{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9592476489028212,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.692053604125977,
      "epoch": 0.003918495297805642,
      "grad_norm": 13.875,
      "learning_rate": 2e-06,
      "loss": 10.821,
      "mean_token_accuracy": 0.00013404826167970896,
      "num_tokens": 9174.0,
      "step": 5
    },
    {
      "entropy": 10.691976261138915,
      "epoch": 0.007836990595611285,
      "grad_norm": 13.25,
      "learning_rate": 4.5e-06,
      "loss": 10.7352,
      "mean_token_accuracy": 0.0,
      "num_tokens": 17790.0,
      "step": 10
    },
    {
      "entropy": 10.6909686088562,
      "epoch": 0.011755485893416929,
      "grad_norm": 9.5,
      "learning_rate": 7e-06,
      "loss": 10.4692,
      "mean_token_accuracy": 0.024429761280771345,
      "num_tokens": 25850.0,
      "step": 15
    },
    {
      "entropy": 10.677701091766357,
      "epoch": 0.01567398119122257,
      "grad_norm": 6.59375,
      "learning_rate": 9.5e-06,
      "loss": 10.2158,
      "mean_token_accuracy": 0.04213750921189785,
      "num_tokens": 35194.0,
      "step": 20
    },
    {
      "entropy": 10.598451328277587,
      "epoch": 0.019592476489028215,
      "grad_norm": 4.375,
      "learning_rate": 1.2e-05,
      "loss": 9.9693,
      "mean_token_accuracy": 0.042495733872056005,
      "num_tokens": 44218.0,
      "step": 25
    },
    {
      "entropy": 10.572938823699952,
      "epoch": 0.023510971786833857,
      "grad_norm": 3.421875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8009,
      "mean_token_accuracy": 0.04398789629340172,
      "num_tokens": 53397.0,
      "step": 30
    },
    {
      "entropy": 10.539127254486084,
      "epoch": 0.0274294670846395,
      "grad_norm": 2.921875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6844,
      "mean_token_accuracy": 0.04925155565142632,
      "num_tokens": 62749.0,
      "step": 35
    },
    {
      "entropy": 10.50910243988037,
      "epoch": 0.03134796238244514,
      "grad_norm": 2.59375,
      "learning_rate": 1.95e-05,
      "loss": 9.6464,
      "mean_token_accuracy": 0.04322206266224384,
      "num_tokens": 71721.0,
      "step": 40
    },
    {
      "entropy": 10.538900089263915,
      "epoch": 0.03526645768025078,
      "grad_norm": 2.765625,
      "learning_rate": 2.2e-05,
      "loss": 9.5818,
      "mean_token_accuracy": 0.04916957952082157,
      "num_tokens": 79844.0,
      "step": 45
    },
    {
      "entropy": 10.527848815917968,
      "epoch": 0.03918495297805643,
      "grad_norm": 2.421875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.567,
      "mean_token_accuracy": 0.05607832483947277,
      "num_tokens": 88866.0,
      "step": 50
    },
    {
      "entropy": 10.449817848205566,
      "epoch": 0.04310344827586207,
      "grad_norm": 2.265625,
      "learning_rate": 2.7e-05,
      "loss": 9.4999,
      "mean_token_accuracy": 0.06071024388074875,
      "num_tokens": 97918.0,
      "step": 55
    },
    {
      "entropy": 10.289769554138184,
      "epoch": 0.047021943573667714,
      "grad_norm": 2.203125,
      "learning_rate": 2.95e-05,
      "loss": 9.42,
      "mean_token_accuracy": 0.06239944435656071,
      "num_tokens": 107043.0,
      "step": 60
    },
    {
      "entropy": 10.339974212646485,
      "epoch": 0.050940438871473356,
      "grad_norm": 2.328125,
      "learning_rate": 3.2e-05,
      "loss": 9.3525,
      "mean_token_accuracy": 0.06377544328570366,
      "num_tokens": 116000.0,
      "step": 65
    },
    {
      "entropy": 10.358237171173096,
      "epoch": 0.054858934169279,
      "grad_norm": 2.15625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2722,
      "mean_token_accuracy": 0.06179721914231777,
      "num_tokens": 124559.0,
      "step": 70
    },
    {
      "entropy": 10.306263256072999,
      "epoch": 0.05877742946708464,
      "grad_norm": 2.015625,
      "learning_rate": 3.7e-05,
      "loss": 9.1558,
      "mean_token_accuracy": 0.0636992421001196,
      "num_tokens": 132868.0,
      "step": 75
    },
    {
      "entropy": 10.310817909240722,
      "epoch": 0.06269592476489028,
      "grad_norm": 2.09375,
      "learning_rate": 3.95e-05,
      "loss": 9.1437,
      "mean_token_accuracy": 0.06109033823013306,
      "num_tokens": 141286.0,
      "step": 80
    },
    {
      "entropy": 10.31660270690918,
      "epoch": 0.06661442006269593,
      "grad_norm": 1.8828125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0017,
      "mean_token_accuracy": 0.06191765181720257,
      "num_tokens": 150406.0,
      "step": 85
    },
    {
      "entropy": 10.19018259048462,
      "epoch": 0.07053291536050156,
      "grad_norm": 1.9140625,
      "learning_rate": 4.45e-05,
      "loss": 8.9406,
      "mean_token_accuracy": 0.06189325153827667,
      "num_tokens": 158770.0,
      "step": 90
    },
    {
      "entropy": 10.224696159362793,
      "epoch": 0.07445141065830721,
      "grad_norm": 1.8515625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8329,
      "mean_token_accuracy": 0.06557972468435765,
      "num_tokens": 167763.0,
      "step": 95
    },
    {
      "entropy": 10.152040767669678,
      "epoch": 0.07836990595611286,
      "grad_norm": 1.5859375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.6853,
      "mean_token_accuracy": 0.06438785269856453,
      "num_tokens": 177306.0,
      "step": 100
    },
    {
      "entropy": 10.053633880615234,
      "epoch": 0.0822884012539185,
      "grad_norm": 1.59375,
      "learning_rate": 5.2e-05,
      "loss": 8.5993,
      "mean_token_accuracy": 0.06454790569841862,
      "num_tokens": 186014.0,
      "step": 105
    },
    {
      "entropy": 9.987529850006103,
      "epoch": 0.08620689655172414,
      "grad_norm": 1.625,
      "learning_rate": 5.45e-05,
      "loss": 8.4292,
      "mean_token_accuracy": 0.06873111054301262,
      "num_tokens": 194122.0,
      "step": 110
    },
    {
      "entropy": 9.858163261413575,
      "epoch": 0.09012539184952978,
      "grad_norm": 1.703125,
      "learning_rate": 5.7e-05,
      "loss": 8.3325,
      "mean_token_accuracy": 0.06681315004825591,
      "num_tokens": 203097.0,
      "step": 115
    },
    {
      "entropy": 9.74828462600708,
      "epoch": 0.09404388714733543,
      "grad_norm": 1.6015625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2069,
      "mean_token_accuracy": 0.06950863264501095,
      "num_tokens": 211413.0,
      "step": 120
    },
    {
      "entropy": 9.593859004974366,
      "epoch": 0.09796238244514106,
      "grad_norm": 1.3828125,
      "learning_rate": 6.2e-05,
      "loss": 8.1517,
      "mean_token_accuracy": 0.06460827700793743,
      "num_tokens": 220550.0,
      "step": 125
    },
    {
      "entropy": 9.457711124420166,
      "epoch": 0.10188087774294671,
      "grad_norm": 1.3828125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9844,
      "mean_token_accuracy": 0.072696752846241,
      "num_tokens": 229197.0,
      "step": 130
    },
    {
      "entropy": 9.283110237121582,
      "epoch": 0.10579937304075235,
      "grad_norm": 1.359375,
      "learning_rate": 6.7e-05,
      "loss": 7.9816,
      "mean_token_accuracy": 0.06841139532625676,
      "num_tokens": 238479.0,
      "step": 135
    },
    {
      "entropy": 9.081107711791992,
      "epoch": 0.109717868338558,
      "grad_norm": 1.515625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8324,
      "mean_token_accuracy": 0.07079579271376132,
      "num_tokens": 246318.0,
      "step": 140
    },
    {
      "entropy": 8.931202793121338,
      "epoch": 0.11363636363636363,
      "grad_norm": 1.3984375,
      "learning_rate": 7.2e-05,
      "loss": 7.8567,
      "mean_token_accuracy": 0.06792247369885444,
      "num_tokens": 254783.0,
      "step": 145
    },
    {
      "entropy": 8.767581272125245,
      "epoch": 0.11755485893416928,
      "grad_norm": 1.375,
      "learning_rate": 7.45e-05,
      "loss": 7.6935,
      "mean_token_accuracy": 0.0719048798084259,
      "num_tokens": 263416.0,
      "step": 150
    },
    {
      "entropy": 8.662985897064209,
      "epoch": 0.12147335423197492,
      "grad_norm": 1.671875,
      "learning_rate": 7.7e-05,
      "loss": 7.712,
      "mean_token_accuracy": 0.07060935273766518,
      "num_tokens": 271930.0,
      "step": 155
    },
    {
      "entropy": 8.506498527526855,
      "epoch": 0.12539184952978055,
      "grad_norm": 1.3671875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.619,
      "mean_token_accuracy": 0.07228946462273597,
      "num_tokens": 280546.0,
      "step": 160
    },
    {
      "entropy": 8.340063381195069,
      "epoch": 0.12931034482758622,
      "grad_norm": 1.359375,
      "learning_rate": 8.2e-05,
      "loss": 7.6031,
      "mean_token_accuracy": 0.07542856186628341,
      "num_tokens": 288813.0,
      "step": 165
    },
    {
      "entropy": 8.317859268188476,
      "epoch": 0.13322884012539185,
      "grad_norm": 1.2265625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.5757,
      "mean_token_accuracy": 0.07388538345694542,
      "num_tokens": 297966.0,
      "step": 170
    },
    {
      "entropy": 8.287030601501465,
      "epoch": 0.1371473354231975,
