{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8368200836820083,
  "eval_steps": 500,
  "global_step": 1000,
  "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
    }
  ],
  "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": 2662990405632000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}