,
      "loss": 4.1432,
      "mean_token_accuracy": 0.26685127764940264,
      "num_tokens": 16913453.0,
      "step": 9155
    },
    {
      "entropy": 4.835260581970215,
      "epoch": 7.665271966527197,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011836168408294513,
      "loss": 4.1343,
      "mean_token_accuracy": 0.2660395398736,
      "num_tokens": 16922392.0,
      "step": 9160
    },
    {
      "entropy": 4.9047647476196286,
      "epoch": 7.669456066945607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011813013888261154,
      "loss": 4.2457,
      "mean_token_accuracy": 0.256240251660347,
      "num_tokens": 16931945.0,
      "step": 9165
    },
    {
      "entropy": 4.857575988769531,
      "epoch": 7.673640167364017,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011789891649523137,
      "loss": 4.1498,
      "mean_token_accuracy": 0.2619936212897301,
      "num_tokens": 16941285.0,
      "step": 9170
    },
    {
      "entropy": 4.879521751403809,
      "epoch": 7.677824267782427,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001176680173966232,
      "loss": 4.1524,
      "mean_token_accuracy": 0.26522725969552996,
      "num_tokens": 16950155.0,
      "step": 9175
    },
    {
      "entropy": 4.867926931381225,
      "epoch": 7.682008368200837,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011743744206194028,
      "loss": 4.2656,
      "mean_token_accuracy": 0.258167140185833,
      "num_tokens": 16959994.0,
      "step": 9180
    },
    {
      "entropy": 4.847810935974121,
      "epoch": 7.686192468619247,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011720719096566946,
      "loss": 4.1614,
      "mean_token_accuracy": 0.2653421118855476,
      "num_tokens": 16969095.0,
      "step": 9185
    },
    {
      "entropy": 4.897336626052857,
      "epoch": 7.690376569037657,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00011697726458163056,
      "loss": 4.2368,
      "mean_token_accuracy": 0.2578914284706116,
      "num_tokens": 16978574.0,
      "step": 9190
    },
    {
      "entropy": 4.899958515167237,
      "epoch": 7.694560669456067,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011674766338297508,
      "loss": 4.2266,
      "mean_token_accuracy": 0.26474476754665377,
      "num_tokens": 16988165.0,
      "step": 9195
    },
    {
      "entropy": 4.799305105209351,
      "epoch": 7.698744769874477,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011651838784218545,
      "loss": 4.128,
      "mean_token_accuracy": 0.2672067165374756,
      "num_tokens": 16998419.0,
      "step": 9200
    },
    {
      "entropy": 4.814904308319091,
      "epoch": 7.702928870292887,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011628943843107372,
      "loss": 4.1484,
      "mean_token_accuracy": 0.2647838518023491,
      "num_tokens": 17008276.0,
      "step": 9205
    },
    {
      "entropy": 4.871384000778198,
      "epoch": 7.707112970711297,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011606081562078113,
      "loss": 4.2123,
      "mean_token_accuracy": 0.2612835064530373,
      "num_tokens": 17017157.0,
      "step": 9210
    },
    {
      "entropy": 4.853985834121704,
      "epoch": 7.711297071129707,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011583251988177664,
      "loss": 4.1493,
      "mean_token_accuracy": 0.27031747698783876,
      "num_tokens": 17027100.0,
      "step": 9215
    },
    {
      "entropy": 4.810659503936767,
      "epoch": 7.7154811715481175,
      "grad_norm": 1.375,
      "learning_rate": 0.00011560455168385601,
      "loss": 4.1081,
      "mean_token_accuracy": 0.2637812554836273,
      "num_tokens": 17036094.0,
      "step": 9220
    },
    {
      "entropy": 4.851053047180176,
      "epoch": 7.7196652719665275,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011537691149614136,
      "loss": 4.118,
      "mean_token_accuracy": 0.2812513068318367,
      "num_tokens": 17044720.0,
      "step": 9225
    },
    {
      "entropy": 4.842622709274292,
      "epoch": 7.7238493723849375,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011514959978707944,
      "loss": 4.1871,
      "mean_token_accuracy": 0.2711881101131439,
      "num_tokens": 17053074.0,
      "step": 9230
    },
    {
      "entropy": 4.900670671463013,
      "epoch": 7.7280334728033475,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001149226170244413,
      "loss": 4.1965,
      "mean_token_accuracy": 0.2604396104812622,
      "num_tokens": 17063036.0,
      "step": 9235
    },
    {
      "entropy": 4.866956424713135,
      "epoch": 7.7322175732217575,
      "grad_norm": 1.5,
      "learning_rate": 0.0001146959636753209,
      "loss": 4.1591,
      "mean_token_accuracy": 0.2665217503905296,
      "num_tokens": 17072722.0,
      "step": 9240
    },
    {
      "entropy": 4.903783893585205,
      "epoch": 7.7364016736401675,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011446964020613448,
      "loss": 4.1701,
      "mean_token_accuracy": 0.2658463090658188,
      "num_tokens": 17081620.0,
      "step": 9245
    },
    {
      "entropy": 4.810437202453613,
      "epoch": 7.7405857740585775,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011424364708261931,
      "loss": 4.05,
      "mean_token_accuracy": 0.28733055889606474,
      "num_tokens": 17090628.0,
      "step": 9250
    },
    {
      "entropy": 4.90362401008606,
      "epoch": 7.7447698744769875,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011401798476983281,
      "loss": 4.2162,
      "mean_token_accuracy": 0.2652573108673096,
      "num_tokens": 17100089.0,
      "step": 9255
    },
    {
      "entropy": 4.860663175582886,
      "epoch": 7.7489539748953975,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00011379265373215189,
      "loss": 4.1491,
      "mean_token_accuracy": 0.26893729716539383,
      "num_tokens": 17108455.0,
      "step": 9260
    },
    {
      "entropy": 4.848731756210327,
