{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.16,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.692202186584472,
      "epoch": 0.0005333333333333334,
      "grad_norm": 15.75,
      "learning_rate": 2e-06,
      "loss": 11.0362,
      "mean_token_accuracy": 7.338551804423333e-05,
      "num_tokens": 81920.0,
      "step": 5
    },
    {
      "entropy": 10.69208927154541,
      "epoch": 0.0010666666666666667,
      "grad_norm": 15.0,
      "learning_rate": 4.5e-06,
      "loss": 10.9165,
      "mean_token_accuracy": 0.0003424657508730888,
      "num_tokens": 163840.0,
      "step": 10
    },
    {
      "entropy": 10.690902614593506,
      "epoch": 0.0016,
      "grad_norm": 10.0625,
      "learning_rate": 7e-06,
      "loss": 10.5112,
      "mean_token_accuracy": 0.04739481443539262,
      "num_tokens": 245760.0,
      "step": 15
    },
    {
      "entropy": 10.675402927398682,
      "epoch": 0.0021333333333333334,
      "grad_norm": 5.65625,
      "learning_rate": 9.5e-06,
      "loss": 10.0191,
      "mean_token_accuracy": 0.06303815953433514,
      "num_tokens": 327680.0,
      "step": 20
    },
    {
      "entropy": 10.631837749481202,
      "epoch": 0.0026666666666666666,
      "grad_norm": 3.546875,
      "learning_rate": 1.2e-05,
      "loss": 9.6081,
      "mean_token_accuracy": 0.0619618397206068,
      "num_tokens": 409600.0,
      "step": 25
    },
    {
      "entropy": 10.630702686309814,
      "epoch": 0.0032,
      "grad_norm": 3.03125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.3958,
      "mean_token_accuracy": 0.06272015646100045,
      "num_tokens": 491520.0,
      "step": 30
    },
    {
      "entropy": 10.616136837005616,
      "epoch": 0.0037333333333333333,
      "grad_norm": 3.046875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.2705,
      "mean_token_accuracy": 0.06243884563446045,
      "num_tokens": 573440.0,
      "step": 35
    },
    {
      "entropy": 10.615250396728516,
      "epoch": 0.004266666666666667,
      "grad_norm": 2.96875,
      "learning_rate": 1.95e-05,
      "loss": 9.0997,
      "mean_token_accuracy": 0.06679305247962475,
      "num_tokens": 655360.0,
      "step": 40
    },
    {
      "entropy": 10.601798248291015,
      "epoch": 0.0048,
      "grad_norm": 2.90625,
      "learning_rate": 2.2e-05,
      "loss": 8.9015,
      "mean_token_accuracy": 0.08314579203724862,
      "num_tokens": 737280.0,
      "step": 45
    },
    {
      "entropy": 10.51156063079834,
      "epoch": 0.005333333333333333,
      "grad_norm": 2.71875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 8.7423,
      "mean_token_accuracy": 0.08902886360883713,
      "num_tokens": 819200.0,
      "step": 50
    },
    {
      "entropy": 10.398311328887939,
      "epoch": 0.005866666666666667,
      "grad_norm": 2.625,
      "learning_rate": 2.7e-05,
      "loss": 8.604,
      "mean_token_accuracy": 0.09022749587893486,
      "num_tokens": 901120.0,
      "step": 55
    },
    {
      "entropy": 10.375072383880616,
      "epoch": 0.0064,
      "grad_norm": 2.6875,
      "learning_rate": 2.95e-05,
      "loss": 8.4781,
      "mean_token_accuracy": 0.08942025527358055,
      "num_tokens": 983040.0,
      "step": 60
    },
    {
      "entropy": 10.373117828369141,
      "epoch": 0.006933333333333333,
      "grad_norm": 2.59375,
      "learning_rate": 3.2e-05,
      "loss": 8.3278,
      "mean_token_accuracy": 0.09103473573923111,
      "num_tokens": 1064960.0,
      "step": 65
    },
    {
      "entropy": 10.389276027679443,
      "epoch": 0.007466666666666667,
      "grad_norm": 2.65625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 8.1902,
      "mean_token_accuracy": 0.08838062658905983,
      "num_tokens": 1146880.0,
      "step": 70
    },
    {
      "entropy": 10.346378612518311,
      "epoch": 0.008,
      "grad_norm": 2.53125,
      "learning_rate": 3.7e-05,
      "loss": 8.0119,
      "mean_token_accuracy": 0.09025195613503456,
      "num_tokens": 1228800.0,
      "step": 75
    },
    {
      "entropy": 10.33881072998047,
      "epoch": 0.008533333333333334,
      "grad_norm": 2.515625,
      "learning_rate": 3.95e-05,
      "loss": 7.8607,
      "mean_token_accuracy": 0.08929794505238534,
      "num_tokens": 1310720.0,
      "step": 80
    },
    {
      "entropy": 10.319882678985596,
      "epoch": 0.009066666666666667,
      "grad_norm": 2.578125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.6599,
      "mean_token_accuracy": 0.09436154589056969,
      "num_tokens": 1392640.0,
      "step": 85
    },
    {
      "entropy": 10.274563980102538,
      "epoch": 0.0096,
      "grad_norm": 2.484375,
      "learning_rate": 4.45e-05,
      "loss": 7.4498,
      "mean_token_accuracy": 0.09596379622817039,
      "num_tokens": 1474560.0,
      "step": 90
    },
    {
      "entropy": 10.254796314239503,
      "epoch": 0.010133333333333333,
      "grad_norm": 2.484375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.2388,
      "mean_token_accuracy": 0.09705234840512275,
      "num_tokens": 1556480.0,
      "step": 95
    },
    {
      "entropy": 10.20362606048584,
      "epoch": 0.010666666666666666,
      "grad_norm": 2.453125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.9913,
      "mean_token_accuracy": 0.10337573289871216,
      "num_tokens": 1638400.0,
      "step": 100
    },
    {
      "entropy": 10.102453422546386,
      "epoch": 0.0112,
      "grad_norm": 2.8125,
      "learning_rate": 5.2e-05,
      "loss": 6.7199,
      "mean_token_accuracy": 0.11284246444702148,
      "num_tokens": 1720320.0,
      "step": 105
    },
    {
      "entropy": 10.014357852935792,
      "epoch": 0.011733333333333333,
      "grad_norm": 3.171875,
      "learning_rate": 5.45e-05,
      "loss": 6.4546,
      "mean_token_accuracy": 0.12188111543655396,
      "num_tokens": 1802240.0,
      "step": 110
    },
    {
      "entropy": 9.865539932250977,
      "epoch": 0.012266666666666667,
      "grad_norm": 2.234375,
      "learning_rate": 5.7e-05,
      "loss": 6.1614,
      "mean_token_accuracy": 0.12569716200232506,
      "num_tokens": 1884160.0,
      "step": 115
    },
    {
      "entropy": 9.654939365386962,
      "epoch": 0.0128,
      "grad_norm": 2.140625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 5.8674,
      "mean_token_accuracy": 0.1292685903608799,
      "num_tokens": 1966080.0,
      "step": 120
    },
    {
      "entropy": 9.285263824462891,
      "epoch": 0.013333333333333334,
      "grad_norm": 2.125,
      "learning_rate": 6.2e-05,
      "loss": 5.5809,
      "mean_token_accuracy": 0.13335371538996696,
      "num_tokens": 2048000.0,
      "step": 125
    },
    {
      "entropy": 8.893216991424561,
      "epoch": 0.013866666666666666,
      "grad_norm": 1.84375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 5.3149,
      "mean_token_accuracy": 0.13040606752038003,
      "num_tokens": 2129920.0,
      "step": 130
    },
    {
      "entropy": 8.3220458984375,
      "epoch": 0.0144,
      "grad_norm": 2.171875,
      "learning_rate": 6.7e-05,
      "loss": 5.0512,
      "mean_token_accuracy": 0.13522505015134811,
      "num_tokens": 2211840.0,
      "step": 135
    },
    {
      "entropy": 7.719539833068848,
      "epoch": 0.014933333333333333,
      "grad_norm": 1.359375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 4.8233,
      "mean_token_accuracy": 0.13852739930152894,
      "num_tokens": 2293760.0,
      "step": 140
    },
    {
      "entropy": 7.1180624008178714,
      "epoch": 0.015466666666666667,
      "grad_norm": 1.265625,
      "learning_rate": 7.2e-05,
      "loss": 4.6641,
      "mean_token_accuracy": 0.13579990416765214,
      "num_tokens": 2375680.0,
      "step": 145
    },
    {
      "entropy": 6.465722370147705,
      "epoch": 0.016,
      "grad_norm": 0.9765625,
      "learning_rate": 7.45e-05,
      "loss": 4.5133,
      "mean_token_accuracy": 0.13872309327125548,
      "num_tokens": 2457600.0,
      "step": 150
    },
    {
      "entropy": 5.973577833175659,
      "epoch": 0.016533333333333334,
      "grad_norm": 1.125,
      "learning_rate": 7.7e-05,
      "loss": 4.4339,
      "mean_token_accuracy": 0.13631360232830048,
      "num_tokens": 2539520.0,
      "step": 155
    },
    {
      "entropy": 5.60869870185852,
      "epoch": 0.017066666666666667,
      "grad_norm": 1.1484375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 4.3677,
      "mean_token_accuracy": 0.13697406873106957,
      "num_tokens": 2621440.0,
      "step": 160
    },
    {
      "entropy": 5.379635667800903,
      "epoch": 0.0176,
      "grad_norm": 1.140625,
      "learning_rate": 8.2e-05,
      "loss": 4.3299,
      "mean_token_accuracy": 0.13913894593715667,
      "num_tokens": 2703360.0,
      "step": 165
    },
    {
      "entropy": 5.209587240219117,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.734375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 4.3036,
      "mean_token_accuracy": 0.1401174172759056,
      "num_tokens": 2785280.0,
      "step": 170
    },
    {
      "entropy": 5.061366176605224,
      "epoch": 0.018666666666666668,
      "grad_norm": 0.73046875,
      "learning_rate": 8.7e-05,
      "loss": 4.2876,
      "mean_token_accuracy": 0.13740215301513672,
      "num_tokens": 2867200.0,
      "step": 175
    },
    {
      "entropy": 4.979794216156006,
      "epoch": 0.0192,
      "grad_norm": 0.6015625,
      "learning_rate": 8.95e-05,
      "loss": 4.2709,
      "mean_token_accuracy": 0.13915117383003234,
      "num_tokens": 2949120.0,
      "step": 180
    },
    {
      "entropy": 4.908948373794556,
      "epoch": 0.019733333333333332,
      "grad_norm": 0.609375,
      "learning_rate": 9.2e-05,
      "loss": 4.2632,
      "mean_token_accuracy": 0.14059442281723022,
      "num_tokens": 3031040.0,
      "step": 185
    },
    {
      "entropy": 4.850634813308716,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.79296875,
      "learning_rate": 9.45e-05,
      "loss": 4.2582,
      "mean_token_accuracy": 0.13990949243307113,
      "num_tokens": 3112960.0,
