{
  "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": 10.692013263702393,
      "epoch": 0.0041841004184100415,
      "grad_norm": 14.4375,
      "learning_rate": 2e-06,
      "loss": 10.8216,
      "mean_token_accuracy": 0.0,
      "num_tokens": 9848.0,
      "step": 5
    },
    {
      "entropy": 10.69194860458374,
      "epoch": 0.008368200836820083,
      "grad_norm": 13.4375,
      "learning_rate": 4.5e-06,
      "loss": 10.7246,
      "mean_token_accuracy": 0.00012674271129071714,
      "num_tokens": 18196.0,
      "step": 10
    },
    {
      "entropy": 10.691147422790527,
      "epoch": 0.012552301255230125,
      "grad_norm": 9.9375,
      "learning_rate": 7e-06,
      "loss": 10.4677,
      "mean_token_accuracy": 0.0241286511416547,
      "num_tokens": 26927.0,
      "step": 15
    },
    {
      "entropy": 10.680734634399414,
      "epoch": 0.016736401673640166,
      "grad_norm": 6.6875,
      "learning_rate": 9.5e-06,
      "loss": 10.115,
      "mean_token_accuracy": 0.04429602064192295,
      "num_tokens": 34987.0,
      "step": 20
    },
    {
      "entropy": 10.62258882522583,
      "epoch": 0.02092050209205021,
      "grad_norm": 4.25,
      "learning_rate": 1.2e-05,
      "loss": 9.8294,
      "mean_token_accuracy": 0.04711197260767221,
      "num_tokens": 44936.0,
      "step": 25
    },
    {
      "entropy": 10.580855178833009,
      "epoch": 0.02510460251046025,
      "grad_norm": 3.515625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.6731,
      "mean_token_accuracy": 0.046292875334620474,
      "num_tokens": 54562.0,
      "step": 30
    },
    {
      "entropy": 10.583562088012695,
      "epoch": 0.029288702928870293,
      "grad_norm": 2.890625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6171,
      "mean_token_accuracy": 0.04133799690753222,
      "num_tokens": 63844.0,
      "step": 35
    },
    {
      "entropy": 10.599079990386963,
      "epoch": 0.03347280334728033,
      "grad_norm": 2.8125,
      "learning_rate": 1.95e-05,
      "loss": 9.6032,
      "mean_token_accuracy": 0.040498660318553446,
      "num_tokens": 74331.0,
      "step": 40
    },
    {
      "entropy": 10.584051704406738,
      "epoch": 0.03765690376569038,
      "grad_norm": 2.796875,
      "learning_rate": 2.2e-05,
      "loss": 9.5052,
      "mean_token_accuracy": 0.04810179658234119,
      "num_tokens": 83028.0,
      "step": 45
    },
    {
      "entropy": 10.553222179412842,
      "epoch": 0.04184100418410042,
      "grad_norm": 2.5625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4503,
      "mean_token_accuracy": 0.04860256612300873,
      "num_tokens": 91395.0,
      "step": 50
    },
    {
      "entropy": 10.53250617980957,
      "epoch": 0.04602510460251046,
      "grad_norm": 2.859375,
      "learning_rate": 2.7e-05,
      "loss": 9.3929,
      "mean_token_accuracy": 0.05782372578978538,
      "num_tokens": 101868.0,
      "step": 55
    },
    {
      "entropy": 10.429417991638184,
      "epoch": 0.0502092050209205,
      "grad_norm": 2.40625,
      "learning_rate": 2.95e-05,
      "loss": 9.2812,
      "mean_token_accuracy": 0.05993606857955456,
      "num_tokens": 110403.0,
      "step": 60
    },
    {
      "entropy": 10.376260757446289,
      "epoch": 0.05439330543933055,
      "grad_norm": 2.5,
      "learning_rate": 3.2e-05,
      "loss": 9.3299,
      "mean_token_accuracy": 0.05404467172920704,
      "num_tokens": 119557.0,
      "step": 65
    },
    {
      "entropy": 10.482521724700927,
      "epoch": 0.058577405857740586,
      "grad_norm": 2.515625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.1462,
      "mean_token_accuracy": 0.05712815783917904,
      "num_tokens": 128903.0,
      "step": 70
    },
    {
      "entropy": 10.432404041290283,
      "epoch": 0.06276150627615062,
      "grad_norm": 2.546875,
      "learning_rate": 3.7e-05,
      "loss": 9.1137,
      "mean_token_accuracy": 0.05686149187386036,
      "num_tokens": 137630.0,
      "step": 75
    },
    {
      "entropy": 10.388175106048584,
      "epoch": 0.06694560669456066,
      "grad_norm": 2.890625,
      "learning_rate": 3.95e-05,
      "loss": 8.8947,
      "mean_token_accuracy": 0.06613010838627816,
      "num_tokens": 146335.0,
      "step": 80
    },
    {
      "entropy": 10.388596439361573,
      "epoch": 0.07112970711297072,
      "grad_norm": 2.4375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.8136,
      "mean_token_accuracy": 0.06454679258167743,
      "num_tokens": 155766.0,
      "step": 85
    },
    {
      "entropy": 10.335438442230224,
      "epoch": 0.07531380753138076,
      "grad_norm": 2.234375,
      "learning_rate": 4.45e-05,
      "loss": 8.7316,
      "mean_token_accuracy": 0.06548036001622677,
      "num_tokens": 165844.0,
      "step": 90
    },
    {
      "entropy": 10.317347526550293,
      "epoch": 0.0794979079497908,
      "grad_norm": 2.34375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6896,
      "mean_token_accuracy": 0.06795907951891422,
      "num_tokens": 174858.0,
      "step": 95
    },
    {
      "entropy": 10.24217710494995,
      "epoch": 0.08368200836820083,
      "grad_norm": 2.21875,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.5453,
      "mean_token_accuracy": 0.07007661424577236,
      "num_tokens": 184020.0,
      "step": 100
    },
    {
      "entropy": 10.19391098022461,
      "epoch": 0.08786610878661087,
      "grad_norm": 2.25,
      "learning_rate": 5.2e-05,
      "loss": 8.4004,
      "mean_token_accuracy": 0.06834917515516281,
      "num_tokens": 193714.0,
      "step": 105
    },
    {
      "entropy": 10.143691921234131,
      "epoch": 0.09205020920502092,
      "grad_norm": 2.265625,
      "learning_rate": 5.45e-05,
      "loss": 8.3625,
      "mean_token_accuracy": 0.07014007195830345,
      "num_tokens": 203139.0,
      "step": 110
    },
    {
      "entropy": 10.033193874359132,
      "epoch": 0.09623430962343096,
      "grad_norm": 2.75,
      "learning_rate": 5.7e-05,
      "loss": 8.2108,
      "mean_token_accuracy": 0.0716083101928234,
      "num_tokens": 212446.0,
      "step": 115
    },
    {
      "entropy": 9.93264389038086,
      "epoch": 0.100418410041841,
      "grad_norm": 2.15625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.0305,
      "mean_token_accuracy": 0.08365688249468803,
      "num_tokens": 220910.0,
      "step": 120
    },
    {
      "entropy": 9.831494235992432,
      "epoch": 0.10460251046025104,
      "grad_norm": 1.9453125,
      "learning_rate": 6.2e-05,
      "loss": 7.975,
      "mean_token_accuracy": 0.07431616149842739,