      "epoch": 7.7531380753138075,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011356765443327146,
      "loss": 4.1582,
      "mean_token_accuracy": 0.27208707928657533,
      "num_tokens": 17117803.0,
      "step": 9265
    },
    {
      "entropy": 4.8879271984100345,
      "epoch": 7.7573221757322175,
      "grad_norm": 1.5,
      "learning_rate": 0.00011334298733620405,
      "loss": 4.1392,
      "mean_token_accuracy": 0.28006619662046434,
      "num_tokens": 17126436.0,
      "step": 9270
    },
    {
      "entropy": 4.8984729766845705,
      "epoch": 7.7615062761506275,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011311865290327827,
      "loss": 4.2405,
      "mean_token_accuracy": 0.25084921419620515,
      "num_tokens": 17135467.0,
      "step": 9275
    },
    {
      "entropy": 4.877915287017823,
      "epoch": 7.7656903765690375,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011289465159613849,
      "loss": 4.2041,
      "mean_token_accuracy": 0.2661248058080673,
      "num_tokens": 17145282.0,
      "step": 9280
    },
    {
      "entropy": 4.8812925815582275,
      "epoch": 7.7698744769874475,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011267098387574328,
      "loss": 4.2174,
      "mean_token_accuracy": 0.2646054789423943,
      "num_tokens": 17155003.0,
      "step": 9285
    },
    {
      "entropy": 4.86104998588562,
      "epoch": 7.7740585774058575,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00011244765020236477,
      "loss": 4.1396,
      "mean_token_accuracy": 0.26692577451467514,
      "num_tokens": 17163784.0,
      "step": 9290
    },
    {
      "entropy": 4.85145902633667,
      "epoch": 7.7782426778242675,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00011222465103558794,
      "loss": 4.1538,
      "mean_token_accuracy": 0.2652158305048943,
      "num_tokens": 17172991.0,
      "step": 9295
    },
    {
      "entropy": 4.849344253540039,
      "epoch": 7.7824267782426775,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011200198683430904,
      "loss": 4.1396,
      "mean_token_accuracy": 0.2656431034207344,
      "num_tokens": 17182035.0,
      "step": 9300
    },
    {
      "entropy": 4.812448358535766,
      "epoch": 7.786610878661088,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011177965805673528,
      "loss": 4.1019,
      "mean_token_accuracy": 0.2739919498562813,
      "num_tokens": 17191257.0,
      "step": 9305
    },
    {
      "entropy": 4.851333665847778,
      "epoch": 7.790794979079498,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011155766516038363,
      "loss": 4.1386,
      "mean_token_accuracy": 0.26525565832853315,
      "num_tokens": 17200792.0,
      "step": 9310
    },
    {
      "entropy": 4.848621129989624,
      "epoch": 7.794979079497908,
      "grad_norm": 1.515625,
      "learning_rate": 0.00011133600860207965,
      "loss": 4.2002,
      "mean_token_accuracy": 0.25173618346452714,
      "num_tokens": 17209833.0,
      "step": 9315
    },
    {
      "entropy": 4.845285701751709,
      "epoch": 7.799163179916318,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011111468883795698,
      "loss": 4.1413,
      "mean_token_accuracy": 0.2666185885667801,
      "num_tokens": 17218982.0,
      "step": 9320
    },
    {
      "entropy": 4.858426094055176,
      "epoch": 7.803347280334728,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011089370632345599,
      "loss": 4.1824,
      "mean_token_accuracy": 0.2646715983748436,
      "num_tokens": 17228557.0,
      "step": 9325
    },
    {
      "entropy": 4.874403715133667,
      "epoch": 7.807531380753138,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011067306151332331,
      "loss": 4.1847,
      "mean_token_accuracy": 0.2682981461286545,
      "num_tokens": 17238081.0,
      "step": 9330
    },
    {
      "entropy": 4.812588739395141,
      "epoch": 7.811715481171548,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011045275486161039,
      "loss": 4.1186,
      "mean_token_accuracy": 0.2741950049996376,
      "num_tokens": 17249000.0,
      "step": 9335
    },
    {
      "entropy": 4.845307159423828,
      "epoch": 7.815899581589958,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001102327868216729,
      "loss": 4.1864,
      "mean_token_accuracy": 0.2645310252904892,
      "num_tokens": 17259374.0,
      "step": 9340
    },
    {
      "entropy": 4.819796133041382,
      "epoch": 7.820083682008368,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00011001315784616988,
      "loss": 4.1221,
      "mean_token_accuracy": 0.26800217777490615,
      "num_tokens": 17268334.0,
      "step": 9345
    },
    {
      "entropy": 4.872894763946533,
      "epoch": 7.824267782426778,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001097938683870622,
      "loss": 4.1132,
      "mean_token_accuracy": 0.26874651163816454,
      "num_tokens": 17277364.0,
      "step": 9350
    },
    {
      "entropy": 4.8521984100341795,
      "epoch": 7.828451882845188,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010957491889561244,
      "loss": 4.12,
      "mean_token_accuracy": 0.2657155245542526,
      "num_tokens": 17286552.0,
      "step": 9355
    },
    {
      "entropy": 4.8429419040679935,
      "epoch": 7.832635983263598,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001093563098223834,
      "loss": 4.1774,
      "mean_token_accuracy": 0.2630805715918541,
      "num_tokens": 17297114.0,
      "step": 9360
    },
    {
      "entropy": 4.884470653533936,
      "epoch": 7.836820083682008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010913804161723742,
      "loss": 4.1887,
      "mean_token_accuracy": 0.265458083152771,
      "num_tokens": 17306469.0,
      "step": 9365
    },
    {
      "entropy": 4.909255743026733,
      "epoch": 7.841004184100418,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010892011472933547,
      "loss": 4.1938,
      "mean_token_accuracy": 0.2699979841709137,