      "step": 190
    },
    {
      "entropy": 4.786648368835449,
      "epoch": 0.0208,
      "grad_norm": 1.0625,
      "learning_rate": 9.7e-05,
      "loss": 4.2419,
      "mean_token_accuracy": 0.14218444079160691,
      "num_tokens": 3194880.0,
      "step": 195
    },
    {
      "entropy": 4.758226346969605,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.8359375,
      "learning_rate": 9.95e-05,
      "loss": 4.2357,
      "mean_token_accuracy": 0.14488747715950012,
      "num_tokens": 3276800.0,
      "step": 200
    },
    {
      "entropy": 4.722541952133179,
      "epoch": 0.021866666666666666,
      "grad_norm": 0.60546875,
      "learning_rate": 0.000102,
      "loss": 4.2313,
      "mean_token_accuracy": 0.14301614165306092,
      "num_tokens": 3358720.0,
      "step": 205
    },
    {
      "entropy": 4.7076897621154785,
      "epoch": 0.0224,
      "grad_norm": 0.71875,
      "learning_rate": 0.00010449999999999999,
      "loss": 4.2287,
      "mean_token_accuracy": 0.1454378679394722,
      "num_tokens": 3440640.0,
      "step": 210
    },
    {
      "entropy": 4.667892217636108,
      "epoch": 0.022933333333333333,
      "grad_norm": 0.62890625,
      "learning_rate": 0.000107,
      "loss": 4.2144,
      "mean_token_accuracy": 0.14744373708963393,
      "num_tokens": 3522560.0,
      "step": 215
    },
    {
      "entropy": 4.640421724319458,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.625,
      "learning_rate": 0.0001095,
      "loss": 4.1981,
      "mean_token_accuracy": 0.15086839497089385,
      "num_tokens": 3604480.0,
      "step": 220
    },
    {
      "entropy": 4.632302331924438,
      "epoch": 0.024,
      "grad_norm": 0.796875,
      "learning_rate": 0.000112,
      "loss": 4.2028,
      "mean_token_accuracy": 0.1504770040512085,
      "num_tokens": 3686400.0,
      "step": 225
    },
    {
      "entropy": 4.586120748519898,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0001145,
      "loss": 4.1824,
      "mean_token_accuracy": 0.15565068572759627,
      "num_tokens": 3768320.0,
      "step": 230
    },
    {
      "entropy": 4.5880763053894045,
      "epoch": 0.025066666666666668,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00011700000000000001,
      "loss": 4.1709,
      "mean_token_accuracy": 0.15836594849824906,
      "num_tokens": 3850240.0,
      "step": 235
    },
    {
      "entropy": 4.547198820114136,
      "epoch": 0.0256,
      "grad_norm": 1.484375,
      "learning_rate": 0.00011949999999999999,
      "loss": 4.1814,
      "mean_token_accuracy": 0.1539872795343399,
      "num_tokens": 3932160.0,
      "step": 240
    },
    {
      "entropy": 4.554472780227661,
      "epoch": 0.026133333333333335,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000122,
      "loss": 4.1648,
      "mean_token_accuracy": 0.15847602784633635,
      "num_tokens": 4014080.0,
      "step": 245
    },
    {
      "entropy": 4.515098428726196,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0001245,
      "loss": 4.1435,
      "mean_token_accuracy": 0.16312377899885178,
      "num_tokens": 4096000.0,
      "step": 250
    },
    {
      "entropy": 4.485965967178345,
      "epoch": 0.0272,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000127,
      "loss": 4.1324,
      "mean_token_accuracy": 0.16243884563446045,
      "num_tokens": 4177920.0,
      "step": 255
    },
    {
      "entropy": 4.493349647521972,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001295,
      "loss": 4.127,
      "mean_token_accuracy": 0.16493395268917083,
      "num_tokens": 4259840.0,
      "step": 260
    },
    {
      "entropy": 4.452348852157593,
      "epoch": 0.028266666666666666,
      "grad_norm": 0.74609375,
      "learning_rate": 0.000132,
      "loss": 4.1071,
      "mean_token_accuracy": 0.16856653690338136,
      "num_tokens": 4341760.0,
      "step": 265
    },
    {
      "entropy": 4.4304321765899655,
      "epoch": 0.0288,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00013450000000000002,
      "loss": 4.0844,
      "mean_token_accuracy": 0.17174657434225082,
      "num_tokens": 4423680.0,
      "step": 270
    },
    {
      "entropy": 4.423278045654297,
      "epoch": 0.029333333333333333,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00013700000000000002,
      "loss": 4.0609,
      "mean_token_accuracy": 0.1772749528288841,
      "num_tokens": 4505600.0,
      "step": 275
    },
    {
      "entropy": 4.391708183288574,
      "epoch": 0.029866666666666666,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001395,
      "loss": 4.0314,
      "mean_token_accuracy": 0.1812133088707924,
      "num_tokens": 4587520.0,
      "step": 280
    },
    {
      "entropy": 4.36551685333252,
      "epoch": 0.0304,
      "grad_norm": 1.015625,
      "learning_rate": 0.00014199999999999998,
      "loss": 4.001,
      "mean_token_accuracy": 0.1842098817229271,
      "num_tokens": 4669440.0,
      "step": 285
    },
    {
      "entropy": 4.3382508754730225,
      "epoch": 0.030933333333333334,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001445,
      "loss": 3.9539,
      "mean_token_accuracy": 0.19268591105937957,
      "num_tokens": 4751360.0,
      "step": 290
    },
    {
      "entropy": 4.313802337646484,
      "epoch": 0.031466666666666664,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000147,
      "loss": 3.9225,
      "mean_token_accuracy": 0.19373776912689208,
      "num_tokens": 4833280.0,
      "step": 295
    },
    {
      "entropy": 4.288891696929932,
      "epoch": 0.032,
      "grad_norm": 0.953125,
      "learning_rate": 0.0001495,
      "loss": 3.8903,
      "mean_token_accuracy": 0.19465508908033372,
      "num_tokens": 4915200.0,
      "step": 300
    },
    {
      "entropy": 4.258247137069702,
      "epoch": 0.03253333333333333,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000152,
      "loss": 3.8566,
      "mean_token_accuracy": 0.19915606528520585,
      "num_tokens": 4997120.0,
      "step": 305
    },
    {
      "entropy": 4.232576513290406,
      "epoch": 0.03306666666666667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00015450000000000001,
      "loss": 3.8529,
      "mean_token_accuracy": 0.19697896540164947,
      "num_tokens": 5079040.0,
      "step": 310
    },
    {
      "entropy": 4.207263469696045,
      "epoch": 0.0336,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000157,
      "loss": 3.8227,
      "mean_token_accuracy": 0.19991438388824462,
      "num_tokens": 5160960.0,
      "step": 315
    },
    {
      "entropy": 4.1795470237731935,
      "epoch": 0.034133333333333335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001595,
      "loss": 3.8027,
      "mean_token_accuracy": 0.20426859110593795,
      "num_tokens": 5242880.0,
      "step": 320
    },
    {
      "entropy": 4.1533502578735355,
      "epoch": 0.034666666666666665,
      "grad_norm": 1.046875,
      "learning_rate": 0.000162,
      "loss": 3.7818,
      "mean_token_accuracy": 0.2062255397439003,
      "num_tokens": 5324800.0,
      "step": 325
    },
    {
      "entropy": 4.1572949409484865,
      "epoch": 0.0352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016450000000000001,
      "loss": 3.7669,
      "mean_token_accuracy": 0.2083047956228256,
      "num_tokens": 5406720.0,
      "step": 330
    },
    {
      "entropy": 4.130109024047852,
      "epoch": 0.03573333333333333,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016700000000000002,
      "loss": 3.747,
      "mean_token_accuracy": 0.20991927534341812,
      "num_tokens": 5488640.0,
      "step": 335
    },
    {
      "entropy": 4.128541278839111,
      "epoch": 0.03626666666666667,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00016950000000000003,
      "loss": 3.7415,
      "mean_token_accuracy": 0.20881849378347397,
      "num_tokens": 5570560.0,
      "step": 340
    },
    {
      "entropy": 4.096694231033325,
      "epoch": 0.0368,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00017199999999999998,
      "loss": 3.713,
      "mean_token_accuracy": 0.2130993127822876,
      "num_tokens": 5652480.0,
      "step": 345
    },
    {
      "entropy": 4.0859273910522464,
      "epoch": 0.037333333333333336,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00017449999999999999,
      "loss": 3.689,
      "mean_token_accuracy": 0.21895792782306672,
      "num_tokens": 5734400.0,
      "step": 350
    },
    {
      "entropy": 4.069272518157959,
      "epoch": 0.037866666666666667,
      "grad_norm": 1.21875,
      "learning_rate": 0.000177,
      "loss": 3.6817,
      "mean_token_accuracy": 0.21903131008148194,
      "num_tokens": 5816320.0,
      "step": 355
    },
    {
      "entropy": 4.05731840133667,
      "epoch": 0.0384,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001795,
      "loss": 3.6675,
      "mean_token_accuracy": 0.21892123222351073,
      "num_tokens": 5898240.0,
      "step": 360
    },
    {
      "entropy": 4.031953763961792,
      "epoch": 0.038933333333333334,
      "grad_norm": 1.328125,
      "learning_rate": 0.000182,
      "loss": 3.6507,
      "mean_token_accuracy": 0.2229207456111908,
      "num_tokens": 5980160.0,
      "step": 365
    },
    {
      "entropy": 4.035226678848266,
      "epoch": 0.039466666666666664,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001845,
      "loss": 3.6486,
      "mean_token_accuracy": 0.22133072316646576,
      "num_tokens": 6062080.0,
      "step": 370
    },
    {
      "entropy": 4.009369921684265,
      "epoch": 0.04,
      "grad_norm": 1.625,
      "learning_rate": 0.000187,
      "loss": 3.6404,
      "mean_token_accuracy": 0.2208047941327095,
      "num_tokens": 6144000.0,
      "step": 375
    },
    {
      "entropy": 4.010466837882996,
      "epoch": 0.04053333333333333,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001895,
      "loss": 3.625,
      "mean_token_accuracy": 0.2228473573923111,
      "num_tokens": 6225920.0,
      "step": 380
    },
    {
      "entropy": 3.9951345205307005,