      "num_tokens": 230522.0,
      "step": 125
    },
    {
      "entropy": 9.695558643341064,
      "epoch": 0.1087866108786611,
      "grad_norm": 1.8046875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.8131,
      "mean_token_accuracy": 0.07820754498243332,
      "num_tokens": 240092.0,
      "step": 130
    },
    {
      "entropy": 9.519502639770508,
      "epoch": 0.11297071129707113,
      "grad_norm": 1.859375,
      "learning_rate": 6.7e-05,
      "loss": 7.7618,
      "mean_token_accuracy": 0.07841893658041954,
      "num_tokens": 249478.0,
      "step": 135
    },
    {
      "entropy": 9.289987373352051,
      "epoch": 0.11715481171548117,
      "grad_norm": 2.578125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.522,
      "mean_token_accuracy": 0.09086068868637084,
      "num_tokens": 257714.0,
      "step": 140
    },
    {
      "entropy": 9.143731307983398,
      "epoch": 0.12133891213389121,
      "grad_norm": 1.6484375,
      "learning_rate": 7.2e-05,
      "loss": 7.5578,
      "mean_token_accuracy": 0.08316774666309357,
      "num_tokens": 266569.0,
      "step": 145
    },
    {
      "entropy": 8.99811429977417,
      "epoch": 0.12552301255230125,
      "grad_norm": 1.4609375,
      "learning_rate": 7.45e-05,
      "loss": 7.4546,
      "mean_token_accuracy": 0.08273414000868798,
      "num_tokens": 276688.0,
      "step": 150
    },
    {
      "entropy": 8.839478302001954,
      "epoch": 0.1297071129707113,
      "grad_norm": 1.6171875,
      "learning_rate": 7.7e-05,
      "loss": 7.3044,
      "mean_token_accuracy": 0.08153865486383438,
      "num_tokens": 285430.0,
      "step": 155
    },
    {
      "entropy": 8.572208499908447,
      "epoch": 0.13389121338912133,
      "grad_norm": 1.8203125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.2963,
      "mean_token_accuracy": 0.08305188342928886,
      "num_tokens": 295727.0,
      "step": 160
    },
    {
      "entropy": 8.346776962280273,
      "epoch": 0.13807531380753138,
      "grad_norm": 1.6171875,
      "learning_rate": 8.2e-05,
      "loss": 7.2537,
      "mean_token_accuracy": 0.0843418687582016,
      "num_tokens": 304854.0,
      "step": 165
    },
    {
      "entropy": 8.35010871887207,
      "epoch": 0.14225941422594143,
      "grad_norm": 1.640625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.2224,
      "mean_token_accuracy": 0.08320909515023231,
      "num_tokens": 313579.0,
      "step": 170
    },
    {
      "entropy": 8.109733581542969,
      "epoch": 0.14644351464435146,
      "grad_norm": 1.4453125,
      "learning_rate": 8.7e-05,
      "loss": 7.2637,
      "mean_token_accuracy": 0.08496511876583099,
      "num_tokens": 323310.0,
      "step": 175
    },
    {
      "entropy": 8.118343544006347,
      "epoch": 0.1506276150627615,
      "grad_norm": 1.4453125,
      "learning_rate": 8.95e-05,
      "loss": 7.1236,
      "mean_token_accuracy": 0.0888123981654644,
      "num_tokens": 331720.0,
      "step": 180
    },
    {
      "entropy": 8.003835868835449,
      "epoch": 0.15481171548117154,
      "grad_norm": 1.6640625,
      "learning_rate": 9.2e-05,
      "loss": 7.0728,
      "mean_token_accuracy": 0.0916049025952816,
      "num_tokens": 340331.0,
      "step": 185
    },
    {
      "entropy": 7.846502780914307,
      "epoch": 0.1589958158995816,
      "grad_norm": 1.3046875,
      "learning_rate": 9.45e-05,
      "loss": 7.0803,
      "mean_token_accuracy": 0.09315933063626289,
      "num_tokens": 349677.0,
      "step": 190
    },
    {
      "entropy": 7.870132970809936,
      "epoch": 0.16317991631799164,
      "grad_norm": 1.5078125,
      "learning_rate": 9.7e-05,
      "loss": 7.0721,
      "mean_token_accuracy": 0.08652734383940697,
      "num_tokens": 358674.0,
      "step": 195
    },
    {
      "entropy": 7.712577533721924,
      "epoch": 0.16736401673640167,
      "grad_norm": 1.4921875,
      "learning_rate": 9.95e-05,
      "loss": 7.0811,
      "mean_token_accuracy": 0.08503658622503281,
      "num_tokens": 368258.0,
      "step": 200
    },
    {
      "entropy": 7.745207118988037,
      "epoch": 0.17154811715481172,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000102,
      "loss": 6.9726,
      "mean_token_accuracy": 0.09250840395689011,
      "num_tokens": 377083.0,
      "step": 205
    },
    {
      "entropy": 7.754372835159302,
      "epoch": 0.17573221757322174,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.0446,
      "mean_token_accuracy": 0.0841131079941988,
      "num_tokens": 385456.0,
      "step": 210
    },
    {
      "entropy": 7.639361429214477,
      "epoch": 0.1799163179916318,
      "grad_norm": 1.40625,
      "learning_rate": 0.000107,
      "loss": 6.9265,
      "mean_token_accuracy": 0.0942225843667984,
      "num_tokens": 395295.0,
      "step": 215
    },
    {
      "entropy": 7.558081722259521,
      "epoch": 0.18410041841004185,
      "grad_norm": 1.796875,
      "learning_rate": 0.0001095,
      "loss": 6.9772,
      "mean_token_accuracy": 0.08982849642634391,
      "num_tokens": 405023.0,
      "step": 220
    },
    {
      "entropy": 7.534724760055542,
      "epoch": 0.18828451882845187,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000112,
      "loss": 6.9996,
      "mean_token_accuracy": 0.08760679811239243,
      "num_tokens": 414224.0,
      "step": 225
    },
    {
      "entropy": 7.567825078964233,
      "epoch": 0.19246861924686193,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001145,
      "loss": 6.8948,
      "mean_token_accuracy": 0.08697729334235191,
      "num_tokens": 423951.0,
      "step": 230
    },
    {
      "entropy": 7.65638952255249,
      "epoch": 0.19665271966527198,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.9173,
      "mean_token_accuracy": 0.08747867494821548,
      "num_tokens": 432427.0,
      "step": 235
    },
    {
      "entropy": 7.392250013351441,
      "epoch": 0.200836820083682,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.8495,
      "mean_token_accuracy": 0.09490769654512406,
      "num_tokens": 441560.0,
      "step": 240
    },
    {
      "entropy": 7.542767906188965,
      "epoch": 0.20502092050209206,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000122,
      "loss": 6.9423,
      "mean_token_accuracy": 0.08346537575125694,
      "num_tokens": 450364.0,
      "step": 245
    },
    {
      "entropy": 7.399922561645508,
      "epoch": 0.20920502092050208,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001245,
      "loss": 6.8922,
      "mean_token_accuracy": 0.08459680303931236,