      "num_tokens": 17315790.0,
      "step": 9370
    },
    {
      "entropy": 4.887000703811646,
      "epoch": 7.845188284518828,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010870252960713587,
      "loss": 4.2367,
      "mean_token_accuracy": 0.25583274513483045,
      "num_tokens": 17324803.0,
      "step": 9375
    },
    {
      "entropy": 4.87223711013794,
      "epoch": 7.849372384937238,
      "grad_norm": 1.25,
      "learning_rate": 0.00010848528669839409,
      "loss": 4.2203,
      "mean_token_accuracy": 0.26636188477277756,
      "num_tokens": 17333856.0,
      "step": 9380
    },
    {
      "entropy": 4.812434291839599,
      "epoch": 7.853556485355648,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010826838645016077,
      "loss": 4.0583,
      "mean_token_accuracy": 0.2741051882505417,
      "num_tokens": 17343399.0,
      "step": 9385
    },
    {
      "entropy": 4.839921617507935,
      "epoch": 7.857740585774058,
      "grad_norm": 1.5,
      "learning_rate": 0.00010805182930878191,
      "loss": 4.0833,
      "mean_token_accuracy": 0.27092849314212797,
      "num_tokens": 17352547.0,
      "step": 9390
    },
    {
      "entropy": 4.890019702911377,
      "epoch": 7.861924686192468,
      "grad_norm": 1.25,
      "learning_rate": 0.00010783561571989739,
      "loss": 4.1667,
      "mean_token_accuracy": 0.2649814262986183,
      "num_tokens": 17362507.0,
      "step": 9395
    },
    {
      "entropy": 4.822741746902466,
      "epoch": 7.866108786610878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010761974612843984,
      "loss": 4.0804,
      "mean_token_accuracy": 0.2789373084902763,
      "num_tokens": 17372133.0,
      "step": 9400
    },
    {
      "entropy": 4.894700241088867,
      "epoch": 7.870292887029288,
      "grad_norm": 1.28125,
      "learning_rate": 0.00010740422097863431,
      "loss": 4.2198,
      "mean_token_accuracy": 0.2555113911628723,
      "num_tokens": 17383178.0,
      "step": 9405
    },
    {
      "entropy": 4.876405477523804,
      "epoch": 7.874476987447698,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001071890407139968,
      "loss": 4.2209,
      "mean_token_accuracy": 0.2598504602909088,
      "num_tokens": 17392357.0,
      "step": 9410
    },
    {
      "entropy": 4.910240650177002,
      "epoch": 7.878661087866108,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010697420577733385,
      "loss": 4.1948,
      "mean_token_accuracy": 0.26959356367588044,
      "num_tokens": 17401101.0,
      "step": 9415
    },
    {
      "entropy": 4.808073568344116,
      "epoch": 7.882845188284519,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00010675971661074102,
      "loss": 4.044,
      "mean_token_accuracy": 0.28112886250019076,
      "num_tokens": 17409909.0,
      "step": 9420
    },
    {
      "entropy": 4.905235719680786,
      "epoch": 7.887029288702929,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00010654557365560261,
      "loss": 4.2274,
      "mean_token_accuracy": 0.26628507375717164,
      "num_tokens": 17419631.0,
      "step": 9425
    },
    {
      "entropy": 4.886778545379639,
      "epoch": 7.891213389121339,
      "grad_norm": 1.515625,
      "learning_rate": 0.00010633177735259047,
      "loss": 4.2161,
      "mean_token_accuracy": 0.2576869174838066,
      "num_tokens": 17428557.0,
      "step": 9430
    },
    {
      "entropy": 4.8537639617919925,
      "epoch": 7.895397489539749,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010611832814166292,
      "loss": 4.149,
      "mean_token_accuracy": 0.2656896740198135,
      "num_tokens": 17438442.0,
      "step": 9435
    },
    {
      "entropy": 4.884250926971435,
      "epoch": 7.899581589958159,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00010590522646206419,
      "loss": 4.1791,
      "mean_token_accuracy": 0.26497264206409454,
      "num_tokens": 17448116.0,
      "step": 9440
    },
    {
      "entropy": 4.827585601806641,
      "epoch": 7.903765690376569,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010569247275232328,
      "loss": 4.1026,
      "mean_token_accuracy": 0.2715082108974457,
      "num_tokens": 17456823.0,
      "step": 9445
    },
    {
      "entropy": 4.8680174350738525,
      "epoch": 7.907949790794979,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010548006745025298,
      "loss": 4.1536,
      "mean_token_accuracy": 0.2710878700017929,
      "num_tokens": 17466313.0,
      "step": 9450
    },
    {
      "entropy": 4.889216423034668,
      "epoch": 7.912133891213389,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010526801099294947,
      "loss": 4.2644,
      "mean_token_accuracy": 0.2634760469198227,
      "num_tokens": 17477448.0,
      "step": 9455
    },
    {
      "entropy": 4.922382116317749,
      "epoch": 7.916317991631799,
      "grad_norm": 1.375,
      "learning_rate": 0.00010505630381679063,
      "loss": 4.2292,
      "mean_token_accuracy": 0.26556815654039384,
      "num_tokens": 17487587.0,
      "step": 9460
    },
    {
      "entropy": 4.867498779296875,
      "epoch": 7.920502092050209,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010484494635743598,
      "loss": 4.1634,
      "mean_token_accuracy": 0.2647510409355164,
      "num_tokens": 17496282.0,
      "step": 9465
    },
    {
      "entropy": 4.887860918045044,
      "epoch": 7.924686192468619,
      "grad_norm": 1.4375,
      "learning_rate": 0.000104633939049825,
      "loss": 4.2149,
      "mean_token_accuracy": 0.2553803905844688,
      "num_tokens": 17505468.0,
      "step": 9470
    },
    {
      "entropy": 4.90014615058899,
      "epoch": 7.928870292887029,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010442328232817702,
      "loss": 4.1537,
      "mean_token_accuracy": 0.26632860153913496,
      "num_tokens": 17515743.0,
      "step": 9475
    },
    {
      "entropy": 4.912510681152344,
      "epoch": 7.933054393305439,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010421297662598955,