      "epoch": 0.04106666666666667,
      "grad_norm": 1.359375,
      "learning_rate": 0.000192,
      "loss": 3.6282,
      "mean_token_accuracy": 0.22034001648426055,
      "num_tokens": 6307840.0,
      "step": 385
    },
    {
      "entropy": 3.9747177362442017,
      "epoch": 0.0416,
      "grad_norm": 1.890625,
      "learning_rate": 0.0001945,
      "loss": 3.5954,
      "mean_token_accuracy": 0.22756849527359008,
      "num_tokens": 6389760.0,
      "step": 390
    },
    {
      "entropy": 3.97040593624115,
      "epoch": 0.042133333333333335,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019700000000000002,
      "loss": 3.5895,
      "mean_token_accuracy": 0.23017368018627166,
      "num_tokens": 6471680.0,
      "step": 395
    },
    {
      "entropy": 3.9500073432922362,
      "epoch": 0.042666666666666665,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00019950000000000002,
      "loss": 3.5823,
      "mean_token_accuracy": 0.22897505015134811,
      "num_tokens": 6553600.0,
      "step": 400
    },
    {
      "entropy": 3.9455363273620607,
      "epoch": 0.0432,
      "grad_norm": 1.296875,
      "learning_rate": 0.000202,
      "loss": 3.5793,
      "mean_token_accuracy": 0.2271648719906807,
      "num_tokens": 6635520.0,
      "step": 405
    },
    {
      "entropy": 3.9388590335845945,
      "epoch": 0.04373333333333333,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020449999999999998,
      "loss": 3.5722,
      "mean_token_accuracy": 0.22818004041910173,
      "num_tokens": 6717440.0,
      "step": 410
    },
    {
      "entropy": 3.921509313583374,
      "epoch": 0.04426666666666667,
      "grad_norm": 1.671875,
      "learning_rate": 0.000207,
      "loss": 3.5599,
      "mean_token_accuracy": 0.22937866896390915,
      "num_tokens": 6799360.0,
      "step": 415
    },
    {
      "entropy": 3.93716356754303,
      "epoch": 0.0448,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002095,
      "loss": 3.5664,
      "mean_token_accuracy": 0.2296355187892914,
      "num_tokens": 6881280.0,
      "step": 420
    },
    {
      "entropy": 3.9049474239349364,
      "epoch": 0.04533333333333334,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000212,
      "loss": 3.5563,
      "mean_token_accuracy": 0.22797211408615112,
      "num_tokens": 6963200.0,
      "step": 425
    },
    {
      "entropy": 3.9092701196670534,
      "epoch": 0.04586666666666667,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0002145,
      "loss": 3.5386,
      "mean_token_accuracy": 0.23233855217695237,
      "num_tokens": 7045120.0,
      "step": 430
    },
    {
      "entropy": 3.8930064916610716,
      "epoch": 0.0464,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021700000000000002,
      "loss": 3.5333,
      "mean_token_accuracy": 0.23288894146680833,
      "num_tokens": 7127040.0,
      "step": 435
    },
    {
      "entropy": 3.8910942554473875,
      "epoch": 0.046933333333333334,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002195,
      "loss": 3.5452,
      "mean_token_accuracy": 0.229415363073349,
      "num_tokens": 7208960.0,
      "step": 440
    },
    {
      "entropy": 3.8705926656723024,
      "epoch": 0.047466666666666664,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000222,
      "loss": 3.5283,
      "mean_token_accuracy": 0.2326076328754425,
      "num_tokens": 7290880.0,
      "step": 445
    },
    {
      "entropy": 3.8792372226715086,
      "epoch": 0.048,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002245,
      "loss": 3.5378,
      "mean_token_accuracy": 0.2297945201396942,
      "num_tokens": 7372800.0,
      "step": 450
    },
    {
      "entropy": 3.86421058177948,
      "epoch": 0.04853333333333333,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022700000000000002,
      "loss": 3.5131,
      "mean_token_accuracy": 0.23609344363212587,
      "num_tokens": 7454720.0,
      "step": 455
    },
    {
      "entropy": 3.8421786546707155,
      "epoch": 0.04906666666666667,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022950000000000002,
      "loss": 3.4924,
      "mean_token_accuracy": 0.23913894295692445,
      "num_tokens": 7536640.0,
      "step": 460
    },
    {
      "entropy": 3.8420616388320923,
      "epoch": 0.0496,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023200000000000003,
      "loss": 3.5053,
      "mean_token_accuracy": 0.23378180116415023,
      "num_tokens": 7618560.0,
      "step": 465
    },
    {
      "entropy": 3.833560061454773,
      "epoch": 0.050133333333333335,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023449999999999998,
      "loss": 3.4973,
      "mean_token_accuracy": 0.23490704447031022,
      "num_tokens": 7700480.0,
      "step": 470
    },
    {
      "entropy": 3.8142281055450438,
      "epoch": 0.050666666666666665,
      "grad_norm": 1.546875,
      "learning_rate": 0.000237,
      "loss": 3.4899,
      "mean_token_accuracy": 0.2363869845867157,
      "num_tokens": 7782400.0,
      "step": 475
    },
    {
      "entropy": 3.8285669088363647,
      "epoch": 0.0512,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002395,
      "loss": 3.4914,
      "mean_token_accuracy": 0.23606898188591002,
      "num_tokens": 7864320.0,
      "step": 480
    },
    {
      "entropy": 3.8081378698349,
      "epoch": 0.05173333333333333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000242,
      "loss": 3.4818,
      "mean_token_accuracy": 0.2354941263794899,
      "num_tokens": 7946240.0,
      "step": 485
    },
    {
      "entropy": 3.8174705028533937,
      "epoch": 0.05226666666666667,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002445,
      "loss": 3.4804,
      "mean_token_accuracy": 0.23588551729917526,
      "num_tokens": 8028160.0,
      "step": 490
    },
    {
      "entropy": 3.807200789451599,
      "epoch": 0.0528,
      "grad_norm": 2.140625,
      "learning_rate": 0.000247,
      "loss": 3.48,
      "mean_token_accuracy": 0.23435665369033815,
      "num_tokens": 8110080.0,
      "step": 495
    },
    {
      "entropy": 3.79546959400177,
      "epoch": 0.05333333333333334,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002495,
      "loss": 3.4675,
      "mean_token_accuracy": 0.2389554798603058,
      "num_tokens": 8192000.0,
      "step": 500
    },
    {
      "entropy": 3.7889782667160032,
      "epoch": 0.05386666666666667,
      "grad_norm": 1.765625,
      "learning_rate": 0.000252,
      "loss": 3.4667,
      "mean_token_accuracy": 0.23646037131547928,
      "num_tokens": 8273920.0,
      "step": 505
    },
    {
      "entropy": 3.785716485977173,
      "epoch": 0.0544,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0002545,
      "loss": 3.455,
      "mean_token_accuracy": 0.24015410989522934,
      "num_tokens": 8355840.0,
      "step": 510
    },
    {
      "entropy": 3.764500856399536,
      "epoch": 0.054933333333333334,
      "grad_norm": 2.6875,
      "learning_rate": 0.000257,
      "loss": 3.4532,
      "mean_token_accuracy": 0.23701076358556747,
      "num_tokens": 8437760.0,
      "step": 515
    },
    {
      "entropy": 3.758707785606384,
      "epoch": 0.055466666666666664,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002595,
      "loss": 3.4453,
      "mean_token_accuracy": 0.23972602486610411,
      "num_tokens": 8519680.0,
      "step": 520
    },
    {
      "entropy": 3.770280122756958,
      "epoch": 0.056,
      "grad_norm": 1.515625,
      "learning_rate": 0.000262,
      "loss": 3.4543,
      "mean_token_accuracy": 0.23757338523864746,
      "num_tokens": 8601600.0,
      "step": 525
    },
    {
      "entropy": 3.7545013189315797,
      "epoch": 0.05653333333333333,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026450000000000003,
      "loss": 3.4479,
      "mean_token_accuracy": 0.2368272989988327,
      "num_tokens": 8683520.0,
      "step": 530
    },
    {
      "entropy": 3.737969398498535,
      "epoch": 0.05706666666666667,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026700000000000004,
      "loss": 3.4376,
      "mean_token_accuracy": 0.24004403203725816,
      "num_tokens": 8765440.0,
      "step": 535
    },
    {
      "entropy": 3.7428213119506837,
      "epoch": 0.0576,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00026950000000000005,
      "loss": 3.4349,
      "mean_token_accuracy": 0.24008072465658187,
      "num_tokens": 8847360.0,
      "step": 540
    },
    {
      "entropy": 3.735646629333496,
      "epoch": 0.058133333333333335,
      "grad_norm": 1.59375,
      "learning_rate": 0.00027200000000000005,
      "loss": 3.4296,
      "mean_token_accuracy": 0.24118150919675826,
      "num_tokens": 8929280.0,
      "step": 545
    },
    {
      "entropy": 3.7235782146453857,
      "epoch": 0.058666666666666666,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002745,
      "loss": 3.4253,
      "mean_token_accuracy": 0.24175636172294618,
      "num_tokens": 9011200.0,
      "step": 550
    },
    {
      "entropy": 3.7225298404693605,
      "epoch": 0.0592,
      "grad_norm": 1.609375,
      "learning_rate": 0.000277,
      "loss": 3.42,
      "mean_token_accuracy": 0.24081458002328873,
      "num_tokens": 9093120.0,
      "step": 555
    },
    {
      "entropy": 3.7187859535217287,
      "epoch": 0.05973333333333333,
      "grad_norm": 1.8125,
      "learning_rate": 0.0002795,
      "loss": 3.4261,
      "mean_token_accuracy": 0.24001957178115846,
      "num_tokens": 9175040.0,
      "step": 560
    },
    {
      "entropy": 3.713302421569824,
      "epoch": 0.06026666666666667,
      "grad_norm": 1.5,
      "learning_rate": 0.00028199999999999997,
      "loss": 3.4172,
      "mean_token_accuracy": 0.24145058691501617,
      "num_tokens": 9256960.0,
      "step": 565
    },
    {
      "entropy": 3.7087379932403564,
      "epoch": 0.0608,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002845,
      "loss": 3.4322,
      "mean_token_accuracy": 0.2365582212805748,
      "num_tokens": 9338880.0,
      "step": 570
    },
    {