      "num_tokens": 460076.0,
      "step": 250
    },
    {
      "entropy": 7.435028028488159,
      "epoch": 0.21338912133891214,
      "grad_norm": 1.265625,
      "learning_rate": 0.000127,
      "loss": 6.8343,
      "mean_token_accuracy": 0.09127931594848633,
      "num_tokens": 468718.0,
      "step": 255
    },
    {
      "entropy": 7.456234979629516,
      "epoch": 0.2175732217573222,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001295,
      "loss": 6.9113,
      "mean_token_accuracy": 0.08921206891536712,
      "num_tokens": 478436.0,
      "step": 260
    },
    {
      "entropy": 7.390152215957642,
      "epoch": 0.2217573221757322,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000132,
      "loss": 6.9278,
      "mean_token_accuracy": 0.0872597984969616,
      "num_tokens": 486833.0,
      "step": 265
    },
    {
      "entropy": 7.41746563911438,
      "epoch": 0.22594142259414227,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.7959,
      "mean_token_accuracy": 0.09919953644275666,
      "num_tokens": 495119.0,
      "step": 270
    },
    {
      "entropy": 7.363225364685059,
      "epoch": 0.2301255230125523,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.9685,
      "mean_token_accuracy": 0.09823801144957542,
      "num_tokens": 505396.0,
      "step": 275
    },
    {
      "entropy": 7.373254871368408,
      "epoch": 0.23430962343096234,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001395,
      "loss": 6.8264,
      "mean_token_accuracy": 0.08988286629319191,
      "num_tokens": 515259.0,
      "step": 280
    },
    {
      "entropy": 7.381401538848877,
      "epoch": 0.2384937238493724,
      "grad_norm": 1.4375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.8366,
      "mean_token_accuracy": 0.08750834614038468,
      "num_tokens": 523608.0,
      "step": 285
    },
    {
      "entropy": 7.272293424606323,
      "epoch": 0.24267782426778242,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001445,
      "loss": 6.8138,
      "mean_token_accuracy": 0.09048807919025421,
      "num_tokens": 534613.0,
      "step": 290
    },
    {
      "entropy": 7.320169878005982,
      "epoch": 0.24686192468619247,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000147,
      "loss": 6.8579,
      "mean_token_accuracy": 0.09750684052705764,
      "num_tokens": 543347.0,
      "step": 295
    },
    {
      "entropy": 7.278151416778565,
      "epoch": 0.2510460251046025,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001495,
      "loss": 6.8097,
      "mean_token_accuracy": 0.08896707370877266,
      "num_tokens": 552371.0,
      "step": 300
    },
    {
      "entropy": 7.234066581726074,
      "epoch": 0.25523012552301255,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000152,
      "loss": 6.7401,
      "mean_token_accuracy": 0.09521848261356354,
      "num_tokens": 560742.0,
      "step": 305
    },
    {
      "entropy": 7.309448719024658,
      "epoch": 0.2594142259414226,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.8506,
      "mean_token_accuracy": 0.09421026632189751,
      "num_tokens": 569960.0,
      "step": 310
    },
    {
      "entropy": 7.26940336227417,
      "epoch": 0.26359832635983266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000157,
      "loss": 6.7952,
      "mean_token_accuracy": 0.09265207946300506,
      "num_tokens": 578859.0,
      "step": 315
    },
    {
      "entropy": 7.1658327102661135,
      "epoch": 0.26778242677824265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001595,
      "loss": 6.802,
      "mean_token_accuracy": 0.09623523280024529,
      "num_tokens": 588146.0,
      "step": 320
    },
    {
      "entropy": 7.232771396636963,
      "epoch": 0.2719665271966527,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000162,
      "loss": 6.7437,
      "mean_token_accuracy": 0.09764528870582581,
      "num_tokens": 597268.0,
      "step": 325
    },
    {
      "entropy": 7.080677461624146,
      "epoch": 0.27615062761506276,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.676,
      "mean_token_accuracy": 0.0969509594142437,
      "num_tokens": 606199.0,
      "step": 330
    },
    {
      "entropy": 7.3103187561035154,
      "epoch": 0.2803347280334728,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8102,
      "mean_token_accuracy": 0.08201513588428497,
      "num_tokens": 615966.0,
      "step": 335
    },
    {
      "entropy": 7.242901706695557,
      "epoch": 0.28451882845188287,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7714,
      "mean_token_accuracy": 0.0932280994951725,
      "num_tokens": 625689.0,
      "step": 340
    },
    {
      "entropy": 7.17906231880188,
      "epoch": 0.28870292887029286,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7957,
      "mean_token_accuracy": 0.08773538246750831,
      "num_tokens": 634877.0,
      "step": 345
    },
    {
      "entropy": 7.1589775562286375,
      "epoch": 0.2928870292887029,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.7211,
      "mean_token_accuracy": 0.09845930486917495,
      "num_tokens": 644042.0,
      "step": 350
    },
    {
      "entropy": 7.122976160049438,
      "epoch": 0.29707112970711297,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000177,
      "loss": 6.7157,
      "mean_token_accuracy": 0.09408211931586266,
      "num_tokens": 653072.0,
      "step": 355
    },
    {
      "entropy": 7.08035979270935,
      "epoch": 0.301255230125523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001795,
      "loss": 6.7206,
      "mean_token_accuracy": 0.09660579040646552,
      "num_tokens": 663686.0,
      "step": 360
    },
    {
      "entropy": 7.087335777282715,
      "epoch": 0.3054393305439331,
      "grad_norm": 1.25,
      "learning_rate": 0.000182,
      "loss": 6.655,
      "mean_token_accuracy": 0.09818260595202447,
      "num_tokens": 672293.0,
      "step": 365
    },
    {
      "entropy": 6.929145002365113,
      "epoch": 0.30962343096234307,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001845,
      "loss": 6.617,
      "mean_token_accuracy": 0.10118300542235374,
      "num_tokens": 681758.0,
      "step": 370
    },
    {
      "entropy": 7.119603776931763,
      "epoch": 0.3138075313807531,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000187,
      "loss": 6.6496,
      "mean_token_accuracy": 0.10070758163928986,
      "num_tokens": 690727.0,
      "step": 375
    },
    {
      "entropy": 7.051980209350586,
      "epoch": 0.3179916317991632,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001895,