      "grad_norm": 1.2421875,
      "learning_rate": 8.7e-05,
      "loss": 7.6313,
      "mean_token_accuracy": 0.07094714678823948,
      "num_tokens": 307135.0,
      "step": 175
    },
    {
      "entropy": 8.15151834487915,
      "epoch": 0.14106583072100312,
      "grad_norm": 1.1875,
      "learning_rate": 8.95e-05,
      "loss": 7.543,
      "mean_token_accuracy": 0.07589463070034981,
      "num_tokens": 315546.0,
      "step": 180
    },
    {
      "entropy": 8.161028385162354,
      "epoch": 0.14498432601880878,
      "grad_norm": 1.2265625,
      "learning_rate": 9.2e-05,
      "loss": 7.5249,
      "mean_token_accuracy": 0.0759074218571186,
      "num_tokens": 323930.0,
      "step": 185
    },
    {
      "entropy": 8.058754682540894,
      "epoch": 0.14890282131661442,
      "grad_norm": 1.2578125,
      "learning_rate": 9.45e-05,
      "loss": 7.392,
      "mean_token_accuracy": 0.08224981352686882,
      "num_tokens": 332419.0,
      "step": 190
    },
    {
      "entropy": 8.08242244720459,
      "epoch": 0.15282131661442006,
      "grad_norm": 1.265625,
      "learning_rate": 9.7e-05,
      "loss": 7.4781,
      "mean_token_accuracy": 0.07802417948842048,
      "num_tokens": 341362.0,
      "step": 195
    },
    {
      "entropy": 7.985074710845947,
      "epoch": 0.15673981191222572,
      "grad_norm": 1.234375,
      "learning_rate": 9.95e-05,
      "loss": 7.3972,
      "mean_token_accuracy": 0.07822780273854732,
      "num_tokens": 349395.0,
      "step": 200
    },
    {
      "entropy": 8.01458387374878,
      "epoch": 0.16065830721003135,
      "grad_norm": 1.546875,
      "learning_rate": 0.000102,
      "loss": 7.4184,
      "mean_token_accuracy": 0.07329343855381013,
      "num_tokens": 358413.0,
      "step": 205
    },
    {
      "entropy": 7.919571828842163,
      "epoch": 0.164576802507837,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.5032,
      "mean_token_accuracy": 0.07889928370714187,
      "num_tokens": 366489.0,
      "step": 210
    },
    {
      "entropy": 7.988800716400147,
      "epoch": 0.16849529780564262,
      "grad_norm": 1.265625,
      "learning_rate": 0.000107,
      "loss": 7.3554,
      "mean_token_accuracy": 0.07564577832818031,
      "num_tokens": 375335.0,
      "step": 215
    },
    {
      "entropy": 7.991494131088257,
      "epoch": 0.1724137931034483,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001095,
      "loss": 7.407,
      "mean_token_accuracy": 0.08004131242632866,
      "num_tokens": 384276.0,
      "step": 220
    },
    {
      "entropy": 7.928954362869263,
      "epoch": 0.17633228840125392,
      "grad_norm": 1.21875,
      "learning_rate": 0.000112,
      "loss": 7.447,
      "mean_token_accuracy": 0.07608951888978481,
      "num_tokens": 393571.0,
      "step": 225
    },
    {
      "entropy": 7.913915538787842,
      "epoch": 0.18025078369905956,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001145,
      "loss": 7.3764,
      "mean_token_accuracy": 0.08197141736745835,
      "num_tokens": 401865.0,
      "step": 230
    },
    {
      "entropy": 7.777705907821655,
      "epoch": 0.1841692789968652,
      "grad_norm": 1.15625,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3381,
      "mean_token_accuracy": 0.08282902240753173,
      "num_tokens": 410518.0,
      "step": 235
    },
    {
      "entropy": 7.857710313796997,
      "epoch": 0.18808777429467086,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.356,
      "mean_token_accuracy": 0.08076885864138603,
      "num_tokens": 419769.0,
      "step": 240
    },
    {
      "entropy": 7.7723448276519775,
      "epoch": 0.1920062695924765,
      "grad_norm": 1.265625,
      "learning_rate": 0.000122,
      "loss": 7.2799,
      "mean_token_accuracy": 0.08509492203593254,
      "num_tokens": 428204.0,
      "step": 245
    },
    {
      "entropy": 7.8825091361999515,
      "epoch": 0.19592476489028213,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001245,
      "loss": 7.3833,
      "mean_token_accuracy": 0.07587559074163437,
      "num_tokens": 436594.0,
      "step": 250
    },
    {
      "entropy": 7.804588508605957,
      "epoch": 0.19984326018808776,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000127,
      "loss": 7.1619,
      "mean_token_accuracy": 0.08455468490719795,
      "num_tokens": 444645.0,
      "step": 255
    },
    {
      "entropy": 7.734504652023316,
      "epoch": 0.20376175548589343,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001295,
      "loss": 7.2047,
      "mean_token_accuracy": 0.0890711173415184,
      "num_tokens": 453016.0,
      "step": 260
    },
    {
      "entropy": 7.689095687866211,
      "epoch": 0.20768025078369906,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000132,
      "loss": 7.2874,
      "mean_token_accuracy": 0.08078921884298325,
      "num_tokens": 462116.0,
      "step": 265
    },
    {
      "entropy": 7.745668983459472,
      "epoch": 0.2115987460815047,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.2359,
      "mean_token_accuracy": 0.08409795872867107,
      "num_tokens": 470807.0,
      "step": 270
    },
    {
      "entropy": 7.6965230941772464,
      "epoch": 0.21551724137931033,
      "grad_norm": 1.546875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.3247,
      "mean_token_accuracy": 0.08189421519637108,
      "num_tokens": 479592.0,
      "step": 275
    },
    {
      "entropy": 7.784852504730225,
      "epoch": 0.219435736677116,
      "grad_norm": 1.25,
      "learning_rate": 0.0001395,
      "loss": 7.2423,
      "mean_token_accuracy": 0.08114428222179412,
      "num_tokens": 488107.0,
      "step": 280
    },
    {
      "entropy": 7.663507890701294,
      "epoch": 0.22335423197492163,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.2547,
      "mean_token_accuracy": 0.08439974337816239,
      "num_tokens": 496775.0,
      "step": 285
    },
    {
      "entropy": 7.740933275222778,
      "epoch": 0.22727272727272727,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001445,
      "loss": 7.2816,
      "mean_token_accuracy": 0.08367991298437119,
      "num_tokens": 505415.0,
      "step": 290
    },
    {
      "entropy": 7.680744886398315,
      "epoch": 0.23119122257053293,
      "grad_norm": 1.28125,
      "learning_rate": 0.000147,
      "loss": 7.2656,
      "mean_token_accuracy": 0.08193187937140464,
      "num_tokens": 514447.0,
      "step": 295
    },
    {
      "entropy": 7.735220909118652,
      "epoch": 0.23510971786833856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001495,
      "loss": 7.2519,
      "mean_token_accuracy": 0.086246570199728,
      "num_tokens": 522998.0,
      "step": 300
    },
    {
      "entropy": 7.693396520614624,
      "epoch": 0.2390282131661442,
      "grad_norm": 1.34375,
      "learning_rate": 0.000152,
      "loss": 7.2173,
      "mean_token_accuracy": 0.08238770589232444,
      "num_tokens": 531542.0,
      "step": 305
    },
    {
      "entropy": 7.604292917251587,
      "epoch": 0.24294670846394983,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.1457,
      "mean_token_accuracy": 0.08177541941404343,
      "num_tokens": 540143.0,
      "step": 310
    },
    {
      "entropy": 7.716537094116211,
      "epoch": 0.2468652037617555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000157,
      "loss": 7.2444,
      "mean_token_accuracy": 0.08475765213370323,
      "num_tokens": 549156.0,
      "step": 315
    },
    {
      "entropy": 7.560835886001587,
      "epoch": 0.2507836990595611,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001595,
      "loss": 7.0897,
      "mean_token_accuracy": 0.08665165156126023,
      "num_tokens": 557522.0,
      "step": 320
    },
    {
      "entropy": 7.5509428024292,
      "epoch": 0.2547021943573668,
      "grad_norm": 1.3125,
      "learning_rate": 0.000162,
      "loss": 7.1604,
      "mean_token_accuracy": 0.0852648138999939,
      "num_tokens": 566650.0,
      "step": 325
    },
    {
      "entropy": 7.643481254577637,
      "epoch": 0.25862068965517243,
      "grad_norm": 1.375,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.1157,
      "mean_token_accuracy": 0.08511288538575172,
      "num_tokens": 576219.0,
      "step": 330
    },
    {
      "entropy": 7.535575675964355,
      "epoch": 0.26253918495297807,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.2172,
      "mean_token_accuracy": 0.08702815398573875,
      "num_tokens": 584304.0,
      "step": 335
    },
    {
      "entropy": 7.477091073989868,
      "epoch": 0.2664576802507837,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.1379,
      "mean_token_accuracy": 0.08963485434651375,
      "num_tokens": 593163.0,
      "step": 340
    },
    {
      "entropy": 7.577557039260864,
      "epoch": 0.27037617554858934,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0509,
      "mean_token_accuracy": 0.08813998103141785,
      "num_tokens": 602077.0,
      "step": 345
    },
    {
      "entropy": 7.504067659378052,
      "epoch": 0.274294670846395,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1645,
      "mean_token_accuracy": 0.08567056804895401,
      "num_tokens": 610009.0,