      "loss": 4.1159,
      "mean_token_accuracy": 0.27275232374668124,
      "num_tokens": 17524350.0,
      "step": 9480
    },
    {
      "entropy": 4.85536298751831,
      "epoch": 7.937238493723849,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010400302237603793,
      "loss": 4.1114,
      "mean_token_accuracy": 0.26450660526752473,
      "num_tokens": 17533439.0,
      "step": 9485
    },
    {
      "entropy": 4.866399240493775,
      "epoch": 7.941422594142259,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010379342001037431,
      "loss": 4.1277,
      "mean_token_accuracy": 0.26386686861515046,
      "num_tokens": 17542782.0,
      "step": 9490
    },
    {
      "entropy": 4.875611305236816,
      "epoch": 7.945606694560669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010358416996032658,
      "loss": 4.176,
      "mean_token_accuracy": 0.2690185859799385,
      "num_tokens": 17552495.0,
      "step": 9495
    },
    {
      "entropy": 4.838320875167847,
      "epoch": 7.949790794979079,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00010337527265649782,
      "loss": 4.1599,
      "mean_token_accuracy": 0.2611872136592865,
      "num_tokens": 17560933.0,
      "step": 9500
    },
    {
      "epoch": 7.949790794979079,
      "eval_entropy": 4.729641100136536,
      "eval_loss": 5.312812328338623,
      "eval_mean_token_accuracy": 0.20367083044871231,
      "eval_num_tokens": 17560933.0,
      "eval_runtime": 2.7575,
      "eval_samples_per_second": 1416.885,
      "eval_steps_per_second": 177.337,
      "step": 9500
    },
    {
      "entropy": 4.87229642868042,
      "epoch": 7.95397489539749,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010316672852876494,
      "loss": 4.1952,
      "mean_token_accuracy": 0.26640831530094145,
      "num_tokens": 17571077.0,
      "step": 9505
    },
    {
      "entropy": 4.905666828155518,
      "epoch": 7.9581589958159,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010295853800627836,
      "loss": 4.1381,
      "mean_token_accuracy": 0.2713698551058769,
      "num_tokens": 17580128.0,
      "step": 9510
    },
    {
      "entropy": 4.913319826126099,
      "epoch": 7.96234309623431,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010275070151746069,
      "loss": 4.2255,
      "mean_token_accuracy": 0.26101440042257307,
      "num_tokens": 17589811.0,
      "step": 9515
    },
    {
      "entropy": 4.8879022121429445,
      "epoch": 7.96652719665272,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010254321949000587,
      "loss": 4.1931,
      "mean_token_accuracy": 0.2617680221796036,
      "num_tokens": 17599018.0,
      "step": 9520
    },
    {
      "entropy": 4.854970645904541,
      "epoch": 7.97071129707113,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010233609235087874,
      "loss": 4.1109,
      "mean_token_accuracy": 0.26325787156820296,
      "num_tokens": 17607416.0,
      "step": 9525
    },
    {
      "entropy": 4.8475573539733885,
      "epoch": 7.97489539748954,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010212932052631351,
      "loss": 4.1954,
      "mean_token_accuracy": 0.2659373477101326,
      "num_tokens": 17616674.0,
      "step": 9530
    },
    {
      "entropy": 4.859478712081909,
      "epoch": 7.97907949790795,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010192290444181347,
      "loss": 4.1518,
      "mean_token_accuracy": 0.2673717737197876,
      "num_tokens": 17626099.0,
      "step": 9535
    },
    {
      "entropy": 4.836061668395996,
      "epoch": 7.98326359832636,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010171684452214959,
      "loss": 4.1466,
      "mean_token_accuracy": 0.2661234512925148,
      "num_tokens": 17636013.0,
      "step": 9540
    },
    {
      "entropy": 4.894927978515625,
      "epoch": 7.98744769874477,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010151114119136024,
      "loss": 4.1876,
      "mean_token_accuracy": 0.2630405753850937,
      "num_tokens": 17645849.0,
      "step": 9545
    },
    {
      "entropy": 4.885028266906739,
      "epoch": 7.99163179916318,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010130579487274969,
      "loss": 4.2104,
      "mean_token_accuracy": 0.26138140708208085,
      "num_tokens": 17655557.0,
      "step": 9550
    },
    {
      "entropy": 4.891262531280518,
      "epoch": 7.99581589958159,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001011008059888876,
      "loss": 4.2295,
      "mean_token_accuracy": 0.2632304608821869,
      "num_tokens": 17665717.0,
      "step": 9555
    },
    {
      "entropy": 4.820310306549072,
      "epoch": 8.0,
      "grad_norm": 1.4375,
      "learning_rate": 0.00010089617496160825,
      "loss": 4.1038,
      "mean_token_accuracy": 0.2660931944847107,
      "num_tokens": 17674576.0,
      "step": 9560
    },
    {
      "entropy": 4.855773401260376,
      "epoch": 8.00418410041841,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010069190221200929,
      "loss": 4.0337,
      "mean_token_accuracy": 0.2871698349714279,
      "num_tokens": 17684147.0,
      "step": 9565
    },
    {
      "entropy": 4.778929233551025,
      "epoch": 8.00836820083682,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010048798816045132,
      "loss": 3.9854,
      "mean_token_accuracy": 0.28262548595666886,
      "num_tokens": 17693082.0,
      "step": 9570
    },
    {
      "entropy": 4.7858538150787355,
      "epoch": 8.01255230125523,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010028443322655652,
      "loss": 4.0367,
      "mean_token_accuracy": 0.2799614518880844,
      "num_tokens": 17702149.0,
      "step": 9575
    },
    {
      "entropy": 4.8438887119293215,
      "epoch": 8.01673640167364,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010008123782920836,
      "loss": 4.0388,
      "mean_token_accuracy": 0.28587919771671294,
      "num_tokens": 17710666.0,
      "step": 9580