      "loss": 6.6052,
      "mean_token_accuracy": 0.09968686252832412,
      "num_tokens": 701210.0,
      "step": 380
    },
    {
      "entropy": 6.929204320907592,
      "epoch": 0.32217573221757323,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000192,
      "loss": 6.5924,
      "mean_token_accuracy": 0.10255414322018623,
      "num_tokens": 710869.0,
      "step": 385
    },
    {
      "entropy": 7.104875373840332,
      "epoch": 0.3263598326359833,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001945,
      "loss": 6.7186,
      "mean_token_accuracy": 0.095282444357872,
      "num_tokens": 721455.0,
      "step": 390
    },
    {
      "entropy": 6.909361505508423,
      "epoch": 0.3305439330543933,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.5982,
      "mean_token_accuracy": 0.10492872297763825,
      "num_tokens": 730917.0,
      "step": 395
    },
    {
      "entropy": 7.0232825756073,
      "epoch": 0.33472803347280333,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6235,
      "mean_token_accuracy": 0.10053492337465286,
      "num_tokens": 739302.0,
      "step": 400
    },
    {
      "entropy": 7.034136819839477,
      "epoch": 0.3389121338912134,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000202,
      "loss": 6.5922,
      "mean_token_accuracy": 0.09860573858022689,
      "num_tokens": 748785.0,
      "step": 405
    },
    {
      "entropy": 6.9551417350769045,
      "epoch": 0.34309623430962344,
      "grad_norm": 1.25,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.5277,
      "mean_token_accuracy": 0.10383330285549164,
      "num_tokens": 757215.0,
      "step": 410
    },
    {
      "entropy": 6.833004713058472,
      "epoch": 0.3472803347280335,
      "grad_norm": 1.390625,
      "learning_rate": 0.000207,
      "loss": 6.4628,
      "mean_token_accuracy": 0.10961654931306838,
      "num_tokens": 765682.0,
      "step": 415
    },
    {
      "entropy": 7.001798439025879,
      "epoch": 0.3514644351464435,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002095,
      "loss": 6.6685,
      "mean_token_accuracy": 0.09785127714276314,
      "num_tokens": 774814.0,
      "step": 420
    },
    {
      "entropy": 6.951812839508056,
      "epoch": 0.35564853556485354,
      "grad_norm": 1.46875,
      "learning_rate": 0.000212,
      "loss": 6.5856,
      "mean_token_accuracy": 0.10029251947999,
      "num_tokens": 784019.0,
      "step": 425
    },
    {
      "entropy": 6.814007663726807,
      "epoch": 0.3598326359832636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002145,
      "loss": 6.4521,
      "mean_token_accuracy": 0.10874552577733994,
      "num_tokens": 793237.0,
      "step": 430
    },
    {
      "entropy": 6.967527246475219,
      "epoch": 0.36401673640167365,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.5148,
      "mean_token_accuracy": 0.1004909910261631,
      "num_tokens": 801548.0,
      "step": 435
    },
    {
      "entropy": 6.829876708984375,
      "epoch": 0.3682008368200837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002195,
      "loss": 6.5539,
      "mean_token_accuracy": 0.10507597774267197,
      "num_tokens": 810974.0,
      "step": 440
    },
    {
      "entropy": 6.837385892868042,
      "epoch": 0.3723849372384937,
      "grad_norm": 1.421875,
      "learning_rate": 0.000222,
      "loss": 6.4384,
      "mean_token_accuracy": 0.10976337790489196,
      "num_tokens": 820023.0,
      "step": 445
    },
    {
      "entropy": 6.836634254455566,
      "epoch": 0.37656903765690375,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002245,
      "loss": 6.5294,
      "mean_token_accuracy": 0.10367678925395012,
      "num_tokens": 828297.0,
      "step": 450
    },
    {
      "entropy": 6.901482248306275,
      "epoch": 0.3807531380753138,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.5793,
      "mean_token_accuracy": 0.09453905671834946,
      "num_tokens": 838268.0,
      "step": 455
    },
    {
      "entropy": 6.938200902938843,
      "epoch": 0.38493723849372385,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.4984,
      "mean_token_accuracy": 0.10036915764212609,
      "num_tokens": 846491.0,
      "step": 460
    },
    {
      "entropy": 6.742056369781494,
      "epoch": 0.3891213389121339,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.4374,
      "mean_token_accuracy": 0.10895640999078751,
      "num_tokens": 856501.0,
      "step": 465
    },
    {
      "entropy": 6.806368494033814,
      "epoch": 0.39330543933054396,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5002,
      "mean_token_accuracy": 0.10559085234999657,
      "num_tokens": 867054.0,
      "step": 470
    },
    {
      "entropy": 6.738630914688111,
      "epoch": 0.39748953974895396,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000237,
      "loss": 6.521,
      "mean_token_accuracy": 0.10245606228709221,
      "num_tokens": 875956.0,
      "step": 475
    },
    {
      "entropy": 6.697034120559692,
      "epoch": 0.401673640167364,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002395,
      "loss": 6.4358,
      "mean_token_accuracy": 0.10774662867188453,
      "num_tokens": 884427.0,
      "step": 480
    },
    {
      "entropy": 6.911193704605102,
      "epoch": 0.40585774058577406,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000242,
      "loss": 6.5239,
      "mean_token_accuracy": 0.1094229944050312,
      "num_tokens": 894181.0,
      "step": 485
    },
    {
      "entropy": 6.690055084228516,
      "epoch": 0.4100418410041841,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002445,
      "loss": 6.3689,
      "mean_token_accuracy": 0.11442587077617646,
      "num_tokens": 904076.0,
      "step": 490
    },
    {
      "entropy": 6.700846529006958,
      "epoch": 0.41422594142259417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000247,
      "loss": 6.5172,
      "mean_token_accuracy": 0.10785161033272743,
      "num_tokens": 913915.0,
      "step": 495
    },
    {
      "entropy": 6.8667717456817625,
      "epoch": 0.41841004184100417,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002495,
      "loss": 6.5828,
      "mean_token_accuracy": 0.10332566574215889,
      "num_tokens": 922876.0,
      "step": 500
    },
    {
      "epoch": 0.41841004184100417,
      "eval_entropy": 6.700536389536166,
      "eval_loss": 6.505311965942383,
      "eval_mean_token_accuracy": 0.10990596924487554,
      "eval_num_tokens": 922876.0,
      "eval_runtime": 2.7776,