      "step": 350
    },
    {
      "entropy": 7.615056896209717,
      "epoch": 0.2782131661442006,
      "grad_norm": 1.171875,
      "learning_rate": 0.000177,
      "loss": 7.1942,
      "mean_token_accuracy": 0.08498035669326783,
      "num_tokens": 619282.0,
      "step": 355
    },
    {
      "entropy": 7.465681743621826,
      "epoch": 0.28213166144200624,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001795,
      "loss": 7.1301,
      "mean_token_accuracy": 0.08965396955609321,
      "num_tokens": 628138.0,
      "step": 360
    },
    {
      "entropy": 7.515298175811767,
      "epoch": 0.28605015673981193,
      "grad_norm": 1.375,
      "learning_rate": 0.000182,
      "loss": 6.9813,
      "mean_token_accuracy": 0.0954482302069664,
      "num_tokens": 637021.0,
      "step": 365
    },
    {
      "entropy": 7.504141473770142,
      "epoch": 0.28996865203761757,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001845,
      "loss": 7.1534,
      "mean_token_accuracy": 0.0818020910024643,
      "num_tokens": 646703.0,
      "step": 370
    },
    {
      "entropy": 7.471045875549317,
      "epoch": 0.2938871473354232,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000187,
      "loss": 7.0346,
      "mean_token_accuracy": 0.09074857905507087,
      "num_tokens": 655616.0,
      "step": 375
    },
    {
      "entropy": 7.431417989730835,
      "epoch": 0.29780564263322884,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001895,
      "loss": 7.0272,
      "mean_token_accuracy": 0.09865856096148491,
      "num_tokens": 663783.0,
      "step": 380
    },
    {
      "entropy": 7.422271394729615,
      "epoch": 0.3017241379310345,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000192,
      "loss": 7.0382,
      "mean_token_accuracy": 0.09664822444319725,
      "num_tokens": 671855.0,
      "step": 385
    },
    {
      "entropy": 7.441711044311523,
      "epoch": 0.3056426332288401,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001945,
      "loss": 7.0574,
      "mean_token_accuracy": 0.0905517354607582,
      "num_tokens": 680981.0,
      "step": 390
    },
    {
      "entropy": 7.409027004241944,
      "epoch": 0.30956112852664575,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.9713,
      "mean_token_accuracy": 0.09158209860324859,
      "num_tokens": 689294.0,
      "step": 395
    },
    {
      "entropy": 7.345873880386352,
      "epoch": 0.31347962382445144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.936,
      "mean_token_accuracy": 0.09149299785494805,
      "num_tokens": 697703.0,
      "step": 400
    },
    {
      "entropy": 7.485394239425659,
      "epoch": 0.31739811912225707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000202,
      "loss": 7.0726,
      "mean_token_accuracy": 0.09204713851213456,
      "num_tokens": 706792.0,
      "step": 405
    },
    {
      "entropy": 7.409369468688965,
      "epoch": 0.3213166144200627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9865,
      "mean_token_accuracy": 0.09500117674469948,
      "num_tokens": 715864.0,
      "step": 410
    },
    {
      "entropy": 7.2893208980560305,
      "epoch": 0.32523510971786834,
      "grad_norm": 1.40625,
      "learning_rate": 0.000207,
      "loss": 7.0287,
      "mean_token_accuracy": 0.0940815657377243,
      "num_tokens": 723921.0,
      "step": 415
    },
    {
      "entropy": 7.437120723724365,
      "epoch": 0.329153605015674,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002095,
      "loss": 7.0524,
      "mean_token_accuracy": 0.08924909234046936,
      "num_tokens": 732797.0,
      "step": 420
    },
    {
      "entropy": 7.3198113441467285,
      "epoch": 0.3330721003134796,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000212,
      "loss": 6.8983,
      "mean_token_accuracy": 0.10327747613191604,
      "num_tokens": 741248.0,
      "step": 425
    },
    {
      "entropy": 7.3485208511352536,
      "epoch": 0.33699059561128525,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002145,
      "loss": 7.0034,
      "mean_token_accuracy": 0.0921932153403759,
      "num_tokens": 749766.0,
      "step": 430
    },
    {
      "entropy": 7.331899881362915,
      "epoch": 0.3409090909090909,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9561,
      "mean_token_accuracy": 0.09300489202141762,
      "num_tokens": 758695.0,
      "step": 435
    },
    {
      "entropy": 7.339229249954224,
      "epoch": 0.3448275862068966,
      "grad_norm": 1.375,
      "learning_rate": 0.0002195,
      "loss": 6.9566,
      "mean_token_accuracy": 0.09157029166817665,
      "num_tokens": 767624.0,
      "step": 440
    },
    {
      "entropy": 7.343198585510254,
      "epoch": 0.3487460815047022,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000222,
      "loss": 6.9819,
      "mean_token_accuracy": 0.09128969609737396,
      "num_tokens": 776616.0,
      "step": 445
    },
    {
      "entropy": 7.2496236801147464,
      "epoch": 0.35266457680250785,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002245,
      "loss": 6.9895,
      "mean_token_accuracy": 0.0921534925699234,
      "num_tokens": 786147.0,
      "step": 450
    },
    {
      "entropy": 7.356827306747436,
      "epoch": 0.3565830721003135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.0808,
      "mean_token_accuracy": 0.08939703032374383,
      "num_tokens": 795213.0,
      "step": 455
    },
    {
      "entropy": 7.332914733886719,
      "epoch": 0.3605015673981191,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8956,
      "mean_token_accuracy": 0.09629729315638542,
      "num_tokens": 803118.0,
      "step": 460
    },
    {
      "entropy": 7.346728849411011,
      "epoch": 0.36442006269592475,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8641,
      "mean_token_accuracy": 0.10322200134396553,
      "num_tokens": 811358.0,
      "step": 465
    },
    {
      "entropy": 7.147189569473267,
      "epoch": 0.3683385579937304,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8718,
      "mean_token_accuracy": 0.09597681015729904,
      "num_tokens": 819653.0,
      "step": 470
    },
    {
      "entropy": 7.195614671707153,
      "epoch": 0.3722570532915361,
      "grad_norm": 1.390625,
      "learning_rate": 0.000237,
      "loss": 6.9088,
      "mean_token_accuracy": 0.09375306144356728,
      "num_tokens": 828462.0,
      "step": 475
    },
    {
      "entropy": 7.361067628860473,
      "epoch": 0.3761755485893417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002395,
      "loss": 6.9995,
      "mean_token_accuracy": 0.09758970066905022,
      "num_tokens": 836204.0,
      "step": 480
    },
    {
      "entropy": 7.163541269302368,
      "epoch": 0.38009404388714735,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000242,
      "loss": 6.8697,
      "mean_token_accuracy": 0.1009294331073761,
      "num_tokens": 845032.0,
      "step": 485
    },
    {
      "entropy": 7.267932987213134,
      "epoch": 0.384012539184953,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002445,
      "loss": 6.8906,
      "mean_token_accuracy": 0.10076733604073525,
      "num_tokens": 853324.0,
      "step": 490
    },
    {
      "entropy": 7.26249213218689,
      "epoch": 0.3879310344827586,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000247,
      "loss": 6.8639,
      "mean_token_accuracy": 0.1036192737519741,
      "num_tokens": 861640.0,
      "step": 495
    },
    {
      "entropy": 7.264737176895141,
      "epoch": 0.39184952978056425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002495,
      "loss": 7.047,
      "mean_token_accuracy": 0.08870344087481499,
      "num_tokens": 870758.0,
      "step": 500
    },
    {
      "epoch": 0.39184952978056425,
      "eval_entropy": 7.3145306674084924,
      "eval_loss": 7.030722141265869,
      "eval_mean_token_accuracy": 0.09521551475708567,
      "eval_num_tokens": 870758.0,
      "eval_runtime": 2.8925,
      "eval_samples_per_second": 1425.051,
      "eval_steps_per_second": 178.391,
      "step": 500
    },
    {
      "entropy": 7.310623741149902,
      "epoch": 0.3957680250783699,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000252,
      "loss": 6.9526,
      "mean_token_accuracy": 0.08988267555832863,
      "num_tokens": 879968.0,
      "step": 505
    },
    {
      "entropy": 7.146830654144287,
      "epoch": 0.3996865203761755,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002545,
      "loss": 6.9074,
      "mean_token_accuracy": 0.09161595106124878,
      "num_tokens": 888804.0,
      "step": 510
    },
    {
      "entropy": 7.1906270503997805,
      "epoch": 0.4036050156739812,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000257,
      "loss": 6.9221,
      "mean_token_accuracy": 0.08995741233229637,
      "num_tokens": 898593.0,
      "step": 515
    },
    {
      "entropy": 7.173298168182373,
      "epoch": 0.40752351097178685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002595,
      "loss": 6.9368,
      "mean_token_accuracy": 0.09437091797590255,
      "num_tokens": 906873.0,
      "step": 520
    },
    {
      "entropy": 7.218977069854736,
      "epoch": 0.4114420062695925,