    },
    {
      "entropy": 4.860464382171631,
      "epoch": 8.02092050209205,
      "grad_norm": 1.46875,
      "learning_rate": 9.987840238655026e-05,
      "loss": 4.0918,
      "mean_token_accuracy": 0.2757610589265823,
      "num_tokens": 17719643.0,
      "step": 9585
    },
    {
      "entropy": 4.9195075035095215,
      "epoch": 8.02510460251046,
      "grad_norm": 1.375,
      "learning_rate": 9.967592731598487e-05,
      "loss": 4.1477,
      "mean_token_accuracy": 0.2713066071271896,
      "num_tokens": 17728714.0,
      "step": 9590
    },
    {
      "entropy": 4.9525594234466555,
      "epoch": 8.02928870292887,
      "grad_norm": 1.421875,
      "learning_rate": 9.947381303417346e-05,
      "loss": 4.1108,
      "mean_token_accuracy": 0.27403611689805984,
      "num_tokens": 17736496.0,
      "step": 9595
    },
    {
      "entropy": 4.917904710769653,
      "epoch": 8.03347280334728,
      "grad_norm": 1.4375,
      "learning_rate": 9.927205995703456e-05,
      "loss": 4.1478,
      "mean_token_accuracy": 0.2689757362008095,
      "num_tokens": 17745909.0,
      "step": 9600
    },
    {
      "entropy": 4.871824073791504,
      "epoch": 8.03765690376569,
      "grad_norm": 1.3046875,
      "learning_rate": 9.907066849974372e-05,
      "loss": 4.0295,
      "mean_token_accuracy": 0.2816247045993805,
      "num_tokens": 17755125.0,
      "step": 9605
    },
    {
      "entropy": 4.850768089294434,
      "epoch": 8.0418410041841,
      "grad_norm": 1.46875,
      "learning_rate": 9.886963907673214e-05,
      "loss": 4.055,
      "mean_token_accuracy": 0.27724531292915344,
      "num_tokens": 17764253.0,
      "step": 9610
    },
    {
      "entropy": 4.834056758880616,
      "epoch": 8.04602510460251,
      "grad_norm": 1.4609375,
      "learning_rate": 9.866897210168608e-05,
      "loss": 4.1071,
      "mean_token_accuracy": 0.27122982144355773,
      "num_tokens": 17773485.0,
      "step": 9615
    },
    {
      "entropy": 4.900292444229126,
      "epoch": 8.05020920502092,
      "grad_norm": 1.6171875,
      "learning_rate": 9.846866798754583e-05,
      "loss": 4.1174,
      "mean_token_accuracy": 0.27364901155233384,
      "num_tokens": 17781942.0,
      "step": 9620
    },
    {
      "entropy": 4.878519439697266,
      "epoch": 8.05439330543933,
      "grad_norm": 1.515625,
      "learning_rate": 9.826872714650509e-05,
      "loss": 4.0519,
      "mean_token_accuracy": 0.2778239533305168,
      "num_tokens": 17791576.0,
      "step": 9625
    },
    {
      "entropy": 4.9013450145721436,
      "epoch": 8.05857740585774,
      "grad_norm": 1.3828125,
      "learning_rate": 9.806914999000998e-05,
      "loss": 4.1113,
      "mean_token_accuracy": 0.2697569772601128,
      "num_tokens": 17801439.0,
      "step": 9630
    },
    {
      "entropy": 4.894170761108398,
      "epoch": 8.06276150627615,
      "grad_norm": 1.4453125,
      "learning_rate": 9.786993692875814e-05,
      "loss": 4.1318,
      "mean_token_accuracy": 0.26804340034723284,
      "num_tokens": 17810523.0,
      "step": 9635
    },
    {
      "entropy": 4.829508876800537,
      "epoch": 8.06694560669456,
      "grad_norm": 1.421875,
      "learning_rate": 9.767108837269808e-05,
      "loss": 4.0437,
      "mean_token_accuracy": 0.274729785323143,
      "num_tokens": 17819050.0,
      "step": 9640
    },
    {
      "entropy": 4.884012365341187,
      "epoch": 8.07112970711297,
      "grad_norm": 1.453125,
      "learning_rate": 9.747260473102829e-05,
      "loss": 4.1137,
      "mean_token_accuracy": 0.268721704185009,
      "num_tokens": 17828215.0,
      "step": 9645
    },
    {
      "entropy": 4.833860588073731,
      "epoch": 8.07531380753138,
      "grad_norm": 1.359375,
      "learning_rate": 9.727448641219591e-05,
      "loss": 4.0851,
      "mean_token_accuracy": 0.27220748364925385,
      "num_tokens": 17837091.0,
      "step": 9650
    },
    {
      "entropy": 4.846099376678467,
      "epoch": 8.07949790794979,
      "grad_norm": 1.4609375,
      "learning_rate": 9.707673382389678e-05,
      "loss": 4.0987,
      "mean_token_accuracy": 0.280666384100914,
      "num_tokens": 17845647.0,
      "step": 9655
    },
    {
      "entropy": 4.895712184906006,
      "epoch": 8.0836820083682,
      "grad_norm": 1.4609375,
      "learning_rate": 9.687934737307386e-05,
      "loss": 4.066,
      "mean_token_accuracy": 0.2752397254109383,
      "num_tokens": 17854096.0,
      "step": 9660
    },
    {
      "entropy": 4.9000965595245365,
      "epoch": 8.087866108786612,
      "grad_norm": 1.4296875,
      "learning_rate": 9.668232746591671e-05,
      "loss": 4.0754,
      "mean_token_accuracy": 0.27713173180818557,
      "num_tokens": 17863239.0,
      "step": 9665
    },
    {
      "entropy": 4.855951499938965,
      "epoch": 8.092050209205022,
      "grad_norm": 1.4296875,
      "learning_rate": 9.648567450786077e-05,
      "loss": 4.0696,
      "mean_token_accuracy": 0.2713649332523346,
      "num_tokens": 17872447.0,
      "step": 9670
    },
    {
      "entropy": 4.799465847015381,
      "epoch": 8.096234309623432,
      "grad_norm": 1.4140625,
      "learning_rate": 9.628938890358604e-05,
      "loss": 4.051,
      "mean_token_accuracy": 0.2743504136800766,
      "num_tokens": 17881862.0,
      "step": 9675
    },
    {
      "entropy": 4.807206392288208,
      "epoch": 8.100418410041842,
      "grad_norm": 1.328125,
      "learning_rate": 9.609347105701691e-05,
      "loss": 4.0987,
      "mean_token_accuracy": 0.26990685909986495,
      "num_tokens": 17891683.0,
      "step": 9680
    },
    {
      "entropy": 4.884832715988159,
      "epoch": 8.104602510460252,
      "grad_norm": 1.3828125,
      "learning_rate": 9.589792137132061e-05,
      "loss": 4.1275,
      "mean_token_accuracy": 0.2744657501578331,
      "num_tokens": 17901105.0,
      "step": 9685
    },
    {
      "entropy": 4.799909400939941,
      "epoch": 8.108786610878662,
      "grad_norm": 1.4765625,