      "entropy": 3.694662022590637,
      "epoch": 0.06133333333333333,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000287,
      "loss": 3.4108,
      "mean_token_accuracy": 0.24277152717113495,
      "num_tokens": 9420800.0,
      "step": 575
    },
    {
      "entropy": 3.7138469219207764,
      "epoch": 0.06186666666666667,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002895,
      "loss": 3.4101,
      "mean_token_accuracy": 0.24241682887077332,
      "num_tokens": 9502720.0,
      "step": 580
    },
    {
      "entropy": 3.7062520503997805,
      "epoch": 0.0624,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000292,
      "loss": 3.4279,
      "mean_token_accuracy": 0.23463796377182006,
      "num_tokens": 9584640.0,
      "step": 585
    },
    {
      "entropy": 3.68607177734375,
      "epoch": 0.06293333333333333,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002945,
      "loss": 3.407,
      "mean_token_accuracy": 0.24049657732248306,
      "num_tokens": 9666560.0,
      "step": 590
    },
    {
      "entropy": 3.689913105964661,
      "epoch": 0.06346666666666667,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000297,
      "loss": 3.4047,
      "mean_token_accuracy": 0.24015410542488097,
      "num_tokens": 9748480.0,
      "step": 595
    },
    {
      "entropy": 3.6928848028182983,
      "epoch": 0.064,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002995,
      "loss": 3.417,
      "mean_token_accuracy": 0.23721868842840194,
      "num_tokens": 9830400.0,
      "step": 600
    },
    {
      "entropy": 3.677217197418213,
      "epoch": 0.06453333333333333,
      "grad_norm": 1.8046875,
      "learning_rate": 0.000302,
      "loss": 3.398,
      "mean_token_accuracy": 0.23994618356227876,
      "num_tokens": 9912320.0,
      "step": 605
    },
    {
      "entropy": 3.673366165161133,
      "epoch": 0.06506666666666666,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003045,
      "loss": 3.391,
      "mean_token_accuracy": 0.24338307231664658,
      "num_tokens": 9994240.0,
      "step": 610
    },
    {
      "entropy": 3.6597073078155518,
      "epoch": 0.0656,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000307,
      "loss": 3.3986,
      "mean_token_accuracy": 0.24030087888240814,
      "num_tokens": 10076160.0,
      "step": 615
    },
    {
      "entropy": 3.661817741394043,
      "epoch": 0.06613333333333334,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003095,
      "loss": 3.3877,
      "mean_token_accuracy": 0.24142612367868424,
      "num_tokens": 10158080.0,
      "step": 620
    },
    {
      "entropy": 3.6684417009353636,
      "epoch": 0.06666666666666667,
      "grad_norm": 1.5,
      "learning_rate": 0.000312,
      "loss": 3.3958,
      "mean_token_accuracy": 0.24153620302677153,
      "num_tokens": 10240000.0,
      "step": 625
    },
    {
      "entropy": 3.655338168144226,
      "epoch": 0.0672,
      "grad_norm": 2.140625,
      "learning_rate": 0.0003145,
      "loss": 3.3992,
      "mean_token_accuracy": 0.2351272001862526,
      "num_tokens": 10321920.0,
      "step": 630
    },
    {
      "entropy": 3.652760720252991,
      "epoch": 0.06773333333333334,
      "grad_norm": 1.96875,
      "learning_rate": 0.000317,
      "loss": 3.3781,
      "mean_token_accuracy": 0.24770058542490006,
      "num_tokens": 10403840.0,
      "step": 635
    },
    {
      "entropy": 3.6520917892456053,
      "epoch": 0.06826666666666667,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003195,
      "loss": 3.3817,
      "mean_token_accuracy": 0.24147504866123198,
      "num_tokens": 10485760.0,
      "step": 640
    },
    {
      "entropy": 3.645389986038208,
      "epoch": 0.0688,
      "grad_norm": 1.296875,
      "learning_rate": 0.000322,
      "loss": 3.3778,
      "mean_token_accuracy": 0.2417563572525978,
      "num_tokens": 10567680.0,
      "step": 645
    },
    {
      "entropy": 3.6351272106170653,
      "epoch": 0.06933333333333333,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00032450000000000003,
      "loss": 3.3924,
      "mean_token_accuracy": 0.23959148526191712,
      "num_tokens": 10649600.0,
      "step": 650
    },
    {
      "entropy": 3.6490575313568114,
      "epoch": 0.06986666666666666,
      "grad_norm": 1.8125,
      "learning_rate": 0.00032700000000000003,
      "loss": 3.3816,
      "mean_token_accuracy": 0.23831947147846222,
      "num_tokens": 10731520.0,
      "step": 655
    },
    {
      "entropy": 3.643259358406067,
      "epoch": 0.0704,
      "grad_norm": 1.953125,
      "learning_rate": 0.00032950000000000004,
      "loss": 3.3803,
      "mean_token_accuracy": 0.23961594849824905,
      "num_tokens": 10813440.0,
      "step": 660
    },
    {
      "entropy": 3.6322882652282713,
      "epoch": 0.07093333333333333,
      "grad_norm": 1.578125,
      "learning_rate": 0.00033200000000000005,
      "loss": 3.3763,
      "mean_token_accuracy": 0.241169273853302,
      "num_tokens": 10895360.0,
      "step": 665
    },
    {
      "entropy": 3.6224374532699586,
      "epoch": 0.07146666666666666,
      "grad_norm": 1.625,
      "learning_rate": 0.00033450000000000005,
      "loss": 3.3591,
      "mean_token_accuracy": 0.24379892349243165,
      "num_tokens": 10977280.0,
      "step": 670
    },
    {
      "entropy": 3.6074854850769045,
      "epoch": 0.072,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000337,
      "loss": 3.3701,
      "mean_token_accuracy": 0.2412426620721817,
      "num_tokens": 11059200.0,
      "step": 675
    },
    {
      "entropy": 3.628856587409973,
      "epoch": 0.07253333333333334,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003395,
      "loss": 3.3693,
      "mean_token_accuracy": 0.2422822892665863,
      "num_tokens": 11141120.0,
      "step": 680
    },
    {
      "entropy": 3.6110815525054933,
      "epoch": 0.07306666666666667,
      "grad_norm": 1.625,
      "learning_rate": 0.000342,
      "loss": 3.3677,
      "mean_token_accuracy": 0.23987279683351517,
      "num_tokens": 11223040.0,
      "step": 685
    },
    {
      "entropy": 3.6009220361709593,
      "epoch": 0.0736,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034449999999999997,
      "loss": 3.3547,
      "mean_token_accuracy": 0.24192759543657302,
      "num_tokens": 11304960.0,
      "step": 690
    },
    {
      "entropy": 3.5970619678497315,
      "epoch": 0.07413333333333333,
      "grad_norm": 1.8125,
      "learning_rate": 0.000347,
      "loss": 3.3463,
      "mean_token_accuracy": 0.24505870640277863,
      "num_tokens": 11386880.0,
      "step": 695
    },
    {
      "entropy": 3.587267446517944,
      "epoch": 0.07466666666666667,
      "grad_norm": 2.3125,
      "learning_rate": 0.0003495,
      "loss": 3.3419,
      "mean_token_accuracy": 0.24344422668218613,
      "num_tokens": 11468800.0,
      "step": 700
    },
    {
      "entropy": 3.5908786058425903,
      "epoch": 0.0752,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000352,
      "loss": 3.3509,
      "mean_token_accuracy": 0.24026418775320052,
      "num_tokens": 11550720.0,
      "step": 705
    },
    {
      "entropy": 3.5929071426391603,
      "epoch": 0.07573333333333333,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003545,
      "loss": 3.3355,
      "mean_token_accuracy": 0.24363992214202881,
      "num_tokens": 11632640.0,
      "step": 710
    },
    {
      "entropy": 3.583715867996216,
      "epoch": 0.07626666666666666,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000357,
      "loss": 3.3393,
      "mean_token_accuracy": 0.2459515631198883,
      "num_tokens": 11714560.0,
      "step": 715
    },
    {
      "entropy": 3.589143085479736,
      "epoch": 0.0768,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0003595,
      "loss": 3.3369,
      "mean_token_accuracy": 0.24209882467985153,
      "num_tokens": 11796480.0,
      "step": 720
    },
    {
      "entropy": 3.5867936611175537,
      "epoch": 0.07733333333333334,
      "grad_norm": 3.03125,
      "learning_rate": 0.000362,
      "loss": 3.3431,
      "mean_token_accuracy": 0.24129158407449722,
      "num_tokens": 11878400.0,
      "step": 725
    },
    {
      "entropy": 3.59276020526886,
      "epoch": 0.07786666666666667,
      "grad_norm": 1.796875,
      "learning_rate": 0.0003645,
      "loss": 3.3434,
      "mean_token_accuracy": 0.24174412786960603,
      "num_tokens": 11960320.0,
      "step": 730
    },
    {
      "entropy": 3.5778300762176514,
      "epoch": 0.0784,
      "grad_norm": 1.71875,
      "learning_rate": 0.000367,
      "loss": 3.3414,
      "mean_token_accuracy": 0.24065557718276978,
      "num_tokens": 12042240.0,
      "step": 735
    },
    {
      "entropy": 3.597053050994873,
      "epoch": 0.07893333333333333,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0003695,
      "loss": 3.3516,
      "mean_token_accuracy": 0.2385273978114128,
      "num_tokens": 12124160.0,
      "step": 740
    },
    {
      "entropy": 3.5768983364105225,
      "epoch": 0.07946666666666667,
      "grad_norm": 1.5625,
      "learning_rate": 0.000372,
      "loss": 3.3231,
      "mean_token_accuracy": 0.2470768079161644,
      "num_tokens": 12206080.0,
      "step": 745
    },
    {
      "entropy": 3.5772984266281127,
      "epoch": 0.08,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0003745,
      "loss": 3.3338,
      "mean_token_accuracy": 0.24061888456344604,
      "num_tokens": 12288000.0,
      "step": 750
    },
    {
      "entropy": 3.566642928123474,
      "epoch": 0.08053333333333333,
      "grad_norm": 1.59375,
      "learning_rate": 0.000377,
      "loss": 3.3082,
      "mean_token_accuracy": 0.24603718072175979,
      "num_tokens": 12369920.0,
      "step": 755
    },
    {
      "entropy": 3.5493921518325804,
      "epoch": 0.08106666666666666,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0003795,
      "loss": 3.3097,
      "mean_token_accuracy": 0.24467954486608506,
      "num_tokens": 12451840.0,
      "step": 760
    },
    {