      "grad_norm": 1.140625,
      "learning_rate": 0.000262,
      "loss": 6.9019,
      "mean_token_accuracy": 0.10439103990793228,
      "num_tokens": 915080.0,
      "step": 525
    },
    {
      "entropy": 7.28611798286438,
      "epoch": 0.4153605015673981,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.911,
      "mean_token_accuracy": 0.0972377523779869,
      "num_tokens": 923741.0,
      "step": 530
    },
    {
      "entropy": 7.100715637207031,
      "epoch": 0.41927899686520376,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8103,
      "mean_token_accuracy": 0.10250237360596656,
      "num_tokens": 932455.0,
      "step": 535
    },
    {
      "entropy": 7.147137594223023,
      "epoch": 0.4231974921630094,
      "grad_norm": 1.25,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8173,
      "mean_token_accuracy": 0.10221570506691932,
      "num_tokens": 941064.0,
      "step": 540
    },
    {
      "entropy": 7.074875783920288,
      "epoch": 0.427115987460815,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8391,
      "mean_token_accuracy": 0.10051099881529808,
      "num_tokens": 950261.0,
      "step": 545
    },
    {
      "entropy": 7.238851642608642,
      "epoch": 0.43103448275862066,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002745,
      "loss": 6.9112,
      "mean_token_accuracy": 0.09698623940348625,
      "num_tokens": 959151.0,
      "step": 550
    },
    {
      "entropy": 7.171272468566895,
      "epoch": 0.43495297805642635,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000277,
      "loss": 6.934,
      "mean_token_accuracy": 0.09741523563861847,
      "num_tokens": 968441.0,
      "step": 555
    },
    {
      "entropy": 7.075249052047729,
      "epoch": 0.438871473354232,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002795,
      "loss": 6.8133,
      "mean_token_accuracy": 0.09270348027348518,
      "num_tokens": 977058.0,
      "step": 560
    },
    {
      "entropy": 7.114323377609253,
      "epoch": 0.4427899686520376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.8275,
      "mean_token_accuracy": 0.10172541886568069,
      "num_tokens": 986531.0,
      "step": 565
    },
    {
      "entropy": 7.010313701629639,
      "epoch": 0.44670846394984326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002845,
      "loss": 6.6613,
      "mean_token_accuracy": 0.10588647425174713,
      "num_tokens": 995081.0,
      "step": 570
    },
    {
      "entropy": 7.006834888458252,
      "epoch": 0.4506269592476489,
      "grad_norm": 1.34375,
      "learning_rate": 0.000287,
      "loss": 6.7384,
      "mean_token_accuracy": 0.10787067040801049,
      "num_tokens": 1003459.0,
      "step": 575
    },
    {
      "entropy": 7.008131837844848,
      "epoch": 0.45454545454545453,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002895,
      "loss": 6.8013,
      "mean_token_accuracy": 0.09799510389566421,
      "num_tokens": 1012420.0,
      "step": 580
    },
    {
      "entropy": 7.163480377197265,
      "epoch": 0.45846394984326017,
      "grad_norm": 1.421875,
      "learning_rate": 0.000292,
      "loss": 6.8601,
      "mean_token_accuracy": 0.09603445902466774,
      "num_tokens": 1021198.0,
      "step": 585
    },
    {
      "entropy": 7.066182661056518,
      "epoch": 0.46238244514106586,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002945,
      "loss": 6.8494,
      "mean_token_accuracy": 0.1006757028400898,
      "num_tokens": 1030023.0,
      "step": 590
    },
    {
      "entropy": 7.019779109954834,
      "epoch": 0.4663009404388715,
      "grad_norm": 1.296875,
      "learning_rate": 0.000297,
      "loss": 6.7127,
      "mean_token_accuracy": 0.10001757815480232,
      "num_tokens": 1038824.0,
      "step": 595
    },
    {
      "entropy": 7.121872615814209,
      "epoch": 0.4702194357366771,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002995,
      "loss": 6.9753,
      "mean_token_accuracy": 0.0965359665453434,
      "num_tokens": 1048207.0,
      "step": 600
    },
    {
      "entropy": 7.160170698165894,
      "epoch": 0.47413793103448276,
      "grad_norm": 1.25,
      "learning_rate": 0.000302,
      "loss": 6.9391,
      "mean_token_accuracy": 0.10044264048337936,
      "num_tokens": 1057354.0,
      "step": 605
    },
    {
      "entropy": 7.145117998123169,
      "epoch": 0.4780564263322884,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003045,
      "loss": 6.8633,
      "mean_token_accuracy": 0.0963246464729309,
      "num_tokens": 1066825.0,
      "step": 610
    },
    {
      "entropy": 7.0773111343383786,
      "epoch": 0.48197492163009403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000307,
      "loss": 6.7953,
      "mean_token_accuracy": 0.0973593220114708,
      "num_tokens": 1075752.0,
      "step": 615
    },
    {
      "entropy": 6.937968826293945,
      "epoch": 0.48589341692789967,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003095,
      "loss": 6.7024,
      "mean_token_accuracy": 0.10333408787846565,
      "num_tokens": 1084826.0,
      "step": 620
    },
    {
      "entropy": 7.09277081489563,
      "epoch": 0.4898119122257053,
      "grad_norm": 1.234375,
      "learning_rate": 0.000312,
      "loss": 6.854,
      "mean_token_accuracy": 0.09841757342219352,
      "num_tokens": 1093425.0,
      "step": 625
    },
    {
      "entropy": 7.049381494522095,
      "epoch": 0.493730407523511,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003145,
      "loss": 6.8692,
      "mean_token_accuracy": 0.09519068449735642,
      "num_tokens": 1102619.0,
      "step": 630
    },
    {
      "entropy": 7.003838348388672,
      "epoch": 0.49764890282131663,
      "grad_norm": 1.203125,
      "learning_rate": 0.000317,
      "loss": 6.7563,
      "mean_token_accuracy": 0.10078713521361352,
      "num_tokens": 1111337.0,
      "step": 635
    },
    {
      "entropy": 6.916393804550171,
      "epoch": 0.5015673981191222,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003195,
      "loss": 6.5811,
      "mean_token_accuracy": 0.11013391688466072,
      "num_tokens": 1119381.0,
      "step": 640
    },
    {
      "entropy": 6.888113069534302,
      "epoch": 0.5054858934169278,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000322,
      "loss": 6.6508,
      "mean_token_accuracy": 0.10688698589801789,
      "num_tokens": 1127825.0,
      "step": 645
    },
    {
      "entropy": 7.003930568695068,
      "epoch": 0.5094043887147336,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.6994,
      "mean_token_accuracy": 0.09896195456385612,
      "num_tokens": 1136750.0,
      "step": 650
    },
    {
      "entropy": 6.882824468612671,
      "epoch": 0.5133228840125392,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.7242,
      "mean_token_accuracy": 0.10310067981481552,
      "num_tokens": 1145739.0,
      "step": 655
    },
    {
      "entropy": 6.920678424835205,
      "epoch": 0.5172413793103449,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.7827,
      "mean_token_accuracy": 0.10021716952323914,
      "num_tokens": 1154415.0,
      "step": 660
    },
    {
      "entropy": 6.999461698532104,
      "epoch": 0.5211598746081505,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6635,
      "mean_token_accuracy": 0.10259130671620369,
      "num_tokens": 1162480.0,
      "step": 665
    },
    {
      "entropy": 6.945166110992432,
      "epoch": 0.5250783699059561,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.7478,
      "mean_token_accuracy": 0.10068432837724686,
      "num_tokens": 1170774.0,
      "step": 670
    },
    {
      "entropy": 6.878892850875855,
      "epoch": 0.5289968652037618,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000337,
      "loss": 6.6879,
      "mean_token_accuracy": 0.10098834857344627,
      "num_tokens": 1179629.0,
      "step": 675
    },
    {
      "entropy": 6.955817317962646,
      "epoch": 0.5329153605015674,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003395,
      "loss": 6.8369,
      "mean_token_accuracy": 0.09804248735308647,
      "num_tokens": 1189111.0,
      "step": 680
    },
    {
      "entropy": 6.9712419509887695,
      "epoch": 0.536833855799373,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000342,
      "loss": 6.8281,
      "mean_token_accuracy": 0.0970684178173542,
      "num_tokens": 1198827.0,
      "step": 685
    },
    {
      "entropy": 7.076535654067993,
      "epoch": 0.5407523510971787,
      "grad_norm": 1.25,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.7525,
      "mean_token_accuracy": 0.10446157231926918,
      "num_tokens": 1207089.0,
      "step": 690
    },
    {
      "entropy": 6.935127210617066,
      "epoch": 0.5446708463949843,
      "grad_norm": 1.28125,
      "learning_rate": 0.000347,
      "loss": 6.736,
      "mean_token_accuracy": 0.09537649154663086,
      "num_tokens": 1216509.0,
      "step": 695
    },
    {
      "entropy": 6.816225671768189,
      "epoch": 0.54858934169279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003495,
      "loss": 6.5657,
      "mean_token_accuracy": 0.11179759725928307,
      "num_tokens": 1225684.0,