      "learning_rate": 9.570274024890704e-05,
      "loss": 4.0507,
      "mean_token_accuracy": 0.28154650032520295,
      "num_tokens": 17910761.0,
      "step": 9690
    },
    {
      "entropy": 4.899988746643066,
      "epoch": 8.112970711297072,
      "grad_norm": 1.359375,
      "learning_rate": 9.550792809142764e-05,
      "loss": 4.0904,
      "mean_token_accuracy": 0.27681459337472913,
      "num_tokens": 17919917.0,
      "step": 9695
    },
    {
      "entropy": 4.912911701202392,
      "epoch": 8.117154811715482,
      "grad_norm": 1.5,
      "learning_rate": 9.531348529977439e-05,
      "loss": 4.148,
      "mean_token_accuracy": 0.2663514018058777,
      "num_tokens": 17928588.0,
      "step": 9700
    },
    {
      "entropy": 4.896832084655761,
      "epoch": 8.121338912133892,
      "grad_norm": 1.4609375,
      "learning_rate": 9.51194122740794e-05,
      "loss": 4.0856,
      "mean_token_accuracy": 0.28034937381744385,
      "num_tokens": 17938142.0,
      "step": 9705
    },
    {
      "entropy": 4.830137443542481,
      "epoch": 8.125523012552302,
      "grad_norm": 1.359375,
      "learning_rate": 9.492570941371365e-05,
      "loss": 4.0215,
      "mean_token_accuracy": 0.2756694495677948,
      "num_tokens": 17947676.0,
      "step": 9710
    },
    {
      "entropy": 4.8204316139221195,
      "epoch": 8.129707112970712,
      "grad_norm": 1.3828125,
      "learning_rate": 9.47323771172866e-05,
      "loss": 4.0843,
      "mean_token_accuracy": 0.2754463732242584,
      "num_tokens": 17957166.0,
      "step": 9715
    },
    {
      "entropy": 4.879590463638306,
      "epoch": 8.133891213389122,
      "grad_norm": 1.5,
      "learning_rate": 9.453941578264499e-05,
      "loss": 4.1201,
      "mean_token_accuracy": 0.2758794605731964,
      "num_tokens": 17965592.0,
      "step": 9720
    },
    {
      "entropy": 4.794753360748291,
      "epoch": 8.138075313807532,
      "grad_norm": 1.4765625,
      "learning_rate": 9.434682580687215e-05,
      "loss": 4.0561,
      "mean_token_accuracy": 0.2794376015663147,
      "num_tokens": 17973949.0,
      "step": 9725
    },
    {
      "entropy": 4.81717882156372,
      "epoch": 8.142259414225942,
      "grad_norm": 1.4296875,
      "learning_rate": 9.415460758628739e-05,
      "loss": 4.067,
      "mean_token_accuracy": 0.2732755482196808,
      "num_tokens": 17982540.0,
      "step": 9730
    },
    {
      "entropy": 4.814676332473755,
      "epoch": 8.146443514644352,
      "grad_norm": 1.359375,
      "learning_rate": 9.396276151644478e-05,
      "loss": 4.0634,
      "mean_token_accuracy": 0.2797649011015892,
      "num_tokens": 17991611.0,
      "step": 9735
    },
    {
      "entropy": 4.8776599884033205,
      "epoch": 8.150627615062762,
      "grad_norm": 1.546875,
      "learning_rate": 9.377128799213286e-05,
      "loss": 4.0434,
      "mean_token_accuracy": 0.27923952639102934,
      "num_tokens": 17999462.0,
      "step": 9740
    },
    {
      "entropy": 4.900128746032715,
      "epoch": 8.154811715481172,
      "grad_norm": 1.296875,
      "learning_rate": 9.358018740737325e-05,
      "loss": 4.179,
      "mean_token_accuracy": 0.27440220564603807,
      "num_tokens": 18009693.0,
      "step": 9745
    },
    {
      "entropy": 4.806037139892578,
      "epoch": 8.158995815899582,
      "grad_norm": 1.2890625,
      "learning_rate": 9.338946015542018e-05,
      "loss": 3.989,
      "mean_token_accuracy": 0.28581259250640867,
      "num_tokens": 18019101.0,
      "step": 9750
    },
    {
      "entropy": 4.8574442863464355,
      "epoch": 8.163179916317992,
      "grad_norm": 1.53125,
      "learning_rate": 9.31991066287598e-05,
      "loss": 4.1947,
      "mean_token_accuracy": 0.26114443093538287,
      "num_tokens": 18028198.0,
      "step": 9755
    },
    {
      "entropy": 4.907324934005738,
      "epoch": 8.167364016736402,
      "grad_norm": 1.296875,
      "learning_rate": 9.300912721910894e-05,
      "loss": 4.1526,
      "mean_token_accuracy": 0.27116800248622897,
      "num_tokens": 18037592.0,
      "step": 9760
    },
    {
      "entropy": 4.875959634780884,
      "epoch": 8.171548117154812,
      "grad_norm": 1.4609375,
      "learning_rate": 9.28195223174148e-05,
      "loss": 4.1214,
      "mean_token_accuracy": 0.271700918674469,
      "num_tokens": 18046988.0,
      "step": 9765
    },
    {
      "entropy": 4.795725917816162,
      "epoch": 8.175732217573222,
      "grad_norm": 1.3515625,
      "learning_rate": 9.263029231385366e-05,
      "loss": 4.0256,
      "mean_token_accuracy": 0.2877350777387619,
      "num_tokens": 18056322.0,
      "step": 9770
    },
    {
      "entropy": 4.808853530883789,
      "epoch": 8.179916317991632,
      "grad_norm": 1.3515625,
      "learning_rate": 9.244143759783058e-05,
      "loss": 4.017,
      "mean_token_accuracy": 0.2852470874786377,
      "num_tokens": 18065157.0,
      "step": 9775
    },
    {
      "entropy": 4.9326530456542965,
      "epoch": 8.184100418410042,
      "grad_norm": 1.4375,
      "learning_rate": 9.225295855797812e-05,
      "loss": 4.1462,
      "mean_token_accuracy": 0.27080668807029723,
      "num_tokens": 18074389.0,
      "step": 9780
    },
    {
      "entropy": 4.896804285049439,
      "epoch": 8.188284518828452,
      "grad_norm": 1.46875,
      "learning_rate": 9.20648555821558e-05,
      "loss": 4.0756,
      "mean_token_accuracy": 0.27991951406002047,
      "num_tokens": 18083434.0,
      "step": 9785
    },
    {
      "entropy": 4.817349767684936,
      "epoch": 8.192468619246862,
      "grad_norm": 1.15625,
      "learning_rate": 9.18771290574494e-05,
      "loss": 4.0529,
      "mean_token_accuracy": 0.27355436980724335,
      "num_tokens": 18094145.0,
      "step": 9790
    },
    {
      "entropy": 4.863154411315918,
      "epoch": 8.196652719665272,
      "grad_norm": 1.34375,
      "learning_rate": 9.168977937016981e-05,
      "loss": 4.1115,
      "mean_token_accuracy": 0.27352896332740784,
      "num_tokens": 18103707.0,