      "entropy": 3.563185977935791,
      "epoch": 0.0816,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000382,
      "loss": 3.3099,
      "mean_token_accuracy": 0.243309685587883,
      "num_tokens": 12533760.0,
      "step": 765
    },
    {
      "entropy": 3.5411034345626833,
      "epoch": 0.08213333333333334,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0003845,
      "loss": 3.3144,
      "mean_token_accuracy": 0.24328522235155106,
      "num_tokens": 12615680.0,
      "step": 770
    },
    {
      "entropy": 3.5681350946426393,
      "epoch": 0.08266666666666667,
      "grad_norm": 2.578125,
      "learning_rate": 0.00038700000000000003,
      "loss": 3.3221,
      "mean_token_accuracy": 0.24020303338766097,
      "num_tokens": 12697600.0,
      "step": 775
    },
    {
      "entropy": 3.5270207643508913,
      "epoch": 0.0832,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038950000000000003,
      "loss": 3.2966,
      "mean_token_accuracy": 0.2470645785331726,
      "num_tokens": 12779520.0,
      "step": 780
    },
    {
      "entropy": 3.5338781118392943,
      "epoch": 0.08373333333333334,
      "grad_norm": 5.625,
      "learning_rate": 0.00039200000000000004,
      "loss": 3.2976,
      "mean_token_accuracy": 0.24847113937139512,
      "num_tokens": 12861440.0,
      "step": 785
    },
    {
      "entropy": 3.5563634634017944,
      "epoch": 0.08426666666666667,
      "grad_norm": 2.1875,
      "learning_rate": 0.00039450000000000005,
      "loss": 3.3168,
      "mean_token_accuracy": 0.24625733941793443,
      "num_tokens": 12943360.0,
      "step": 790
    },
    {
      "entropy": 3.523889255523682,
      "epoch": 0.0848,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039700000000000005,
      "loss": 3.3121,
      "mean_token_accuracy": 0.24505870789289474,
      "num_tokens": 13025280.0,
      "step": 795
    },
    {
      "entropy": 3.5672359466552734,
      "epoch": 0.08533333333333333,
      "grad_norm": 3.5625,
      "learning_rate": 0.0003995,
      "loss": 3.3124,
      "mean_token_accuracy": 0.24388453960418702,
      "num_tokens": 13107200.0,
      "step": 800
    },
    {
      "entropy": 3.524198603630066,
      "epoch": 0.08586666666666666,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000402,
      "loss": 3.3066,
      "mean_token_accuracy": 0.24294275790452957,
      "num_tokens": 13189120.0,
      "step": 805
    },
    {
      "entropy": 3.535692071914673,
      "epoch": 0.0864,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004045,
      "loss": 3.299,
      "mean_token_accuracy": 0.24140166342258454,
      "num_tokens": 13271040.0,
      "step": 810
    },
    {
      "entropy": 3.5244675397872927,
      "epoch": 0.08693333333333333,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00040699999999999997,
      "loss": 3.2988,
      "mean_token_accuracy": 0.24274706244468688,
      "num_tokens": 13352960.0,
      "step": 815
    },
    {
      "entropy": 3.5047380208969114,
      "epoch": 0.08746666666666666,
      "grad_norm": 3.625,
      "learning_rate": 0.0004095,
      "loss": 3.2758,
      "mean_token_accuracy": 0.24700342416763305,
      "num_tokens": 13434880.0,
      "step": 820
    },
    {
      "entropy": 3.5343822240829468,
      "epoch": 0.088,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000412,
      "loss": 3.2917,
      "mean_token_accuracy": 0.2418297454714775,
      "num_tokens": 13516800.0,
      "step": 825
    },
    {
      "entropy": 3.5038384437561034,
      "epoch": 0.08853333333333334,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004145,
      "loss": 3.2755,
      "mean_token_accuracy": 0.24538894444704057,
      "num_tokens": 13598720.0,
      "step": 830
    },
    {
      "entropy": 3.486983823776245,
      "epoch": 0.08906666666666667,
      "grad_norm": 1.5625,
      "learning_rate": 0.000417,
      "loss": 3.2778,
      "mean_token_accuracy": 0.24581702351570128,
      "num_tokens": 13680640.0,
      "step": 835
    },
    {
      "entropy": 3.516296362876892,
      "epoch": 0.0896,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004195,
      "loss": 3.2707,
      "mean_token_accuracy": 0.24487524032592772,
      "num_tokens": 13762560.0,
      "step": 840
    },
    {
      "entropy": 3.5094950437545775,
      "epoch": 0.09013333333333333,
      "grad_norm": 1.765625,
      "learning_rate": 0.000422,
      "loss": 3.2893,
      "mean_token_accuracy": 0.24033757150173188,
      "num_tokens": 13844480.0,
      "step": 845
    },
    {
      "entropy": 3.4691035747528076,
      "epoch": 0.09066666666666667,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004245,
      "loss": 3.2601,
      "mean_token_accuracy": 0.24925391525030136,
      "num_tokens": 13926400.0,
      "step": 850
    },
    {
      "entropy": 3.5137253999710083,
      "epoch": 0.0912,
      "grad_norm": 2.015625,
      "learning_rate": 0.000427,
      "loss": 3.2983,
      "mean_token_accuracy": 0.24137720018625258,
      "num_tokens": 14008320.0,
      "step": 855
    },
    {
      "entropy": 3.4819485664367678,
      "epoch": 0.09173333333333333,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004295,
      "loss": 3.2709,
      "mean_token_accuracy": 0.24374999701976777,
      "num_tokens": 14090240.0,
      "step": 860
    },
    {
      "entropy": 3.4891334533691407,
      "epoch": 0.09226666666666666,
      "grad_norm": 1.609375,
      "learning_rate": 0.000432,
      "loss": 3.2745,
      "mean_token_accuracy": 0.2452421709895134,
      "num_tokens": 14172160.0,
      "step": 865
    },
    {
      "entropy": 3.4707372188568115,
      "epoch": 0.0928,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004345,
      "loss": 3.2686,
      "mean_token_accuracy": 0.24537671208381653,
      "num_tokens": 14254080.0,
      "step": 870
    },
    {
      "entropy": 3.4907490968704225,
      "epoch": 0.09333333333333334,
      "grad_norm": 1.765625,
      "learning_rate": 0.000437,
      "loss": 3.2725,
      "mean_token_accuracy": 0.24230675101280214,
      "num_tokens": 14336000.0,
      "step": 875
    },
    {
      "entropy": 3.470463490486145,
      "epoch": 0.09386666666666667,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004395,
      "loss": 3.2667,
      "mean_token_accuracy": 0.24382338672876358,
      "num_tokens": 14417920.0,
      "step": 880
    },
    {
      "entropy": 3.458239459991455,
      "epoch": 0.0944,
      "grad_norm": 2.71875,
      "learning_rate": 0.000442,
      "loss": 3.2557,
      "mean_token_accuracy": 0.24576810002326965,
      "num_tokens": 14499840.0,
      "step": 885
    },
    {
      "entropy": 3.44415397644043,
      "epoch": 0.09493333333333333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445,
      "loss": 3.2588,
      "mean_token_accuracy": 0.24496086090803146,
      "num_tokens": 14581760.0,
      "step": 890
    },
    {
      "entropy": 3.490961694717407,
      "epoch": 0.09546666666666667,
      "grad_norm": 2.140625,
      "learning_rate": 0.000447,
      "loss": 3.2802,
      "mean_token_accuracy": 0.2430283769965172,
      "num_tokens": 14663680.0,
      "step": 895
    },
    {
      "entropy": 3.4478488683700563,
      "epoch": 0.096,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044950000000000003,
      "loss": 3.2655,
      "mean_token_accuracy": 0.24223336577415466,
      "num_tokens": 14745600.0,
      "step": 900
    },
    {
      "entropy": 3.433678936958313,
      "epoch": 0.09653333333333333,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045200000000000004,
      "loss": 3.2531,
      "mean_token_accuracy": 0.2459760308265686,
      "num_tokens": 14827520.0,
      "step": 905
    },
    {
      "entropy": 3.4747411966323853,
      "epoch": 0.09706666666666666,
      "grad_norm": 4.15625,
      "learning_rate": 0.00045450000000000004,
      "loss": 3.2898,
      "mean_token_accuracy": 0.24159735888242723,
      "num_tokens": 14909440.0,
      "step": 910
    },
    {
      "entropy": 3.4487999200820925,
      "epoch": 0.0976,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00045700000000000005,
      "loss": 3.2673,
      "mean_token_accuracy": 0.24547455757856368,
      "num_tokens": 14991360.0,
      "step": 915
    },
    {
      "entropy": 3.4375893115997314,
      "epoch": 0.09813333333333334,
      "grad_norm": 2.390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 3.2549,
      "mean_token_accuracy": 0.24494863152503968,
      "num_tokens": 15073280.0,
      "step": 920
    },
    {
      "entropy": 3.4646376848220823,
      "epoch": 0.09866666666666667,
      "grad_norm": 3.78125,
      "learning_rate": 0.000462,
      "loss": 3.2728,
      "mean_token_accuracy": 0.24405577182769775,
      "num_tokens": 15155200.0,
      "step": 925
    },
    {
      "entropy": 3.445885944366455,
      "epoch": 0.0992,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004645,
      "loss": 3.2696,
      "mean_token_accuracy": 0.24417808204889296,
      "num_tokens": 15237120.0,
      "step": 930
    },
    {
      "entropy": 3.439109373092651,
      "epoch": 0.09973333333333333,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000467,
      "loss": 3.2607,
      "mean_token_accuracy": 0.24268590956926345,
      "num_tokens": 15319040.0,
      "step": 935
    },
    {
      "entropy": 3.443026566505432,
      "epoch": 0.10026666666666667,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004695,
      "loss": 3.2566,
      "mean_token_accuracy": 0.24390900284051895,
      "num_tokens": 15400960.0,
      "step": 940
    },
    {
      "entropy": 3.4182130336761474,
      "epoch": 0.1008,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000472,
      "loss": 3.2598,
      "mean_token_accuracy": 0.24549902081489564,
      "num_tokens": 15482880.0,
      "step": 945
    },
    {
      "entropy": 3.4506216049194336,
      "epoch": 0.10133333333333333,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004745,
      "loss": 3.2705,
      "mean_token_accuracy": 0.24322407245635985,
      "num_tokens": 15564800.0,