      "step": 700
    },
    {
      "entropy": 6.831858444213867,
      "epoch": 0.5525078369905956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000352,
      "loss": 6.6121,
      "mean_token_accuracy": 0.10936803743243217,
      "num_tokens": 1233675.0,
      "step": 705
    },
    {
      "entropy": 6.846443557739258,
      "epoch": 0.5564263322884012,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003545,
      "loss": 6.621,
      "mean_token_accuracy": 0.1022804893553257,
      "num_tokens": 1242147.0,
      "step": 710
    },
    {
      "entropy": 6.9075483798980715,
      "epoch": 0.5603448275862069,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000357,
      "loss": 6.7254,
      "mean_token_accuracy": 0.10935754925012589,
      "num_tokens": 1251127.0,
      "step": 715
    },
    {
      "entropy": 6.810165548324585,
      "epoch": 0.5642633228840125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003595,
      "loss": 6.6885,
      "mean_token_accuracy": 0.10543572083115578,
      "num_tokens": 1260589.0,
      "step": 720
    },
    {
      "entropy": 6.9715752601623535,
      "epoch": 0.5681818181818182,
      "grad_norm": 1.234375,
      "learning_rate": 0.000362,
      "loss": 6.7526,
      "mean_token_accuracy": 0.10584053248167039,
      "num_tokens": 1269281.0,
      "step": 725
    },
    {
      "entropy": 6.892933082580567,
      "epoch": 0.5721003134796239,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003645,
      "loss": 6.6119,
      "mean_token_accuracy": 0.1029752716422081,
      "num_tokens": 1278033.0,
      "step": 730
    },
    {
      "entropy": 6.8791444301605225,
      "epoch": 0.5760188087774295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000367,
      "loss": 6.6888,
      "mean_token_accuracy": 0.10946752950549125,
      "num_tokens": 1285847.0,
      "step": 735
    },
    {
      "entropy": 6.881000947952271,
      "epoch": 0.5799373040752351,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003695,
      "loss": 6.6685,
      "mean_token_accuracy": 0.1023192547261715,
      "num_tokens": 1294770.0,
      "step": 740
    },
    {
      "entropy": 6.975464582443237,
      "epoch": 0.5838557993730408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372,
      "loss": 6.8736,
      "mean_token_accuracy": 0.09663443863391877,
      "num_tokens": 1304282.0,
      "step": 745
    },
    {
      "entropy": 6.9174909591674805,
      "epoch": 0.5877742946708464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745,
      "loss": 6.7056,
      "mean_token_accuracy": 0.10080768540501595,
      "num_tokens": 1313218.0,
      "step": 750
    },
    {
      "entropy": 6.916672325134277,
      "epoch": 0.591692789968652,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000377,
      "loss": 6.6713,
      "mean_token_accuracy": 0.10844261273741722,
      "num_tokens": 1322093.0,
      "step": 755
    },
    {
      "entropy": 6.670488548278809,
      "epoch": 0.5956112852664577,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003795,
      "loss": 6.7051,
      "mean_token_accuracy": 0.10560382381081582,
      "num_tokens": 1330718.0,
      "step": 760
    },
    {
      "entropy": 6.887430095672608,
      "epoch": 0.5995297805642633,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000382,
      "loss": 6.5598,
      "mean_token_accuracy": 0.1074932336807251,
      "num_tokens": 1339672.0,
      "step": 765
    },
    {
      "entropy": 6.805772209167481,
      "epoch": 0.603448275862069,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003845,
      "loss": 6.5984,
      "mean_token_accuracy": 0.1087186798453331,
      "num_tokens": 1348587.0,
      "step": 770
    },
    {
      "entropy": 6.776034593582153,
      "epoch": 0.6073667711598746,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5901,
      "mean_token_accuracy": 0.11292538270354271,
      "num_tokens": 1357597.0,
      "step": 775
    },
    {
      "entropy": 6.818142938613891,
      "epoch": 0.6112852664576802,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.6842,
      "mean_token_accuracy": 0.10455552414059639,
      "num_tokens": 1366246.0,
      "step": 780
    },
    {
      "entropy": 6.812788200378418,
      "epoch": 0.6152037617554859,
      "grad_norm": 1.25,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.596,
      "mean_token_accuracy": 0.10609478428959847,
      "num_tokens": 1374664.0,
      "step": 785
    },
    {
      "entropy": 6.736437702178955,
      "epoch": 0.6191222570532915,
      "grad_norm": 1.25,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.6125,
      "mean_token_accuracy": 0.10844489559531212,
      "num_tokens": 1382765.0,
      "step": 790
    },
    {
      "entropy": 6.722091817855835,
      "epoch": 0.6230407523510971,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5665,
      "mean_token_accuracy": 0.10965784117579461,
      "num_tokens": 1391708.0,
      "step": 795
    },
    {
      "entropy": 6.765339946746826,
      "epoch": 0.6269592476489029,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003995,
      "loss": 6.7346,
      "mean_token_accuracy": 0.10098599940538407,
      "num_tokens": 1401373.0,
      "step": 800
    },
    {
      "entropy": 6.851631355285645,
      "epoch": 0.6308777429467085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 6.5225,
      "mean_token_accuracy": 0.11173290610313416,
      "num_tokens": 1409702.0,
      "step": 805
    },
    {
      "entropy": 6.767954397201538,
      "epoch": 0.6347962382445141,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004045,
      "loss": 6.7358,
      "mean_token_accuracy": 0.10304761379957199,
      "num_tokens": 1419596.0,
      "step": 810
    },
    {
      "entropy": 6.824428558349609,
      "epoch": 0.6387147335423198,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.6758,
      "mean_token_accuracy": 0.1033481739461422,
      "num_tokens": 1428437.0,
      "step": 815
    },
    {
      "entropy": 6.820134544372559,
      "epoch": 0.6426332288401254,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004095,
      "loss": 6.5833,
      "mean_token_accuracy": 0.10861420258879662,
      "num_tokens": 1436764.0,
      "step": 820
    },
    {
      "entropy": 6.72536416053772,
      "epoch": 0.646551724137931,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000412,
      "loss": 6.6309,
      "mean_token_accuracy": 0.11201521828770637,
      "num_tokens": 1445755.0,
      "step": 825
    },
    {
      "entropy": 6.854031276702881,
      "epoch": 0.6504702194357367,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004145,
      "loss": 6.6969,
      "mean_token_accuracy": 0.10224815532565117,
      "num_tokens": 1454550.0,
      "step": 830
    },
    {
      "entropy": 6.7012652397155765,
      "epoch": 0.6543887147335423,
      "grad_norm": 1.171875,
      "learning_rate": 0.000417,
      "loss": 6.5611,
      "mean_token_accuracy": 0.11515542790293694,
      "num_tokens": 1463039.0,
      "step": 835
    },
    {
      "entropy": 6.802173805236817,
      "epoch": 0.658307210031348,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004195,
      "loss": 6.623,
      "mean_token_accuracy": 0.10527266189455986,
      "num_tokens": 1472029.0,
      "step": 840
    },
    {
      "entropy": 6.709125900268555,
      "epoch": 0.6622257053291536,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422,
      "loss": 6.6453,
      "mean_token_accuracy": 0.10808795914053917,
      "num_tokens": 1481588.0,
      "step": 845
    },
    {
      "entropy": 6.68847393989563,
      "epoch": 0.6661442006269592,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004245,
      "loss": 6.545,
      "mean_token_accuracy": 0.10692245215177536,
      "num_tokens": 1489915.0,
      "step": 850
    },
    {
      "entropy": 6.784275102615356,
      "epoch": 0.6700626959247649,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000427,
      "loss": 6.6288,
      "mean_token_accuracy": 0.10608619600534439,
      "num_tokens": 1497920.0,
      "step": 855
    },
    {
      "entropy": 6.68899974822998,
      "epoch": 0.6739811912225705,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004295,
      "loss": 6.4869,
      "mean_token_accuracy": 0.10574968531727791,
      "num_tokens": 1506965.0,
      "step": 860
    },
    {
      "entropy": 6.722927284240723,
      "epoch": 0.6778996865203761,
      "grad_norm": 1.1875,
      "learning_rate": 0.000432,
      "loss": 6.5838,
      "mean_token_accuracy": 0.10984734967350959,
      "num_tokens": 1516028.0,
      "step": 865
    },
    {
      "entropy": 6.755547428131104,
      "epoch": 0.6818181818181818,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004345,
      "loss": 6.6197,
      "mean_token_accuracy": 0.10767295137047768,
      "num_tokens": 1524579.0,
      "step": 870
    },
    {
      "entropy": 6.617767429351806,
      "epoch": 0.6857366771159875,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 6.5326,
      "mean_token_accuracy": 0.11009506732225419,
      "num_tokens": 1534285.0,
      "step": 875
    },
    {
      "entropy": 6.716221046447754,