      "step": 9795
    },
    {
      "entropy": 4.814683628082276,
      "epoch": 8.200836820083682,
      "grad_norm": 1.40625,
      "learning_rate": 9.150280690585271e-05,
      "loss": 4.0695,
      "mean_token_accuracy": 0.2786334067583084,
      "num_tokens": 18113416.0,
      "step": 9800
    },
    {
      "entropy": 4.797551393508911,
      "epoch": 8.205020920502092,
      "grad_norm": 1.375,
      "learning_rate": 9.131621204925724e-05,
      "loss": 4.043,
      "mean_token_accuracy": 0.26871120631694795,
      "num_tokens": 18122809.0,
      "step": 9805
    },
    {
      "entropy": 4.908729171752929,
      "epoch": 8.209205020920502,
      "grad_norm": 1.40625,
      "learning_rate": 9.112999518436574e-05,
      "loss": 4.1748,
      "mean_token_accuracy": 0.25636289417743685,
      "num_tokens": 18132518.0,
      "step": 9810
    },
    {
      "entropy": 4.880013751983642,
      "epoch": 8.213389121338912,
      "grad_norm": 1.4375,
      "learning_rate": 9.09441566943825e-05,
      "loss": 4.1178,
      "mean_token_accuracy": 0.26793241649866106,
      "num_tokens": 18141499.0,
      "step": 9815
    },
    {
      "entropy": 4.8772594928741455,
      "epoch": 8.217573221757322,
      "grad_norm": 1.5625,
      "learning_rate": 9.075869696173321e-05,
      "loss": 4.0977,
      "mean_token_accuracy": 0.27343538850545884,
      "num_tokens": 18150330.0,
      "step": 9820
    },
    {
      "entropy": 4.856595134735107,
      "epoch": 8.221757322175732,
      "grad_norm": 1.484375,
      "learning_rate": 9.057361636806428e-05,
      "loss": 4.0815,
      "mean_token_accuracy": 0.27211769819259646,
      "num_tokens": 18158925.0,
      "step": 9825
    },
    {
      "entropy": 4.841387033462524,
      "epoch": 8.225941422594142,
      "grad_norm": 1.3984375,
      "learning_rate": 9.038891529424173e-05,
      "loss": 4.0553,
      "mean_token_accuracy": 0.27836519479751587,
      "num_tokens": 18167612.0,
      "step": 9830
    },
    {
      "entropy": 4.887151050567627,
      "epoch": 8.230125523012552,
      "grad_norm": 1.4453125,
      "learning_rate": 9.020459412035075e-05,
      "loss": 4.0779,
      "mean_token_accuracy": 0.2778461843729019,
      "num_tokens": 18176727.0,
      "step": 9835
    },
    {
      "entropy": 4.884986209869385,
      "epoch": 8.234309623430962,
      "grad_norm": 1.4609375,
      "learning_rate": 9.002065322569464e-05,
      "loss": 4.0734,
      "mean_token_accuracy": 0.2777773171663284,
      "num_tokens": 18186542.0,
      "step": 9840
    },
    {
      "entropy": 4.834874391555786,
      "epoch": 8.238493723849372,
      "grad_norm": 1.5234375,
      "learning_rate": 8.983709298879424e-05,
      "loss": 4.0719,
      "mean_token_accuracy": 0.27270805686712263,
      "num_tokens": 18195514.0,
      "step": 9845
    },
    {
      "entropy": 4.830412530899048,
      "epoch": 8.242677824267782,
      "grad_norm": 1.3828125,
      "learning_rate": 8.965391378738692e-05,
      "loss": 4.0732,
      "mean_token_accuracy": 0.2700021222233772,
      "num_tokens": 18204481.0,
      "step": 9850
    },
    {
      "entropy": 4.89267897605896,
      "epoch": 8.246861924686192,
      "grad_norm": 1.3984375,
      "learning_rate": 8.947111599842606e-05,
      "loss": 4.072,
      "mean_token_accuracy": 0.2800348311662674,
      "num_tokens": 18213342.0,
      "step": 9855
    },
    {
      "entropy": 4.887956523895264,
      "epoch": 8.251046025104603,
      "grad_norm": 1.4375,
      "learning_rate": 8.928869999808017e-05,
      "loss": 4.1331,
      "mean_token_accuracy": 0.2714578866958618,
      "num_tokens": 18223083.0,
      "step": 9860
    },
    {
      "entropy": 4.835611724853516,
      "epoch": 8.255230125523013,
      "grad_norm": 1.4296875,
      "learning_rate": 8.910666616173196e-05,
      "loss": 4.0577,
      "mean_token_accuracy": 0.2757794424891472,
      "num_tokens": 18231792.0,
      "step": 9865
    },
    {
      "entropy": 4.857382345199585,
      "epoch": 8.259414225941423,
      "grad_norm": 1.4921875,
      "learning_rate": 8.89250148639779e-05,
      "loss": 4.0907,
      "mean_token_accuracy": 0.2777420789003372,
      "num_tokens": 18240099.0,
      "step": 9870
    },
    {
      "entropy": 4.848942518234253,
      "epoch": 8.263598326359833,
      "grad_norm": 1.4609375,
      "learning_rate": 8.874374647862704e-05,
      "loss": 4.1019,
      "mean_token_accuracy": 0.26991742551326753,
      "num_tokens": 18249237.0,
      "step": 9875
    },
    {
      "entropy": 4.872313547134399,
      "epoch": 8.267782426778243,
      "grad_norm": 1.453125,
      "learning_rate": 8.856286137870068e-05,
      "loss": 4.1082,
      "mean_token_accuracy": 0.272709047794342,
      "num_tokens": 18257896.0,
      "step": 9880
    },
    {
      "entropy": 4.8599803924560545,
      "epoch": 8.271966527196653,
      "grad_norm": 1.484375,
      "learning_rate": 8.838235993643125e-05,
      "loss": 4.0515,
      "mean_token_accuracy": 0.27912245988845824,
      "num_tokens": 18266251.0,
      "step": 9885
    },
    {
      "entropy": 4.861873435974121,
      "epoch": 8.276150627615063,
      "grad_norm": 1.3671875,
      "learning_rate": 8.820224252326161e-05,
      "loss": 4.0626,
      "mean_token_accuracy": 0.2780997663736343,
      "num_tokens": 18275729.0,
      "step": 9890
    },
    {
      "entropy": 4.924075174331665,
      "epoch": 8.280334728033473,
      "grad_norm": 1.421875,
      "learning_rate": 8.802250950984461e-05,
      "loss": 4.1443,
      "mean_token_accuracy": 0.269752436876297,
      "num_tokens": 18284625.0,
      "step": 9895
    },
    {
      "entropy": 4.808215808868408,
      "epoch": 8.284518828451883,
      "grad_norm": 1.3984375,
      "learning_rate": 8.784316126604169e-05,
      "loss": 4.0171,
      "mean_token_accuracy": 0.2875951319932938,
      "num_tokens": 18293390.0,
      "step": 9900
    },
    {
      "entropy": 4.836294317245484,
      "epoch": 8.288702928870293,
      "grad_norm": 1.34375,