      "step": 950
    },
    {
      "entropy": 3.4289287090301515,
      "epoch": 0.10186666666666666,
      "grad_norm": 2.046875,
      "learning_rate": 0.000477,
      "loss": 3.2487,
      "mean_token_accuracy": 0.245193250477314,
      "num_tokens": 15646720.0,
      "step": 955
    },
    {
      "entropy": 3.391101622581482,
      "epoch": 0.1024,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004795,
      "loss": 3.2481,
      "mean_token_accuracy": 0.24557240456342697,
      "num_tokens": 15728640.0,
      "step": 960
    },
    {
      "entropy": 3.4241345167160033,
      "epoch": 0.10293333333333334,
      "grad_norm": 2.671875,
      "learning_rate": 0.000482,
      "loss": 3.2734,
      "mean_token_accuracy": 0.24163405001163482,
      "num_tokens": 15810560.0,
      "step": 965
    },
    {
      "entropy": 3.4429741144180297,
      "epoch": 0.10346666666666667,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004845,
      "loss": 3.2746,
      "mean_token_accuracy": 0.2399951085448265,
      "num_tokens": 15892480.0,
      "step": 970
    },
    {
      "entropy": 3.415851044654846,
      "epoch": 0.104,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000487,
      "loss": 3.2572,
      "mean_token_accuracy": 0.24127935618162155,
      "num_tokens": 15974400.0,
      "step": 975
    },
    {
      "entropy": 3.3974807024002076,
      "epoch": 0.10453333333333334,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004895,
      "loss": 3.2386,
      "mean_token_accuracy": 0.2478840485215187,
      "num_tokens": 16056320.0,
      "step": 980
    },
    {
      "entropy": 3.4176392793655395,
      "epoch": 0.10506666666666667,
      "grad_norm": 1.671875,
      "learning_rate": 0.000492,
      "loss": 3.2544,
      "mean_token_accuracy": 0.24623287618160247,
      "num_tokens": 16138240.0,
      "step": 985
    },
    {
      "entropy": 3.3827028274536133,
      "epoch": 0.1056,
      "grad_norm": 1.375,
      "learning_rate": 0.0004945,
      "loss": 3.2418,
      "mean_token_accuracy": 0.24362768977880478,
      "num_tokens": 16220160.0,
      "step": 990
    },
    {
      "entropy": 3.3921529054641724,
      "epoch": 0.10613333333333333,
      "grad_norm": 2.578125,
      "learning_rate": 0.000497,
      "loss": 3.2305,
      "mean_token_accuracy": 0.2463918760418892,
      "num_tokens": 16302080.0,
      "step": 995
    },
    {
      "entropy": 3.3886776208877563,
      "epoch": 0.10666666666666667,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995,
      "loss": 3.2352,
      "mean_token_accuracy": 0.24684442281723024,
      "num_tokens": 16384000.0,
      "step": 1000
    },
    {
      "entropy": 3.3688613176345825,
      "epoch": 0.1072,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499998026082006,
      "loss": 3.2335,
      "mean_token_accuracy": 0.243615460395813,
      "num_tokens": 16465920.0,
      "step": 1005
    },
    {
      "entropy": 3.3817806243896484,
      "epoch": 0.10773333333333333,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004999900070995136,
      "loss": 3.2333,
      "mean_token_accuracy": 0.24441046714782716,
      "num_tokens": 16547840.0,
      "step": 1010
    },
    {
      "entropy": 3.36424081325531,
      "epoch": 0.10826666666666666,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999758199023239,
      "loss": 3.2249,
      "mean_token_accuracy": 0.244019079208374,
      "num_tokens": 16629760.0,
      "step": 1015
    },
    {
      "entropy": 3.381114053726196,
      "epoch": 0.1088,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004999554648793858,
      "loss": 3.2317,
      "mean_token_accuracy": 0.24527886658906936,
      "num_tokens": 16711680.0,
      "step": 1020
    },
    {
      "entropy": 3.3893242359161375,
      "epoch": 0.10933333333333334,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004999289425887425,
      "loss": 3.2447,
      "mean_token_accuracy": 0.24214774817228318,
      "num_tokens": 16793600.0,
      "step": 1025
    },
    {
      "entropy": 3.356892943382263,
      "epoch": 0.10986666666666667,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004998962537575161,
      "loss": 3.2355,
      "mean_token_accuracy": 0.24450831413269042,
      "num_tokens": 16875520.0,
      "step": 1030
    },
    {
      "entropy": 3.3502973079681397,
      "epoch": 0.1104,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004998573992818874,
      "loss": 3.2217,
      "mean_token_accuracy": 0.24554794579744338,
      "num_tokens": 16957440.0,
      "step": 1035
    },
    {
      "entropy": 3.371893858909607,
      "epoch": 0.11093333333333333,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004998123802270715,
      "loss": 3.2305,
      "mean_token_accuracy": 0.24557240754365922,
      "num_tokens": 17039360.0,
      "step": 1040
    },
    {
      "entropy": 3.380908799171448,
      "epoch": 0.11146666666666667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997611978272886,
      "loss": 3.254,
      "mean_token_accuracy": 0.23850293606519699,
      "num_tokens": 17121280.0,
      "step": 1045
    },
    {
      "entropy": 3.369134211540222,
      "epoch": 0.112,
      "grad_norm": 1.625,
      "learning_rate": 0.0004997038534857298,
      "loss": 3.225,
      "mean_token_accuracy": 0.24552348107099534,
      "num_tokens": 17203200.0,
      "step": 1050
    },
    {
      "entropy": 3.374810314178467,
      "epoch": 0.11253333333333333,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996403487745194,
      "loss": 3.2274,
      "mean_token_accuracy": 0.24568248689174652,
      "num_tokens": 17285120.0,
      "step": 1055
    },
    {
      "entropy": 3.3373070478439333,
      "epoch": 0.11306666666666666,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499570685434671,
      "loss": 3.224,
      "mean_token_accuracy": 0.24245352447032928,
      "num_tokens": 17367040.0,
      "step": 1060
    },
    {
      "entropy": 3.3667542695999146,
      "epoch": 0.1136,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004994948653760405,
      "loss": 3.2327,
      "mean_token_accuracy": 0.24299168288707734,
      "num_tokens": 17448960.0,
      "step": 1065
    },
    {
      "entropy": 3.3659775018692017,
      "epoch": 0.11413333333333334,
      "grad_norm": 2.0,
      "learning_rate": 0.0004994128906772729,
      "loss": 3.2345,
      "mean_token_accuracy": 0.2416829749941826,
      "num_tokens": 17530880.0,
      "step": 1070
    },
    {
      "entropy": 3.3636478424072265,
      "epoch": 0.11466666666666667,
      "grad_norm": 1.703125,
      "learning_rate": 0.000499324763585746,
      "loss": 3.2322,
      "mean_token_accuracy": 0.24304060637950897,
      "num_tokens": 17612800.0,
      "step": 1075
    },
    {
      "entropy": 3.347209930419922,
      "epoch": 0.1152,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004992304865175085,
      "loss": 3.2187,
      "mean_token_accuracy": 0.24414139091968537,
      "num_tokens": 17694720.0,
      "step": 1080
    },
    {
      "entropy": 3.3256459712982176,
      "epoch": 0.11573333333333333,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004991300620572138,
      "loss": 3.2122,
      "mean_token_accuracy": 0.2480919763445854,
      "num_tokens": 17776640.0,
      "step": 1085
    },
    {
      "entropy": 3.3572380781173705,
      "epoch": 0.11626666666666667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990234929580494,
      "loss": 3.2218,
      "mean_token_accuracy": 0.24530332982540132,
      "num_tokens": 17858560.0,
      "step": 1090
    },
    {
      "entropy": 3.3488968133926393,
      "epoch": 0.1168,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004989107821416609,
      "loss": 3.22,
      "mean_token_accuracy": 0.24153620153665542,
      "num_tokens": 17940480.0,
      "step": 1095
    },
    {
      "entropy": 3.3309558391571046,
      "epoch": 0.11733333333333333,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004987919326980723,
      "loss": 3.2169,
      "mean_token_accuracy": 0.24518101513385773,
      "num_tokens": 18022400.0,
      "step": 1100
    },
    {
      "entropy": 3.332407236099243,
      "epoch": 0.11786666666666666,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004986669478856011,
      "loss": 3.2205,
      "mean_token_accuracy": 0.24242906123399735,
      "num_tokens": 18104320.0,
      "step": 1105
    },
    {
      "entropy": 3.349472188949585,
      "epoch": 0.1184,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985358311307688,
      "loss": 3.2266,
      "mean_token_accuracy": 0.2414628192782402,
      "num_tokens": 18186240.0,
      "step": 1110
    },
    {
      "entropy": 3.331104302406311,
      "epoch": 0.11893333333333334,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004983985860282081,
      "loss": 3.2127,
      "mean_token_accuracy": 0.24697896242141723,
      "num_tokens": 18268160.0,
      "step": 1115
    },
    {
      "entropy": 3.319834756851196,
      "epoch": 0.11946666666666667,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982552163405623,
      "loss": 3.2088,
      "mean_token_accuracy": 0.24534002393484117,
      "num_tokens": 18350080.0,
      "step": 1120
    },
    {
      "entropy": 3.317161226272583,
      "epoch": 0.12,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981057259983839,
      "loss": 3.1988,
      "mean_token_accuracy": 0.2484955981373787,
      "num_tokens": 18432000.0,
      "step": 1125
    },
    {
      "entropy": 3.314718413352966,
      "epoch": 0.12053333333333334,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979501191000262,
      "loss": 3.2039,
      "mean_token_accuracy": 0.2452421709895134,
      "num_tokens": 18513920.0,
      "step": 1130
    },
    {
      "entropy": 3.3259057283401487,
      "epoch": 0.12106666666666667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004977883999115311,
      "loss": 3.209,
      "mean_token_accuracy": 0.24386007934808732,
      "num_tokens": 18595840.0,
      "step": 1135
    },
    {