      "epoch": 0.6896551724137931,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004395,
      "loss": 6.5287,
      "mean_token_accuracy": 0.11053434684872628,
      "num_tokens": 1543701.0,
      "step": 880
    },
    {
      "entropy": 6.79583706855774,
      "epoch": 0.6935736677115988,
      "grad_norm": 1.0625,
      "learning_rate": 0.000442,
      "loss": 6.7919,
      "mean_token_accuracy": 0.09309381395578384,
      "num_tokens": 1552715.0,
      "step": 885
    },
    {
      "entropy": 6.501706457138061,
      "epoch": 0.6974921630094044,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445,
      "loss": 6.4006,
      "mean_token_accuracy": 0.11031684949994087,
      "num_tokens": 1561523.0,
      "step": 890
    },
    {
      "entropy": 6.685598754882813,
      "epoch": 0.70141065830721,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000447,
      "loss": 6.6289,
      "mean_token_accuracy": 0.11112729758024216,
      "num_tokens": 1570057.0,
      "step": 895
    },
    {
      "entropy": 6.727664756774902,
      "epoch": 0.7053291536050157,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4786,
      "mean_token_accuracy": 0.11088098585605621,
      "num_tokens": 1578795.0,
      "step": 900
    },
    {
      "entropy": 6.55871696472168,
      "epoch": 0.7092476489028213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5632,
      "mean_token_accuracy": 0.10584904998540878,
      "num_tokens": 1588108.0,
      "step": 905
    },
    {
      "entropy": 6.729115009307861,
      "epoch": 0.713166144200627,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.5254,
      "mean_token_accuracy": 0.11291411146521568,
      "num_tokens": 1596545.0,
      "step": 910
    },
    {
      "entropy": 6.622742557525635,
      "epoch": 0.7170846394984326,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.5094,
      "mean_token_accuracy": 0.11332128643989563,
      "num_tokens": 1605098.0,
      "step": 915
    },
    {
      "entropy": 6.602113914489746,
      "epoch": 0.7210031347962382,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.5232,
      "mean_token_accuracy": 0.1159414529800415,
      "num_tokens": 1613760.0,
      "step": 920
    },
    {
      "entropy": 6.651577711105347,
      "epoch": 0.7249216300940439,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000462,
      "loss": 6.5629,
      "mean_token_accuracy": 0.10667054280638695,
      "num_tokens": 1622345.0,
      "step": 925
    },
    {
      "entropy": 6.7364403247833256,
      "epoch": 0.7288401253918495,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004645,
      "loss": 6.5521,
      "mean_token_accuracy": 0.11116818189620972,
      "num_tokens": 1630706.0,
      "step": 930
    },
    {
      "entropy": 6.705845355987549,
      "epoch": 0.7327586206896551,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000467,
      "loss": 6.624,
      "mean_token_accuracy": 0.10690407380461693,
      "num_tokens": 1639880.0,
      "step": 935
    },
    {
      "entropy": 6.593637418746948,
      "epoch": 0.7366771159874608,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004695,
      "loss": 6.4648,
      "mean_token_accuracy": 0.10600230619311332,
      "num_tokens": 1648325.0,
      "step": 940
    },
    {
      "entropy": 6.491366624832153,
      "epoch": 0.7405956112852664,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000472,
      "loss": 6.4219,
      "mean_token_accuracy": 0.11530695110559464,
      "num_tokens": 1656951.0,
      "step": 945
    },
    {
      "entropy": 6.594129180908203,
      "epoch": 0.7445141065830722,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004745,
      "loss": 6.5604,
      "mean_token_accuracy": 0.11204110458493233,
      "num_tokens": 1665832.0,
      "step": 950
    },
    {
      "entropy": 6.592755603790283,
      "epoch": 0.7484326018808778,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000477,
      "loss": 6.4841,
      "mean_token_accuracy": 0.10641552805900574,
      "num_tokens": 1675116.0,
      "step": 955
    },
    {
      "entropy": 6.512726259231568,
      "epoch": 0.7523510971786834,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004795,
      "loss": 6.4685,
      "mean_token_accuracy": 0.11560196354985237,
      "num_tokens": 1684195.0,
      "step": 960
    },
    {
      "entropy": 6.613996267318726,
      "epoch": 0.7562695924764891,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000482,
      "loss": 6.4407,
      "mean_token_accuracy": 0.11013077944517136,
      "num_tokens": 1693025.0,
      "step": 965
    },
    {
      "entropy": 6.462671899795533,
      "epoch": 0.7601880877742947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004845,
      "loss": 6.4294,
      "mean_token_accuracy": 0.11215238943696022,
      "num_tokens": 1701210.0,
      "step": 970
    },
    {
      "entropy": 6.602805709838867,
      "epoch": 0.7641065830721003,
      "grad_norm": 1.25,
      "learning_rate": 0.000487,
      "loss": 6.5362,
      "mean_token_accuracy": 0.10191687792539597,
      "num_tokens": 1711026.0,
      "step": 975
    },
    {
      "entropy": 6.6010805606842045,
      "epoch": 0.768025078369906,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004895,
      "loss": 6.5314,
      "mean_token_accuracy": 0.11036548092961311,
      "num_tokens": 1719441.0,
      "step": 980
    },
    {
      "entropy": 6.571003103256226,
      "epoch": 0.7719435736677116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000492,
      "loss": 6.5292,
      "mean_token_accuracy": 0.10878988429903984,
      "num_tokens": 1728300.0,
      "step": 985
    },
    {
      "entropy": 6.600493288040161,
      "epoch": 0.7758620689655172,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004945,
      "loss": 6.4193,
      "mean_token_accuracy": 0.11556374430656433,
      "num_tokens": 1736364.0,
      "step": 990
    },
    {
      "entropy": 6.384331512451172,
      "epoch": 0.7797805642633229,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497,
      "loss": 6.259,
      "mean_token_accuracy": 0.12344810739159584,
      "num_tokens": 1745043.0,
      "step": 995
    },
    {
      "entropy": 6.534456634521485,
      "epoch": 0.7836990595611285,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995,
      "loss": 6.4121,
      "mean_token_accuracy": 0.11444604694843293,
      "num_tokens": 1754008.0,
      "step": 1000
    },
    {
      "epoch": 0.7836990595611285,
      "eval_entropy": 6.353288079417029,
      "eval_loss": 6.573922157287598,
      "eval_mean_token_accuracy": 0.11509622025570666,
      "eval_num_tokens": 1754008.0,
      "eval_runtime": 2.8912,
      "eval_samples_per_second": 1425.689,
      "eval_steps_per_second": 178.471,
      "step": 1000
    },
    {
      "entropy": 6.410930061340332,
      "epoch": 0.7876175548589341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999998713243189,
      "loss": 6.4673,
      "mean_token_accuracy": 0.10981100052595139,
      "num_tokens": 1763592.0,
      "step": 1005
    },
    {
      "entropy": 6.565805721282959,
      "epoch": 0.7915360501567398,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999993485796164,
      "loss": 6.3337,
      "mean_token_accuracy": 0.11553042531013488,
      "num_tokens": 1773111.0,
      "step": 1010
    },
    {
      "entropy": 6.497714471817017,
      "epoch": 0.7954545454545454,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999984237245962,
      "loss": 6.4084,
      "mean_token_accuracy": 0.12054920345544815,
      "num_tokens": 1781230.0,
      "step": 1015
    },
    {
      "entropy": 6.660019779205323,
      "epoch": 0.799373040752351,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004999970967609109,
      "loss": 6.6371,
      "mean_token_accuracy": 0.11188422217965126,
      "num_tokens": 1790324.0,
      "step": 1020
    },
    {
      "entropy": 6.494213390350342,
      "epoch": 0.8032915360501567,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999953676909322,
      "loss": 6.4729,
      "mean_token_accuracy": 0.10698359534144401,
      "num_tokens": 1798849.0,
      "step": 1025
    },
    {
      "entropy": 6.546792268753052,
      "epoch": 0.8072100313479624,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004999932365177502,
      "loss": 6.3552,
      "mean_token_accuracy": 0.11793587505817413,
      "num_tokens": 1808019.0,
      "step": 1030
    },
    {
      "entropy": 6.554946851730347,
      "epoch": 0.8111285266457681,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999907032451735,
      "loss": 6.467,
      "mean_token_accuracy": 0.11289061605930328,
      "num_tokens": 1816550.0,
      "step": 1035
    },
    {
      "entropy": 6.432303428649902,
      "epoch": 0.8150470219435737,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999877678777296,
      "loss": 6.3661,
      "mean_token_accuracy": 0.11930875778198242,
      "num_tokens": 1825238.0,
      "step": 1040
    },
    {
      "entropy": 6.448621463775635,
      "epoch": 0.8189655172413793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999844304206645,
      "loss": 6.3297,
      "mean_token_accuracy": 0.11632873341441155,
      "num_tokens": 1833645.0,
      "step": 1045