      "eval_samples_per_second": 1406.597,
      "eval_steps_per_second": 176.05,
      "step": 500
    },
    {
      "entropy": 6.676968288421631,
      "epoch": 0.4225941422594142,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000252,
      "loss": 6.3727,
      "mean_token_accuracy": 0.11308198049664497,
      "num_tokens": 931998.0,
      "step": 505
    },
    {
      "entropy": 6.805904197692871,
      "epoch": 0.42677824267782427,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002545,
      "loss": 6.4411,
      "mean_token_accuracy": 0.10866107791662216,
      "num_tokens": 941387.0,
      "step": 510
    },
    {
      "entropy": 6.625282573699951,
      "epoch": 0.4309623430962343,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000257,
      "loss": 6.4789,
      "mean_token_accuracy": 0.10362649038434028,
      "num_tokens": 951866.0,
      "step": 515
    },
    {
      "entropy": 6.657781028747559,
      "epoch": 0.4351464435146444,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002595,
      "loss": 6.4117,
      "mean_token_accuracy": 0.10945287123322486,
      "num_tokens": 961252.0,
      "step": 520
    },
    {
      "entropy": 6.821563005447388,
      "epoch": 0.4393305439330544,
      "grad_norm": 1.5,
      "learning_rate": 0.000262,
      "loss": 6.4206,
      "mean_token_accuracy": 0.11085275113582611,
      "num_tokens": 969857.0,
      "step": 525
    },
    {
      "entropy": 6.679222440719604,
      "epoch": 0.4435146443514644,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.4176,
      "mean_token_accuracy": 0.11243949085474014,
      "num_tokens": 978754.0,
      "step": 530
    },
    {
      "entropy": 6.723608922958374,
      "epoch": 0.4476987447698745,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.391,
      "mean_token_accuracy": 0.1103011853992939,
      "num_tokens": 987935.0,
      "step": 535
    },
    {
      "entropy": 6.682740306854248,
      "epoch": 0.45188284518828453,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.3588,
      "mean_token_accuracy": 0.11257985085248948,
      "num_tokens": 996873.0,
      "step": 540
    },
    {
      "entropy": 6.741128253936767,
      "epoch": 0.4560669456066946,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.4534,
      "mean_token_accuracy": 0.10576379746198654,
      "num_tokens": 1006487.0,
      "step": 545
    },
    {
      "entropy": 6.482868623733521,
      "epoch": 0.4602510460251046,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002745,
      "loss": 6.2856,
      "mean_token_accuracy": 0.11558721140027046,
      "num_tokens": 1015715.0,
      "step": 550
    },
    {
      "entropy": 6.656096649169922,
      "epoch": 0.46443514644351463,
      "grad_norm": 1.4375,
      "learning_rate": 0.000277,
      "loss": 6.4102,
      "mean_token_accuracy": 0.11223409548401833,
      "num_tokens": 1025091.0,
      "step": 555
    },
    {
      "entropy": 6.6403412342071535,
      "epoch": 0.4686192468619247,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002795,
      "loss": 6.342,
      "mean_token_accuracy": 0.10930742770433426,
      "num_tokens": 1034174.0,
      "step": 560
    },
    {
      "entropy": 6.697489976882935,
      "epoch": 0.47280334728033474,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.549,
      "mean_token_accuracy": 0.09722351431846618,
      "num_tokens": 1043777.0,
      "step": 565
    },
    {
      "entropy": 6.652381849288941,
      "epoch": 0.4769874476987448,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002845,
      "loss": 6.3832,
      "mean_token_accuracy": 0.11293103024363518,
      "num_tokens": 1052749.0,
      "step": 570
    },
    {
      "entropy": 6.7199958801269535,
      "epoch": 0.4811715481171548,
      "grad_norm": 1.203125,
      "learning_rate": 0.000287,
      "loss": 6.4071,
      "mean_token_accuracy": 0.11299470067024231,
      "num_tokens": 1061564.0,
      "step": 575
    },
    {
      "entropy": 6.540546655654907,
      "epoch": 0.48535564853556484,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002895,
      "loss": 6.3267,
      "mean_token_accuracy": 0.11745149567723275,
      "num_tokens": 1070006.0,
      "step": 580
    },
    {
      "entropy": 6.592943572998047,
      "epoch": 0.4895397489539749,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000292,
      "loss": 6.3903,
      "mean_token_accuracy": 0.1095091424882412,
      "num_tokens": 1079799.0,
      "step": 585
    },
    {
      "entropy": 6.655527019500733,
      "epoch": 0.49372384937238495,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002945,
      "loss": 6.2219,
      "mean_token_accuracy": 0.120665642619133,
      "num_tokens": 1088842.0,
      "step": 590
    },
    {
      "entropy": 6.531441926956177,
      "epoch": 0.497907949790795,
      "grad_norm": 1.34375,
      "learning_rate": 0.000297,
      "loss": 6.3064,
      "mean_token_accuracy": 0.12399455532431602,
      "num_tokens": 1097960.0,
      "step": 595
    },
    {
      "entropy": 6.42484769821167,
      "epoch": 0.502092050209205,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002995,
      "loss": 6.2793,
      "mean_token_accuracy": 0.11625320315361024,
      "num_tokens": 1106797.0,
      "step": 600
    },
    {
      "entropy": 6.580197429656982,
      "epoch": 0.5062761506276151,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000302,
      "loss": 6.4432,
      "mean_token_accuracy": 0.11806596219539642,
      "num_tokens": 1116384.0,
      "step": 605
    },
    {
      "entropy": 6.462054252624512,
      "epoch": 0.5104602510460251,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003045,
      "loss": 6.2024,
      "mean_token_accuracy": 0.12296005710959435,
      "num_tokens": 1125206.0,
      "step": 610
    },
    {
      "entropy": 6.673148584365845,
      "epoch": 0.5146443514644351,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000307,
      "loss": 6.4218,
      "mean_token_accuracy": 0.11573300361633301,
      "num_tokens": 1134110.0,
      "step": 615
    },
    {
      "entropy": 6.7004382610321045,
      "epoch": 0.5188284518828452,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003095,
      "loss": 6.417,
      "mean_token_accuracy": 0.11261885166168213,
      "num_tokens": 1142833.0,
      "step": 620
    },
    {
      "entropy": 6.535337591171265,
      "epoch": 0.5230125523012552,
      "grad_norm": 1.375,
      "learning_rate": 0.000312,
      "loss": 6.3455,
      "mean_token_accuracy": 0.11737915500998497,
      "num_tokens": 1152006.0,
      "step": 625
    },