      "entropy": 3.31879608631134,
      "epoch": 0.1216,
      "grad_norm": 1.25,
      "learning_rate": 0.0004976205728665113,
      "loss": 3.2092,
      "mean_token_accuracy": 0.2425880625844002,
      "num_tokens": 18677760.0,
      "step": 1140
    },
    {
      "entropy": 3.326778697967529,
      "epoch": 0.12213333333333333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974466425660307,
      "loss": 3.2106,
      "mean_token_accuracy": 0.24189089983701706,
      "num_tokens": 18759680.0,
      "step": 1145
    },
    {
      "entropy": 3.3221657514572143,
      "epoch": 0.12266666666666666,
      "grad_norm": 1.25,
      "learning_rate": 0.0004972666137784759,
      "loss": 3.2046,
      "mean_token_accuracy": 0.2446428582072258,
      "num_tokens": 18841600.0,
      "step": 1150
    },
    {
      "entropy": 3.311302924156189,
      "epoch": 0.1232,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004970804914394271,
      "loss": 3.2044,
      "mean_token_accuracy": 0.2443126231431961,
      "num_tokens": 18923520.0,
      "step": 1155
    },
    {
      "entropy": 3.3029479503631594,
      "epoch": 0.12373333333333333,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004968882806515225,
      "loss": 3.195,
      "mean_token_accuracy": 0.24726027250289917,
      "num_tokens": 19005440.0,
      "step": 1160
    },
    {
      "entropy": 3.306261348724365,
      "epoch": 0.12426666666666666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004966899866843177,
      "loss": 3.21,
      "mean_token_accuracy": 0.24412915855646133,
      "num_tokens": 19087360.0,
      "step": 1165
    },
    {
      "entropy": 3.3145808219909667,
      "epoch": 0.1248,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000496485614974142,
      "loss": 3.2126,
      "mean_token_accuracy": 0.24129158407449722,
      "num_tokens": 19169280.0,
      "step": 1170
    },
    {
      "entropy": 3.314248561859131,
      "epoch": 0.12533333333333332,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004962751711239492,
      "loss": 3.1896,
      "mean_token_accuracy": 0.24520548284053803,
      "num_tokens": 19251200.0,
      "step": 1175
    },
    {
      "entropy": 3.296343231201172,
      "epoch": 0.12586666666666665,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004960586609031636,
      "loss": 3.2084,
      "mean_token_accuracy": 0.24050880819559098,
      "num_tokens": 19333120.0,
      "step": 1180
    },
    {
      "entropy": 3.2957814216613768,
      "epoch": 0.1264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958360902475224,
      "loss": 3.1944,
      "mean_token_accuracy": 0.2450220137834549,
      "num_tokens": 19415040.0,
      "step": 1185
    },
    {
      "entropy": 3.2972240924835203,
      "epoch": 0.12693333333333334,
      "grad_norm": 1.75,
      "learning_rate": 0.0004956074652589125,
      "loss": 3.1947,
      "mean_token_accuracy": 0.2469055771827698,
      "num_tokens": 19496960.0,
      "step": 1190
    },
    {
      "entropy": 3.302608609199524,
      "epoch": 0.12746666666666667,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004953727922052035,
      "loss": 3.1965,
      "mean_token_accuracy": 0.24475293457508088,
      "num_tokens": 19578880.0,
      "step": 1195
    },
    {
      "entropy": 3.2857021570205687,
      "epoch": 0.128,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004951320775200756,
      "loss": 3.1931,
      "mean_token_accuracy": 0.2490949109196663,
      "num_tokens": 19660800.0,
      "step": 1200
    },
    {
      "entropy": 3.2887937784194947,
      "epoch": 0.12853333333333333,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004948853278028436,
      "loss": 3.1823,
      "mean_token_accuracy": 0.24741927534341812,
      "num_tokens": 19742720.0,
      "step": 1205
    },
    {
      "entropy": 3.2885008096694945,
      "epoch": 0.12906666666666666,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004946325498182755,
      "loss": 3.2001,
      "mean_token_accuracy": 0.24372553527355195,
      "num_tokens": 19824640.0,
      "step": 1210
    },
    {
      "entropy": 3.285250186920166,
      "epoch": 0.1296,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004943737504964076,
      "loss": 3.1976,
      "mean_token_accuracy": 0.24319960922002792,
      "num_tokens": 19906560.0,
      "step": 1215
    },
    {
      "entropy": 3.285680651664734,
      "epoch": 0.13013333333333332,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494108936932354,
      "loss": 3.1927,
      "mean_token_accuracy": 0.24404354393482208,
      "num_tokens": 19988480.0,
      "step": 1220
    },
    {
      "entropy": 3.2831942081451415,
      "epoch": 0.13066666666666665,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004938381163861124,
      "loss": 3.1895,
      "mean_token_accuracy": 0.24496086239814757,
      "num_tokens": 20070400.0,
      "step": 1225
    },
    {
      "entropy": 3.3090481758117676,
      "epoch": 0.1312,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004935612962823645,
      "loss": 3.199,
      "mean_token_accuracy": 0.2423801377415657,
      "num_tokens": 20152320.0,
      "step": 1230
    },
    {
      "entropy": 3.279662609100342,
      "epoch": 0.13173333333333334,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004932784842102739,
      "loss": 3.1845,
      "mean_token_accuracy": 0.24876467883586884,
      "num_tokens": 20234240.0,
      "step": 1235
    },
    {
      "entropy": 3.2735153436660767,
      "epoch": 0.13226666666666667,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004929896879232758,
      "loss": 3.1864,
      "mean_token_accuracy": 0.24636741429567338,
      "num_tokens": 20316160.0,
      "step": 1240
    },
    {
      "entropy": 3.2701792240142824,
      "epoch": 0.1328,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004926949153388668,
      "loss": 3.1827,
      "mean_token_accuracy": 0.24416584968566896,
      "num_tokens": 20398080.0,
      "step": 1245
    },
    {
      "entropy": 3.2787445068359373,
      "epoch": 0.13333333333333333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004923941745383859,
      "loss": 3.177,
      "mean_token_accuracy": 0.24897260069847107,
      "num_tokens": 20480000.0,
      "step": 1250
    },
    {
      "entropy": 3.2674917697906496,
      "epoch": 0.13386666666666666,
      "grad_norm": 1.125,
      "learning_rate": 0.000492087473766794,
      "loss": 3.174,
      "mean_token_accuracy": 0.2497064560651779,
      "num_tokens": 20561920.0,
      "step": 1255
    },
    {
      "entropy": 3.2698441982269286,
      "epoch": 0.1344,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000491774821432448,
      "loss": 3.1754,
      "mean_token_accuracy": 0.24757827669382096,
      "num_tokens": 20643840.0,
      "step": 1260
    },
    {
      "entropy": 3.2541027069091797,
      "epoch": 0.13493333333333332,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004914562261068693,
      "loss": 3.1745,
      "mean_token_accuracy": 0.24631849080324172,
      "num_tokens": 20725760.0,
      "step": 1265
    },
    {
      "entropy": 3.2766933679580688,
      "epoch": 0.13546666666666668,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004911316965245098,
      "loss": 3.1727,
      "mean_token_accuracy": 0.24894814044237137,
      "num_tokens": 20807680.0,
      "step": 1270
    },
    {
      "entropy": 3.2813777208328245,
      "epoch": 0.136,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000490801241582512,
      "loss": 3.1915,
      "mean_token_accuracy": 0.2451198622584343,
      "num_tokens": 20889600.0,
      "step": 1275
    },
    {
      "entropy": 3.2646867990493775,
      "epoch": 0.13653333333333334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490464870340465,
      "loss": 3.1757,
      "mean_token_accuracy": 0.24669765084981918,
      "num_tokens": 20971520.0,
      "step": 1280
    },
    {
      "entropy": 3.2562508583068848,
      "epoch": 0.13706666666666667,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004901225920201563,
      "loss": 3.1728,
      "mean_token_accuracy": 0.24716242402791977,
      "num_tokens": 21053440.0,
      "step": 1285
    },
    {
      "entropy": 3.253575849533081,
      "epoch": 0.1376,
      "grad_norm": 1.765625,
      "learning_rate": 0.000489774416005319,
      "loss": 3.1774,
      "mean_token_accuracy": 0.2478595867753029,
      "num_tokens": 21135360.0,
      "step": 1290
    },
    {
      "entropy": 3.2708191871643066,
      "epoch": 0.13813333333333333,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004894203518413742,
      "loss": 3.1756,
      "mean_token_accuracy": 0.2494618386030197,
      "num_tokens": 21217280.0,
      "step": 1295
    },
    {
      "entropy": 3.259015607833862,
      "epoch": 0.13866666666666666,
      "grad_norm": 1.0,
      "learning_rate": 0.0004890604092351701,
      "loss": 3.1755,
      "mean_token_accuracy": 0.24667318910360336,
      "num_tokens": 21299200.0,
      "step": 1300
    },
    {
      "entropy": 3.251534938812256,
      "epoch": 0.1392,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000488694598054715,
      "loss": 3.172,
      "mean_token_accuracy": 0.24642856866121293,
      "num_tokens": 21381120.0,
      "step": 1305
    },
    {
      "entropy": 3.2675605535507204,
      "epoch": 0.13973333333333332,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004883229283289071,
      "loss": 3.182,
      "mean_token_accuracy": 0.24386007934808732,
      "num_tokens": 21463040.0,
      "step": 1310
    },
    {
      "entropy": 3.25923867225647,
      "epoch": 0.14026666666666668,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048794541024725993,
      "loss": 3.1757,
      "mean_token_accuracy": 0.2474315032362938,
      "num_tokens": 21544960.0,
      "step": 1315
    },
    {
      "entropy": 3.250070834159851,
      "epoch": 0.1408,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004875620541596221,
      "loss": 3.1563,
      "mean_token_accuracy": 0.25125978142023087,