    },
    {
      "entropy": 6.457649612426758,
      "epoch": 0.822884012539185,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999806908799428,
      "loss": 6.4009,
      "mean_token_accuracy": 0.11582175865769387,
      "num_tokens": 1842892.0,
      "step": 1050
    },
    {
      "entropy": 6.440515804290771,
      "epoch": 0.8268025078369906,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999765492622474,
      "loss": 6.4502,
      "mean_token_accuracy": 0.11736933290958404,
      "num_tokens": 1851400.0,
      "step": 1055
    },
    {
      "entropy": 6.566648626327515,
      "epoch": 0.8307210031347962,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999720055749804,
      "loss": 6.3827,
      "mean_token_accuracy": 0.11811394691467285,
      "num_tokens": 1860088.0,
      "step": 1060
    },
    {
      "entropy": 6.509423923492432,
      "epoch": 0.8346394984326019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999670598262619,
      "loss": 6.4453,
      "mean_token_accuracy": 0.1164441391825676,
      "num_tokens": 1868558.0,
      "step": 1065
    },
    {
      "entropy": 6.43409571647644,
      "epoch": 0.8385579937304075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999617120249308,
      "loss": 6.4301,
      "mean_token_accuracy": 0.11259575486183167,
      "num_tokens": 1877624.0,
      "step": 1070
    },
    {
      "entropy": 6.559784078598023,
      "epoch": 0.8424764890282131,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999559621805445,
      "loss": 6.5347,
      "mean_token_accuracy": 0.1132730558514595,
      "num_tokens": 1886505.0,
      "step": 1075
    },
    {
      "entropy": 6.372117948532105,
      "epoch": 0.8463949843260188,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999498103033788,
      "loss": 6.2892,
      "mean_token_accuracy": 0.12123534381389618,
      "num_tokens": 1895701.0,
      "step": 1080
    },
    {
      "entropy": 6.388390398025512,
      "epoch": 0.8503134796238244,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999432564044284,
      "loss": 6.2646,
      "mean_token_accuracy": 0.11819472461938858,
      "num_tokens": 1904640.0,
      "step": 1085
    },
    {
      "entropy": 6.498187494277954,
      "epoch": 0.85423197492163,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999363004954058,
      "loss": 6.4879,
      "mean_token_accuracy": 0.11307620778679847,
      "num_tokens": 1914216.0,
      "step": 1090
    },
    {
      "entropy": 6.445917701721191,
      "epoch": 0.8581504702194357,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999289425887425,
      "loss": 6.3985,
      "mean_token_accuracy": 0.12216848358511925,
      "num_tokens": 1923280.0,
      "step": 1095
    },
    {
      "entropy": 6.525809192657471,
      "epoch": 0.8620689655172413,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999211826975884,
      "loss": 6.4166,
      "mean_token_accuracy": 0.12132912650704383,
      "num_tokens": 1931403.0,
      "step": 1100
    },
    {
      "entropy": 6.520140981674194,
      "epoch": 0.8659874608150471,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999130208358114,
      "loss": 6.429,
      "mean_token_accuracy": 0.11339464783668518,
      "num_tokens": 1940774.0,
      "step": 1105
    },
    {
      "entropy": 6.503523302078247,
      "epoch": 0.8699059561128527,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999044570179983,
      "loss": 6.5221,
      "mean_token_accuracy": 0.1174374520778656,
      "num_tokens": 1950045.0,
      "step": 1110
    },
    {
      "entropy": 6.526077318191528,
      "epoch": 0.8738244514106583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998954912594541,
      "loss": 6.3923,
      "mean_token_accuracy": 0.11855587661266327,
      "num_tokens": 1959303.0,
      "step": 1115
    },
    {
      "entropy": 6.462242603302002,
      "epoch": 0.877742946708464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998861235762018,
      "loss": 6.3603,
      "mean_token_accuracy": 0.11869992539286614,
      "num_tokens": 1968583.0,
      "step": 1120
    },
    {
      "entropy": 6.474194145202636,
      "epoch": 0.8816614420062696,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998763539849832,
      "loss": 6.4832,
      "mean_token_accuracy": 0.11621114015579223,
      "num_tokens": 1977489.0,
      "step": 1125
    },
    {
      "entropy": 6.394860315322876,
      "epoch": 0.8855799373040752,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998661825032579,
      "loss": 6.2376,
      "mean_token_accuracy": 0.12140937894582748,
      "num_tokens": 1985308.0,
      "step": 1130
    },
    {
      "entropy": 6.435638904571533,
      "epoch": 0.8894984326018809,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998556091492043,
      "loss": 6.4123,
      "mean_token_accuracy": 0.10934653729200364,
      "num_tokens": 1993930.0,
      "step": 1135
    },
    {
      "entropy": 6.399746656417847,
      "epoch": 0.8934169278996865,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998446339417184,
      "loss": 6.4064,
      "mean_token_accuracy": 0.11175781339406968,
      "num_tokens": 2003802.0,
      "step": 1140
    },
    {
      "entropy": 6.482501697540283,
      "epoch": 0.8973354231974922,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998332569004149,
      "loss": 6.3189,
      "mean_token_accuracy": 0.12161151990294457,
      "num_tokens": 2011957.0,
      "step": 1145
    },
    {
      "entropy": 6.3887701511383055,
      "epoch": 0.9012539184952978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998214780456263,
      "loss": 6.314,
      "mean_token_accuracy": 0.12478633299469948,
      "num_tokens": 2020741.0,
      "step": 1150
    },
    {
      "entropy": 6.356020259857178,
      "epoch": 0.9051724137931034,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998092973984033,
      "loss": 6.41,
      "mean_token_accuracy": 0.11564783379435539,
      "num_tokens": 2029804.0,
      "step": 1155
    },
    {
      "entropy": 6.55821795463562,
      "epoch": 0.9090909090909091,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997967149805147,
      "loss": 6.4492,
      "mean_token_accuracy": 0.11650254800915719,
      "num_tokens": 2038732.0,
      "step": 1160
    },
    {
      "entropy": 6.412743139266968,
      "epoch": 0.9130094043887147,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997837308144474,
      "loss": 6.3322,
      "mean_token_accuracy": 0.1288091853260994,
      "num_tokens": 2046362.0,
      "step": 1165
    },
    {
      "entropy": 6.40589394569397,
      "epoch": 0.9169278996865203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997703449234062,
      "loss": 6.3273,
      "mean_token_accuracy": 0.11811700761318207,
      "num_tokens": 2056018.0,
      "step": 1170
    },
    {
      "entropy": 6.326677036285401,
      "epoch": 0.920846394984326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997565573313139,
      "loss": 6.3065,
      "mean_token_accuracy": 0.11553404182195663,
      "num_tokens": 2065077.0,
      "step": 1175
    },
    {
      "entropy": 6.469640636444092,
      "epoch": 0.9247648902821317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997423680628109,
      "loss": 6.295,
      "mean_token_accuracy": 0.11463245674967766,
      "num_tokens": 2073405.0,
      "step": 1180
    },
    {
      "entropy": 6.21384630203247,
      "epoch": 0.9286833855799373,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499727777143256,
      "loss": 6.2785,
      "mean_token_accuracy": 0.11823253557085991,
      "num_tokens": 2081891.0,
      "step": 1185
    },
    {
      "entropy": 6.534471082687378,
      "epoch": 0.932601880877743,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997127845987255,
      "loss": 6.3293,
      "mean_token_accuracy": 0.12212807089090347,
      "num_tokens": 2090732.0,
      "step": 1190
    },
    {
      "entropy": 6.3433879852294925,
      "epoch": 0.9365203761755486,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996973904560135,
      "loss": 6.2735,
      "mean_token_accuracy": 0.12710270062088966,
      "num_tokens": 2099352.0,
      "step": 1195
    },
    {
      "entropy": 6.382493019104004,
      "epoch": 0.9404388714733543,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996815947426317,
      "loss": 6.2946,
      "mean_token_accuracy": 0.1185976468026638,
      "num_tokens": 2107365.0,
      "step": 1200
    },
    {
      "entropy": 6.328860521316528,
      "epoch": 0.9443573667711599,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996653974868097,
      "loss": 6.1709,
      "mean_token_accuracy": 0.12477589845657348,
      "num_tokens": 2116192.0,
      "step": 1205
    },
    {
      "entropy": 6.279233121871949,
      "epoch": 0.9482758620689655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996487987174946,
      "loss": 6.2726,
      "mean_token_accuracy": 0.11757147014141082,
      "num_tokens": 2124410.0,
      "step": 1210
    },
    {
      "entropy": 6.326535558700561,
      "epoch": 0.9521943573667712,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996317984643512,
      "loss": 6.2848,