    {
      "entropy": 6.4866430282592775,
      "epoch": 0.5271966527196653,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003145,
      "loss": 6.2844,
      "mean_token_accuracy": 0.11385276094079018,
      "num_tokens": 1161061.0,
      "step": 630
    },
    {
      "entropy": 6.597733354568481,
      "epoch": 0.5313807531380753,
      "grad_norm": 1.171875,
      "learning_rate": 0.000317,
      "loss": 6.3047,
      "mean_token_accuracy": 0.11783810779452324,
      "num_tokens": 1170754.0,
      "step": 635
    },
    {
      "entropy": 6.399986743927002,
      "epoch": 0.5355648535564853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003195,
      "loss": 6.2454,
      "mean_token_accuracy": 0.11883974969387054,
      "num_tokens": 1180854.0,
      "step": 640
    },
    {
      "entropy": 6.562164402008056,
      "epoch": 0.5397489539748954,
      "grad_norm": 1.21875,
      "learning_rate": 0.000322,
      "loss": 6.3273,
      "mean_token_accuracy": 0.11495971083641052,
      "num_tokens": 1190567.0,
      "step": 645
    },
    {
      "entropy": 6.4136192321777346,
      "epoch": 0.5439330543933054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.1772,
      "mean_token_accuracy": 0.12412636205554009,
      "num_tokens": 1200247.0,
      "step": 650
    },
    {
      "entropy": 6.411277961730957,
      "epoch": 0.5481171548117155,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.29,
      "mean_token_accuracy": 0.11657113656401634,
      "num_tokens": 1209358.0,
      "step": 655
    },
    {
      "entropy": 6.452598428726196,
      "epoch": 0.5523012552301255,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.3091,
      "mean_token_accuracy": 0.12059221267700196,
      "num_tokens": 1217746.0,
      "step": 660
    },
    {
      "entropy": 6.468178224563599,
      "epoch": 0.5564853556485355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.273,
      "mean_token_accuracy": 0.12242133840918541,
      "num_tokens": 1226829.0,
      "step": 665
    },
    {
      "entropy": 6.409869623184204,
      "epoch": 0.5606694560669456,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.1811,
      "mean_token_accuracy": 0.1252070352435112,
      "num_tokens": 1236642.0,
      "step": 670
    },
    {
      "entropy": 6.611316394805908,
      "epoch": 0.5648535564853556,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000337,
      "loss": 6.3054,
      "mean_token_accuracy": 0.12332511767745018,
      "num_tokens": 1244934.0,
      "step": 675
    },
    {
      "entropy": 6.345842742919922,
      "epoch": 0.5690376569037657,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003395,
      "loss": 6.1933,
      "mean_token_accuracy": 0.11780603304505348,
      "num_tokens": 1254786.0,
      "step": 680
    },
    {
      "entropy": 6.343827581405639,
      "epoch": 0.5732217573221757,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 6.1659,
      "mean_token_accuracy": 0.12387140616774558,
      "num_tokens": 1263745.0,
      "step": 685
    },
    {
      "entropy": 6.4398839473724365,
      "epoch": 0.5774058577405857,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3144,
      "mean_token_accuracy": 0.12714288905262947,
      "num_tokens": 1273101.0,
      "step": 690
    },
    {
      "entropy": 6.485600519180298,
      "epoch": 0.5815899581589958,
      "grad_norm": 1.453125,
      "learning_rate": 0.000347,
      "loss": 6.2931,
      "mean_token_accuracy": 0.11583595350384712,
      "num_tokens": 1282547.0,
      "step": 695
    },
    {
      "entropy": 6.455876684188842,
      "epoch": 0.5857740585774058,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003495,
      "loss": 6.3152,
      "mean_token_accuracy": 0.11679297238588333,
      "num_tokens": 1291594.0,
      "step": 700
    },
    {
      "entropy": 6.429898071289062,
      "epoch": 0.5899581589958159,
      "grad_norm": 1.34375,
      "learning_rate": 0.000352,
      "loss": 6.2199,
      "mean_token_accuracy": 0.12549455389380454,
      "num_tokens": 1300282.0,
      "step": 705
    },
    {
      "entropy": 6.407459688186646,
      "epoch": 0.5941422594142259,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003545,
      "loss": 6.3096,
      "mean_token_accuracy": 0.12226130589842796,
      "num_tokens": 1309304.0,
      "step": 710
    },
    {
      "entropy": 6.392927360534668,
      "epoch": 0.5983263598326359,
      "grad_norm": 1.34375,
      "learning_rate": 0.000357,
      "loss": 6.1472,
      "mean_token_accuracy": 0.12595504522323608,
      "num_tokens": 1317638.0,
      "step": 715
    },
    {
      "entropy": 6.5016227722167965,
      "epoch": 0.602510460251046,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003595,
      "loss": 6.353,
      "mean_token_accuracy": 0.11902206242084504,
      "num_tokens": 1327242.0,
      "step": 720
    },
    {
      "entropy": 6.401812124252319,
      "epoch": 0.606694560669456,
      "grad_norm": 1.1875,
      "learning_rate": 0.000362,
      "loss": 6.1596,
      "mean_token_accuracy": 0.12535706236958505,
      "num_tokens": 1335880.0,
      "step": 725
    },
    {
      "entropy": 6.383457088470459,
      "epoch": 0.6108786610878661,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003645,
      "loss": 6.2361,
      "mean_token_accuracy": 0.12056306451559066,
      "num_tokens": 1344909.0,
      "step": 730
    },
    {
      "entropy": 6.346205854415894,
      "epoch": 0.6150627615062761,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000367,
      "loss": 6.1104,
      "mean_token_accuracy": 0.12955476492643356,
      "num_tokens": 1353997.0,
      "step": 735
    },
    {
      "entropy": 6.379322576522827,
      "epoch": 0.6192468619246861,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003695,
      "loss": 6.2733,
      "mean_token_accuracy": 0.12443547993898392,
      "num_tokens": 1363293.0,
      "step": 740
    },
    {
      "entropy": 6.49610013961792,
      "epoch": 0.6234309623430963,
      "grad_norm": 1.21875,
      "learning_rate": 0.000372,
      "loss": 6.2785,
      "mean_token_accuracy": 0.11846815049648285,
      "num_tokens": 1372174.0,
      "step": 745
    },
    {
      "entropy": 6.406333255767822,
      "epoch": 0.6276150627615062,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003745,
      "loss": 6.1697,
      "mean_token_accuracy": 0.12304460108280182,
      "num_tokens": 1381067.0,
      "step": 750
    },
    {
      "entropy": 6.325730276107788,
      "epoch": 0.6317991631799164,
      "grad_norm": 1.1875,
      "learning_rate": 0.000377,