      "num_tokens": 21626880.0,
      "step": 1320
    },
    {
      "entropy": 3.247203159332275,
      "epoch": 0.14133333333333334,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048717287057589454,
      "loss": 3.1715,
      "mean_token_accuracy": 0.24865460246801377,
      "num_tokens": 21708800.0,
      "step": 1325
    },
    {
      "entropy": 3.259696698188782,
      "epoch": 0.14186666666666667,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004867778701657417,
      "loss": 3.1756,
      "mean_token_accuracy": 0.24743150621652604,
      "num_tokens": 21790720.0,
      "step": 1330
    },
    {
      "entropy": 3.2434436321258544,
      "epoch": 0.1424,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048637706375829955,
      "loss": 3.1706,
      "mean_token_accuracy": 0.24798190146684645,
      "num_tokens": 21872640.0,
      "step": 1335
    },
    {
      "entropy": 3.2555607080459597,
      "epoch": 0.14293333333333333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000485970462341878,
      "loss": 3.1817,
      "mean_token_accuracy": 0.24480185955762862,
      "num_tokens": 21954560.0,
      "step": 1340
    },
    {
      "entropy": 3.2555992364883424,
      "epoch": 0.14346666666666666,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048555807706366044,
      "loss": 3.1661,
      "mean_token_accuracy": 0.24701565653085708,
      "num_tokens": 22036480.0,
      "step": 1345
    },
    {
      "entropy": 3.2635560274124145,
      "epoch": 0.144,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048513991922939756,
      "loss": 3.1719,
      "mean_token_accuracy": 0.2476638972759247,
      "num_tokens": 22118400.0,
      "step": 1350
    },
    {
      "entropy": 3.2353002548217775,
      "epoch": 0.14453333333333335,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048471600030309744,
      "loss": 3.1687,
      "mean_token_accuracy": 0.24576810002326965,
      "num_tokens": 22200320.0,
      "step": 1355
    },
    {
      "entropy": 3.2604353427886963,
      "epoch": 0.14506666666666668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048428633190671186,
      "loss": 3.1833,
      "mean_token_accuracy": 0.24412915706634522,
      "num_tokens": 22282240.0,
      "step": 1360
    },
    {
      "entropy": 3.250130367279053,
      "epoch": 0.1456,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004838509258198167,
      "loss": 3.1613,
      "mean_token_accuracy": 0.24606164395809174,
      "num_tokens": 22364160.0,
      "step": 1365
    },
    {
      "entropy": 3.239962911605835,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048340979397929,
      "loss": 3.1646,
      "mean_token_accuracy": 0.24450831711292267,
      "num_tokens": 22446080.0,
      "step": 1370
    },
    {
      "entropy": 3.2506635189056396,
      "epoch": 0.14666666666666667,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048296294847898386,
      "loss": 3.1618,
      "mean_token_accuracy": 0.2475171208381653,
      "num_tokens": 22528000.0,
      "step": 1375
    },
    {
      "entropy": 3.238317131996155,
      "epoch": 0.1472,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004825104015693934,
      "loss": 3.1581,
      "mean_token_accuracy": 0.2483855202794075,
      "num_tokens": 22609920.0,
      "step": 1380
    },
    {
      "entropy": 3.2297689437866213,
      "epoch": 0.14773333333333333,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004820521656573208,
      "loss": 3.1523,
      "mean_token_accuracy": 0.24914383441209792,
      "num_tokens": 22691840.0,
      "step": 1385
    },
    {
      "entropy": 3.226832556724548,
      "epoch": 0.14826666666666666,
      "grad_norm": 1.0,
      "learning_rate": 0.00048158825330553505,
      "loss": 3.161,
      "mean_token_accuracy": 0.2477372795343399,
      "num_tokens": 22773760.0,
      "step": 1390
    },
    {
      "entropy": 3.241289162635803,
      "epoch": 0.1488,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048111867723242763,
      "loss": 3.1622,
      "mean_token_accuracy": 0.24646526277065278,
      "num_tokens": 22855680.0,
      "step": 1395
    },
    {
      "entropy": 3.230481195449829,
      "epoch": 0.14933333333333335,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004806434503116637,
      "loss": 3.1512,
      "mean_token_accuracy": 0.2504525452852249,
      "num_tokens": 22937600.0,
      "step": 1400
    },
    {
      "entropy": 3.239303541183472,
      "epoch": 0.14986666666666668,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004801625855718296,
      "loss": 3.1624,
      "mean_token_accuracy": 0.2447773963212967,
      "num_tokens": 23019520.0,
      "step": 1405
    },
    {
      "entropy": 3.2366477489471435,
      "epoch": 0.1504,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047967609619607477,
      "loss": 3.164,
      "mean_token_accuracy": 0.24387230724096298,
      "num_tokens": 23101440.0,
      "step": 1410
    },
    {
      "entropy": 3.23808913230896,
      "epoch": 0.15093333333333334,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004791839955217513,
      "loss": 3.1671,
      "mean_token_accuracy": 0.24540117383003235,
      "num_tokens": 23183360.0,
      "step": 1415
    },
    {
      "entropy": 3.2244601249694824,
      "epoch": 0.15146666666666667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047868629704004786,
      "loss": 3.1429,
      "mean_token_accuracy": 0.2505626201629639,
      "num_tokens": 23265280.0,
      "step": 1420
    },
    {
      "entropy": 3.2507676601409914,
      "epoch": 0.152,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047818301439561965,
      "loss": 3.169,
      "mean_token_accuracy": 0.2452421724796295,
      "num_tokens": 23347200.0,
      "step": 1425
    },
    {
      "entropy": 3.2121356248855593,
      "epoch": 0.15253333333333333,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047767416138621454,
      "loss": 3.1471,
      "mean_token_accuracy": 0.25017123073339465,
      "num_tokens": 23429120.0,
      "step": 1430
    },
    {
      "entropy": 3.218046474456787,
      "epoch": 0.15306666666666666,
      "grad_norm": 0.953125,
      "learning_rate": 0.000477159751962295,
      "loss": 3.1436,
      "mean_token_accuracy": 0.25108855217695236,
      "num_tokens": 23511040.0,
      "step": 1435
    },
    {
      "entropy": 3.2176970481872558,
      "epoch": 0.1536,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047663980022665507,
      "loss": 3.1525,
      "mean_token_accuracy": 0.2482999011874199,
      "num_tokens": 23592960.0,
      "step": 1440
    },
    {
      "entropy": 3.2244531154632567,
      "epoch": 0.15413333333333334,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047611432043403437,
      "loss": 3.1472,
      "mean_token_accuracy": 0.24842221140861512,
      "num_tokens": 23674880.0,
      "step": 1445
    },
    {
      "entropy": 3.2204693078994753,
      "epoch": 0.15466666666666667,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004755833269907267,
      "loss": 3.1497,
      "mean_token_accuracy": 0.24949852973222733,
      "num_tokens": 23756800.0,
      "step": 1450
    },
    {
      "entropy": 3.2204687356948853,
      "epoch": 0.1552,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004750468344541857,
      "loss": 3.1528,
      "mean_token_accuracy": 0.24694227129220964,
      "num_tokens": 23838720.0,
      "step": 1455
    },
    {
      "entropy": 3.2242907762527464,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.859375,
      "learning_rate": 0.00047450485753262525,
      "loss": 3.1501,
      "mean_token_accuracy": 0.24770058542490006,
      "num_tokens": 23920640.0,
      "step": 1460
    },
    {
      "entropy": 3.208418536186218,
      "epoch": 0.15626666666666666,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047395741108461633,
      "loss": 3.1442,
      "mean_token_accuracy": 0.24839774817228316,
      "num_tokens": 24002560.0,
      "step": 1465
    },
    {
      "entropy": 3.2096371412277223,
      "epoch": 0.1568,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047340451011867985,
      "loss": 3.1389,
      "mean_token_accuracy": 0.2509784743189812,
      "num_tokens": 24084480.0,
      "step": 1470
    },
    {
      "entropy": 3.207978129386902,
      "epoch": 0.15733333333333333,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047284616979287515,
      "loss": 3.1386,
      "mean_token_accuracy": 0.2520303353667259,
      "num_tokens": 24166400.0,
      "step": 1475
    },
    {
      "entropy": 3.2250426769256593,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.765625,
      "learning_rate": 0.00047228240541438433,
      "loss": 3.1525,
      "mean_token_accuracy": 0.24716242402791977,
      "num_tokens": 24248320.0,
      "step": 1480
    },
    {
      "entropy": 3.216737985610962,
      "epoch": 0.1584,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00047171323243909257,
      "loss": 3.1438,
      "mean_token_accuracy": 0.24896036833524704,
      "num_tokens": 24330240.0,
      "step": 1485
    },
    {
      "entropy": 3.216948318481445,
      "epoch": 0.15893333333333334,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047113866647116457,
      "loss": 3.1477,
      "mean_token_accuracy": 0.24820205420255662,
      "num_tokens": 24412160.0,
      "step": 1490
    },
    {
      "entropy": 3.215489459037781,
      "epoch": 0.15946666666666667,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004705587232626164,
      "loss": 3.1494,
      "mean_token_accuracy": 0.24653865098953248,
      "num_tokens": 24494080.0,
      "step": 1495
    },
    {
      "entropy": 3.216701889038086,
      "epoch": 0.16,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00046997341871288424,
      "loss": 3.1436,
      "mean_token_accuracy": 0.24893590807914734,
      "num_tokens": 24576000.0,
      "step": 1500
    }
  ],
  "logging_steps": 5,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2542017536e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}