      "learning_rate": 8.766419816092284e-05,
      "loss": 4.0501,
      "mean_token_accuracy": 0.27825399935245515,
      "num_tokens": 18302955.0,
      "step": 9905
    },
    {
      "entropy": 4.865026617050171,
      "epoch": 8.292887029288703,
      "grad_norm": 1.5546875,
      "learning_rate": 8.748562056276535e-05,
      "loss": 4.119,
      "mean_token_accuracy": 0.27067701816558837,
      "num_tokens": 18311614.0,
      "step": 9910
    },
    {
      "entropy": 4.889970159530639,
      "epoch": 8.297071129707113,
      "grad_norm": 1.3515625,
      "learning_rate": 8.730742883905321e-05,
      "loss": 4.0928,
      "mean_token_accuracy": 0.27779437601566315,
      "num_tokens": 18320588.0,
      "step": 9915
    },
    {
      "entropy": 4.808910846710205,
      "epoch": 8.301255230125523,
      "grad_norm": 1.3359375,
      "learning_rate": 8.712962335647629e-05,
      "loss": 3.9827,
      "mean_token_accuracy": 0.2875602304935455,
      "num_tokens": 18329976.0,
      "step": 9920
    },
    {
      "entropy": 4.8472771644592285,
      "epoch": 8.305439330543933,
      "grad_norm": 1.484375,
      "learning_rate": 8.695220448092964e-05,
      "loss": 4.1054,
      "mean_token_accuracy": 0.27041110396385193,
      "num_tokens": 18338495.0,
      "step": 9925
    },
    {
      "entropy": 4.855547332763672,
      "epoch": 8.309623430962343,
      "grad_norm": 1.390625,
      "learning_rate": 8.677517257751286e-05,
      "loss": 4.1059,
      "mean_token_accuracy": 0.2736356511712074,
      "num_tokens": 18348212.0,
      "step": 9930
    },
    {
      "entropy": 4.7582862854003904,
      "epoch": 8.313807531380753,
      "grad_norm": 1.46875,
      "learning_rate": 8.659852801052915e-05,
      "loss": 3.9643,
      "mean_token_accuracy": 0.2846452325582504,
      "num_tokens": 18356803.0,
      "step": 9935
    },
    {
      "entropy": 4.9080237865448,
      "epoch": 8.317991631799163,
      "grad_norm": 1.28125,
      "learning_rate": 8.64222711434846e-05,
      "loss": 4.2149,
      "mean_token_accuracy": 0.26953286230564116,
      "num_tokens": 18366822.0,
      "step": 9940
    },
    {
      "entropy": 4.88466796875,
      "epoch": 8.322175732217573,
      "grad_norm": 1.3515625,
      "learning_rate": 8.62464023390876e-05,
      "loss": 4.1056,
      "mean_token_accuracy": 0.26902188211679456,
      "num_tokens": 18376495.0,
      "step": 9945
    },
    {
      "entropy": 4.796904897689819,
      "epoch": 8.326359832635983,
      "grad_norm": 1.453125,
      "learning_rate": 8.607092195924768e-05,
      "loss": 3.9896,
      "mean_token_accuracy": 0.2918999820947647,
      "num_tokens": 18385652.0,
      "step": 9950
    },
    {
      "entropy": 4.837851476669312,
      "epoch": 8.330543933054393,
      "grad_norm": 1.5078125,
      "learning_rate": 8.58958303650754e-05,
      "loss": 4.0484,
      "mean_token_accuracy": 0.2803361743688583,
      "num_tokens": 18394604.0,
      "step": 9955
    },
    {
      "entropy": 4.8861125946044925,
      "epoch": 8.334728033472803,
      "grad_norm": 1.4375,
      "learning_rate": 8.572112791688105e-05,
      "loss": 4.0394,
      "mean_token_accuracy": 0.28187002539634703,
      "num_tokens": 18403477.0,
      "step": 9960
    },
    {
      "entropy": 4.850705099105835,
      "epoch": 8.338912133891213,
      "grad_norm": 1.4375,
      "learning_rate": 8.554681497417416e-05,
      "loss": 4.0623,
      "mean_token_accuracy": 0.27977425009012225,
      "num_tokens": 18412866.0,
      "step": 9965
    },
    {
      "entropy": 4.858709096908569,
      "epoch": 8.343096234309623,
      "grad_norm": 1.2890625,
      "learning_rate": 8.537289189566284e-05,
      "loss": 4.0334,
      "mean_token_accuracy": 0.27595295161008837,
      "num_tokens": 18421742.0,
      "step": 9970
    },
    {
      "entropy": 4.893074607849121,
      "epoch": 8.347280334728033,
      "grad_norm": 1.359375,
      "learning_rate": 8.519935903925271e-05,
      "loss": 4.1198,
      "mean_token_accuracy": 0.26314457654953005,
      "num_tokens": 18431331.0,
      "step": 9975
    },
    {
      "entropy": 4.8549974918365475,
      "epoch": 8.351464435146443,
      "grad_norm": 1.390625,
      "learning_rate": 8.502621676204666e-05,
      "loss": 4.183,
      "mean_token_accuracy": 0.2665374428033829,
      "num_tokens": 18440906.0,
      "step": 9980
    },
    {
      "entropy": 4.843502712249756,
      "epoch": 8.355648535564853,
      "grad_norm": 1.34375,
      "learning_rate": 8.485346542034343e-05,
      "loss": 4.0998,
      "mean_token_accuracy": 0.26991347372531893,
      "num_tokens": 18450093.0,
      "step": 9985
    },
    {
      "entropy": 4.782234811782837,
      "epoch": 8.359832635983263,
      "grad_norm": 1.5078125,
      "learning_rate": 8.468110536963763e-05,
      "loss": 4.0234,
      "mean_token_accuracy": 0.2773934602737427,
      "num_tokens": 18458536.0,
      "step": 9990
    },
    {
      "entropy": 4.85179967880249,
      "epoch": 8.364016736401673,
      "grad_norm": 1.5078125,
      "learning_rate": 8.450913696461856e-05,
      "loss": 4.0464,
      "mean_token_accuracy": 0.28092043101787567,
      "num_tokens": 18466910.0,
      "step": 9995
    },
    {
      "entropy": 4.87721004486084,
      "epoch": 8.368200836820083,
      "grad_norm": 1.421875,
      "learning_rate": 8.433756055916944e-05,
      "loss": 4.1489,
      "mean_token_accuracy": 0.26924076676368713,
      "num_tokens": 18476098.0,
      "step": 10000
    },
    {
      "epoch": 8.368200836820083,
      "eval_entropy": 4.732347448423108,
      "eval_loss": 5.31918478012085,
      "eval_mean_token_accuracy": 0.20304432726162344,
      "eval_num_tokens": 18476098.0,
      "eval_runtime": 2.7319,
      "eval_samples_per_second": 1430.136,
      "eval_steps_per_second": 178.996,
      "step": 10000
    }
  ],
  "logging_steps": 5,
  "max_steps": 11950,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6910980794368e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}