      "loss": 6.1643,
      "mean_token_accuracy": 0.12812735885381699,
      "num_tokens": 1390072.0,
      "step": 755
    },
    {
      "entropy": 6.292504358291626,
      "epoch": 0.6359832635983264,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003795,
      "loss": 6.156,
      "mean_token_accuracy": 0.12865231558680534,
      "num_tokens": 1399391.0,
      "step": 760
    },
    {
      "entropy": 6.304660320281982,
      "epoch": 0.6401673640167364,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000382,
      "loss": 6.1631,
      "mean_token_accuracy": 0.12419576495885849,
      "num_tokens": 1408054.0,
      "step": 765
    },
    {
      "entropy": 6.39186725616455,
      "epoch": 0.6443514644351465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003845,
      "loss": 6.1785,
      "mean_token_accuracy": 0.11784627810120582,
      "num_tokens": 1417406.0,
      "step": 770
    },
    {
      "entropy": 6.323220872879029,
      "epoch": 0.6485355648535565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.237,
      "mean_token_accuracy": 0.11933397874236107,
      "num_tokens": 1426907.0,
      "step": 775
    },
    {
      "entropy": 6.324512672424317,
      "epoch": 0.6527196652719666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3007,
      "mean_token_accuracy": 0.11848932132124901,
      "num_tokens": 1436085.0,
      "step": 780
    },
    {
      "entropy": 6.432321214675904,
      "epoch": 0.6569037656903766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.208,
      "mean_token_accuracy": 0.12037489041686059,
      "num_tokens": 1444823.0,
      "step": 785
    },
    {
      "entropy": 6.354400157928467,
      "epoch": 0.6610878661087866,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.2424,
      "mean_token_accuracy": 0.1207665555179119,
      "num_tokens": 1454621.0,
      "step": 790
    },
    {
      "entropy": 6.24983959197998,
      "epoch": 0.6652719665271967,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.1116,
      "mean_token_accuracy": 0.12615672051906585,
      "num_tokens": 1463784.0,
      "step": 795
    },
    {
      "entropy": 6.329709482192993,
      "epoch": 0.6694560669456067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995,
      "loss": 6.2278,
      "mean_token_accuracy": 0.12423735633492469,
      "num_tokens": 1473257.0,
      "step": 800
    },
    {
      "entropy": 6.327875566482544,
      "epoch": 0.6736401673640168,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000402,
      "loss": 6.1715,
      "mean_token_accuracy": 0.13215391784906388,
      "num_tokens": 1482195.0,
      "step": 805
    },
    {
      "entropy": 6.267716884613037,
      "epoch": 0.6778242677824268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004045,
      "loss": 6.2104,
      "mean_token_accuracy": 0.12745170965790747,
      "num_tokens": 1491590.0,
      "step": 810
    },
    {
      "entropy": 6.32491455078125,
      "epoch": 0.6820083682008368,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.0945,
      "mean_token_accuracy": 0.1320909932255745,
      "num_tokens": 1500208.0,
      "step": 815
    },
    {
      "entropy": 6.329631280899048,
      "epoch": 0.6861924686192469,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004095,
      "loss": 6.2269,
      "mean_token_accuracy": 0.12872799932956697,
      "num_tokens": 1509412.0,
      "step": 820
    },
    {
      "entropy": 6.2926709175109865,
      "epoch": 0.6903765690376569,
      "grad_norm": 1.265625,
      "learning_rate": 0.000412,
      "loss": 6.2233,
      "mean_token_accuracy": 0.12587290704250337,
      "num_tokens": 1518426.0,
      "step": 825
    },
    {
      "entropy": 6.372130298614502,
      "epoch": 0.694560669456067,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004145,
      "loss": 6.173,
      "mean_token_accuracy": 0.12464739456772804,
      "num_tokens": 1528605.0,
      "step": 830
    },
    {
      "entropy": 6.198592662811279,
      "epoch": 0.698744769874477,
      "grad_norm": 1.3125,
      "learning_rate": 0.000417,
      "loss": 6.1198,
      "mean_token_accuracy": 0.12950538620352745,
      "num_tokens": 1537449.0,
      "step": 835
    },
    {
      "entropy": 6.2445250988006595,
      "epoch": 0.702928870292887,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004195,
      "loss": 6.1531,
      "mean_token_accuracy": 0.12678662985563277,
      "num_tokens": 1546743.0,
      "step": 840
    },
    {
      "entropy": 6.198931741714477,
      "epoch": 0.7071129707112971,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422,
      "loss": 6.1512,
      "mean_token_accuracy": 0.12723555639386178,
      "num_tokens": 1555691.0,
      "step": 845
    },
    {
      "entropy": 6.251468992233276,
      "epoch": 0.7112970711297071,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 6.1,
      "mean_token_accuracy": 0.13119761049747466,
      "num_tokens": 1564957.0,
      "step": 850
    },
    {
      "entropy": 6.330463075637818,
      "epoch": 0.7154811715481172,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000427,
      "loss": 6.1739,
      "mean_token_accuracy": 0.12831180393695832,
      "num_tokens": 1573901.0,
      "step": 855
    },
    {
      "entropy": 6.1841747760772705,
      "epoch": 0.7196652719665272,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004295,
      "loss": 6.1514,
      "mean_token_accuracy": 0.13142475932836534,
      "num_tokens": 1582644.0,
      "step": 860
    },
    {
      "entropy": 6.314925765991211,
      "epoch": 0.7238493723849372,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432,
      "loss": 6.2846,
      "mean_token_accuracy": 0.12040768340229988,
      "num_tokens": 1592363.0,
      "step": 865
    },
    {
      "entropy": 6.307796049118042,
      "epoch": 0.7280334728033473,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004345,
      "loss": 6.2164,
      "mean_token_accuracy": 0.12550526708364487,
      "num_tokens": 1602012.0,
      "step": 870
    },
    {
      "entropy": 6.193496656417847,
      "epoch": 0.7322175732217573,
      "grad_norm": 1.21875,
      "learning_rate": 0.000437,
      "loss": 6.1262,
      "mean_token_accuracy": 0.12875815853476524,
      "num_tokens": 1611625.0,
      "step": 875
    },
    {
      "entropy": 6.1236217498779295,
      "epoch": 0.7364016736401674,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004395,
      "loss": 6.0812,
      "mean_token_accuracy": 0.12527592107653618,
      "num_tokens": 1620941.0,
      "step": 880
    },
    {
      "entropy": 6.263450574874878,
      "epoch": 0.7405857740585774,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000442,
      "loss": 6.1654,
      "mean_token_accuracy": 0.12486932873725891,
      "num_tokens": 1629225.0,
      "step": 885
    },
    {
      "entropy": 6.240986633300781,
      "epoch": 0.7447698744769874,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004445,
      "loss": 6.1445,
      "mean_token_accuracy": 0.12441490367054939,
      "num_tokens": 1639551.0,
      "step": 890
    },
    {
      "entropy": 6.223584127426148,
      "epoch": 0.7489539748953975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000447,
      "loss": 6.1279,
      "mean_token_accuracy": 0.12395419999957084,
      "num_tokens": 1648963.0,
      "step": 895
    },
    {
      "entropy": 6.18436336517334,
      "epoch": 0.7531380753138075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.0355,
      "mean_token_accuracy": 0.13170942068099975,
      "num_tokens": 1657735.0,
      "step": 900
    },
    {
      "entropy": 6.318353271484375,
      "epoch": 0.7573221757322176,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.0891,
      "mean_token_accuracy": 0.1331306591629982,
      "num_tokens": 1666349.0,
      "step": 905
    },
    {
      "entropy": 6.016112899780273,
      "epoch": 0.7615062761506276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.0271,
      "mean_token_accuracy": 0.1322978489100933,
      "num_tokens": 1675920.0,
      "step": 910
    },
    {
      "entropy": 6.141036748886108,
      "epoch": 0.7656903765690377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.0628,
      "mean_token_accuracy": 0.13066673129796982,
      "num_tokens": 1685626.0,
      "step": 915
    },
    {
      "entropy": 6.278197765350342,
      "epoch": 0.7698744769874477,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.1888,
      "mean_token_accuracy": 0.1338045746088028,
      "num_tokens": 1694972.0,
      "step": 920
    },
    {
      "entropy": 6.141257333755493,
      "epoch": 0.7740585774058577,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000462,
      "loss": 6.0668,
      "mean_token_accuracy": 0.13685579001903533,
      "num_tokens": 1703806.0,
      "step": 925
    },
    {
      "entropy": 6.192854118347168,
      "epoch": 0.7782426778242678,
      "grad_norm": 1.125,
      "learning_rate": 0.0004645,
      "loss": 6.1659,
      "mean_token_accuracy": 0.13321068361401558,
      "num_tokens": 1712949.0,
      "step": 930
    },
    {
      "entropy": 6.347252273559571,
      "epoch": 0.7824267782426778,
      "grad_norm": 1.171875,
      "learning_rate": 0.000467,
      "loss": 6.287,
      "mean_token_accuracy": 0.12331678792834282,
      "num_tokens": 1722328.0,
      "step": 935
    },
    {
      "entropy": 6.2889646053314205,
      "epoch": 0.7866108786610879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004695,
      "loss": 6.253,
      "mean_token_accuracy": 0.13125553727149963,
      "num_tokens": 1731880.0,
      "step": 940
    },
    {
      "entropy": 6.093116664886475,
      "epoch": 0.7907949790794979,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000472,
      "loss": 6.0693,
      "mean_token_accuracy": 0.13057696744799613,
      "num_tokens": 1740719.0,
      "step": 945
    },
    {
      "entropy": 6.186368799209594,
      "epoch": 0.7949790794979079,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004745,
      "loss": 6.1042,
      "mean_token_accuracy": 0.12656014263629914,
      "num_tokens": 1749753.0,
      "step": 950
    },
    {
      "entropy": 6.255837249755859,
      "epoch": 0.799163179916318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477,
      "loss": 6.1211,
      "mean_token_accuracy": 0.1266303114593029,
      "num_tokens": 1758913.0,
      "step": 955
    },
    {
      "entropy": 6.229157781600952,
      "epoch": 0.803347280334728,
      "grad_norm": 1.125,
      "learning_rate": 0.0004795,
      "loss": 6.1442,
      "mean_token_accuracy": 0.12871945798397064,
      "num_tokens": 1768536.0,
      "step": 960
    },
    {
      "entropy": 6.113988447189331,
      "epoch": 0.8075313807531381,
      "grad_norm": 1.171875,
      "learning_rate": 0.000482,
      "loss": 6.0653,
      "mean_token_accuracy": 0.12936384379863738,
      "num_tokens": 1777291.0,
      "step": 965
    },
    {
      "entropy": 6.186689996719361,
      "epoch": 0.8117154811715481,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004845,
      "loss": 6.1634,
      "mean_token_accuracy": 0.12880500480532647,
      "num_tokens": 1787186.0,
      "step": 970
    },
    {
      "entropy": 6.11544098854065,
      "epoch": 0.8158995815899581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000487,
      "loss": 5.9697,
      "mean_token_accuracy": 0.13747073709964752,
      "num_tokens": 1795581.0,
      "step": 975
    },
    {
      "entropy": 5.999046802520752,
      "epoch": 0.8200836820083682,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004895,
      "loss": 5.9825,
      "mean_token_accuracy": 0.1357624500989914,
      "num_tokens": 1804627.0,
      "step": 980
    },
    {
      "entropy": 6.198841238021851,
      "epoch": 0.8242677824267782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000492,
      "loss": 6.1134,
      "mean_token_accuracy": 0.13310517519712448,
      "num_tokens": 1813808.0,
      "step": 985
    },
    {
      "entropy": 6.064494943618774,
      "epoch": 0.8284518828451883,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004945,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1350795842707157,
      "num_tokens": 1823422.0,
      "step": 990
    },
    {
      "entropy": 6.163973569869995,
      "epoch": 0.8326359832635983,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497,
      "loss": 6.1526,
      "mean_token_accuracy": 0.13110492676496505,
      "num_tokens": 1832437.0,
      "step": 995
    },
    {
      "entropy": 6.180834436416626,
      "epoch": 0.8368200836820083,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995,
      "loss": 6.0707,
      "mean_token_accuracy": 0.12942072749137878,
      "num_tokens": 1842063.0,
      "step": 1000
    },
    {
      "epoch": 0.8368200836820083,
      "eval_entropy": 5.93860879558727,
      "eval_loss": 6.10989236831665,
      "eval_mean_token_accuracy": 0.14120837797721228,
      "eval_num_tokens": 1842063.0,
      "eval_runtime": 2.7411,
      "eval_samples_per_second": 1425.332,
      "eval_steps_per_second": 178.395,
      "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
}