      "mean_token_accuracy": 0.12036795765161515,
      "num_tokens": 2132799.0,
      "step": 1215
    },
    {
      "entropy": 6.467973375320435,
      "epoch": 0.9561128526645768,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996143967577615,
      "loss": 6.4498,
      "mean_token_accuracy": 0.11489988937973976,
      "num_tokens": 2141589.0,
      "step": 1220
    },
    {
      "entropy": 6.386198425292969,
      "epoch": 0.9600313479623824,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995965936288254,
      "loss": 6.3403,
      "mean_token_accuracy": 0.12025970742106437,
      "num_tokens": 2150704.0,
      "step": 1225
    },
    {
      "entropy": 6.3113349914550785,
      "epoch": 0.9639498432601881,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995783891093597,
      "loss": 6.2607,
      "mean_token_accuracy": 0.12228266969323158,
      "num_tokens": 2159587.0,
      "step": 1230
    },
    {
      "entropy": 6.234415531158447,
      "epoch": 0.9678683385579937,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499559783231899,
      "loss": 6.2409,
      "mean_token_accuracy": 0.11728857606649398,
      "num_tokens": 2169071.0,
      "step": 1235
    },
    {
      "entropy": 6.3988946914672855,
      "epoch": 0.9717868338557993,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995407760296952,
      "loss": 6.3251,
      "mean_token_accuracy": 0.11511705070734024,
      "num_tokens": 2178248.0,
      "step": 1240
    },
    {
      "entropy": 6.396841096878052,
      "epoch": 0.975705329153605,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995213675367169,
      "loss": 6.229,
      "mean_token_accuracy": 0.11794086545705795,
      "num_tokens": 2187520.0,
      "step": 1245
    },
    {
      "entropy": 6.2435925006866455,
      "epoch": 0.9796238244514106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995015577876504,
      "loss": 6.3007,
      "mean_token_accuracy": 0.11987854838371277,
      "num_tokens": 2196861.0,
      "step": 1250
    },
    {
      "entropy": 6.314880323410034,
      "epoch": 0.9835423197492164,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499481346817899,
      "loss": 6.2143,
      "mean_token_accuracy": 0.12193886116147042,
      "num_tokens": 2205523.0,
      "step": 1255
    },
    {
      "entropy": 6.318439245223999,
      "epoch": 0.987460815047022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994607346635829,
      "loss": 6.1652,
      "mean_token_accuracy": 0.12974872291088105,
      "num_tokens": 2214291.0,
      "step": 1260
    },
    {
      "entropy": 6.366824340820313,
      "epoch": 0.9913793103448276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994397213615394,
      "loss": 6.2818,
      "mean_token_accuracy": 0.12176446989178658,
      "num_tokens": 2223093.0,
      "step": 1265
    },
    {
      "entropy": 6.331199264526367,
      "epoch": 0.9952978056426333,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994183069493225,
      "loss": 6.2364,
      "mean_token_accuracy": 0.11954401284456254,
      "num_tokens": 2231551.0,
      "step": 1270
    },
    {
      "entropy": 6.184268236160278,
      "epoch": 0.9992163009404389,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993964914652034,
      "loss": 6.2306,
      "mean_token_accuracy": 0.12992196902632713,
      "num_tokens": 2240650.0,
      "step": 1275
    },
    {
      "entropy": 6.461467027664185,
      "epoch": 1.0031347962382444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993742749481699,
      "loss": 6.1323,
      "mean_token_accuracy": 0.12565431445837022,
      "num_tokens": 2249392.0,
      "step": 1280
    },
    {
      "entropy": 6.129501914978027,
      "epoch": 1.0070532915360502,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993516574379265,
      "loss": 5.9558,
      "mean_token_accuracy": 0.13117464035749435,
      "num_tokens": 2258100.0,
      "step": 1285
    },
    {
      "entropy": 6.233500862121582,
      "epoch": 1.0109717868338557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993286389748943,
      "loss": 6.0332,
      "mean_token_accuracy": 0.12863276079297065,
      "num_tokens": 2266661.0,
      "step": 1290
    },
    {
      "entropy": 6.204838132858276,
      "epoch": 1.0148902821316614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993052196002112,
      "loss": 6.1322,
      "mean_token_accuracy": 0.12399615719914436,
      "num_tokens": 2275475.0,
      "step": 1295
    },
    {
      "entropy": 6.244740676879883,
      "epoch": 1.0188087774294672,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992813993557312,
      "loss": 6.0349,
      "mean_token_accuracy": 0.13146774247288703,
      "num_tokens": 2284077.0,
      "step": 1300
    },
    {
      "entropy": 6.1790135383605955,
      "epoch": 1.0227272727272727,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499257178284025,
      "loss": 6.0488,
      "mean_token_accuracy": 0.12352109551429749,
      "num_tokens": 2293029.0,
      "step": 1305
    },
    {
      "entropy": 6.212559843063355,
      "epoch": 1.0266457680250785,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992325564283797,
      "loss": 5.9596,
      "mean_token_accuracy": 0.14028860479593278,
      "num_tokens": 2301574.0,
      "step": 1310
    },
    {
      "entropy": 6.061412286758423,
      "epoch": 1.030564263322884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992075338327984,
      "loss": 5.9615,
      "mean_token_accuracy": 0.13310449346899986,
      "num_tokens": 2310366.0,
      "step": 1315
    },
    {
      "entropy": 6.29543924331665,
      "epoch": 1.0344827586206897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991821105420006,
      "loss": 6.0925,
      "mean_token_accuracy": 0.12402413934469222,
      "num_tokens": 2318943.0,
      "step": 1320
    },
    {
      "entropy": 6.23583607673645,
      "epoch": 1.0384012539184952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991562866014219,
      "loss": 6.0991,
      "mean_token_accuracy": 0.1307850494980812,
      "num_tokens": 2327173.0,
      "step": 1325
    },
    {
      "entropy": 6.260228252410888,
      "epoch": 1.042319749216301,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991300620572138,
      "loss": 6.0618,
      "mean_token_accuracy": 0.12725646793842316,
      "num_tokens": 2336194.0,
      "step": 1330
    },
    {
      "entropy": 6.191347360610962,
      "epoch": 1.0462382445141065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991034369562438,
      "loss": 6.1144,
      "mean_token_accuracy": 0.12965716123580934,
      "num_tokens": 2344721.0,
      "step": 1335
    },
    {
      "entropy": 6.197571229934693,
      "epoch": 1.0501567398119123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990764113460953,
      "loss": 6.0465,
      "mean_token_accuracy": 0.12880703136324884,
      "num_tokens": 2353406.0,
      "step": 1340
    },
    {
      "entropy": 6.303739309310913,
      "epoch": 1.0540752351097178,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990489852750675,
      "loss": 6.2162,
      "mean_token_accuracy": 0.11973949447274208,
      "num_tokens": 2361896.0,
      "step": 1345
    },
    {
      "entropy": 6.172640943527222,
      "epoch": 1.0579937304075235,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990211587921751,
      "loss": 6.0466,
      "mean_token_accuracy": 0.12820285186171532,
      "num_tokens": 2371861.0,
      "step": 1350
    },
    {
      "entropy": 6.191204833984375,
      "epoch": 1.061912225705329,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989929319471487,
      "loss": 5.9731,
      "mean_token_accuracy": 0.1299687497317791,
      "num_tokens": 2381555.0,
      "step": 1355
    },
    {
      "entropy": 6.136105632781982,
      "epoch": 1.0658307210031348,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989643047904341,
      "loss": 5.9618,
      "mean_token_accuracy": 0.13669807985424995,
      "num_tokens": 2389851.0,
      "step": 1360
    },
    {
      "entropy": 6.0609746932983395,
      "epoch": 1.0697492163009406,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989352773731928,
      "loss": 5.8998,
      "mean_token_accuracy": 0.13646639958024026,
      "num_tokens": 2397764.0,
      "step": 1365
    },
    {
      "entropy": 6.258017063140869,
      "epoch": 1.073667711598746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989058497473014,
      "loss": 6.0497,
      "mean_token_accuracy": 0.13653347492218018,
      "num_tokens": 2406688.0,
      "step": 1370
    },
    {
      "entropy": 6.219746732711792,
      "epoch": 1.0775862068965518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988760219653518,
      "loss": 6.1178,
      "mean_token_accuracy": 0.12561060190200807,
      "num_tokens": 2415475.0,
      "step": 1375
    },
    {
      "entropy": 6.158693790435791,
      "epoch": 1.0815047021943573,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988457940806513,
      "loss": 6.0725,
      "mean_token_accuracy": 0.12764996886253357,
      "num_tokens": 2425719.0,
      "step": 1380
    },
    {
      "entropy": 6.134065866470337,
      "epoch": 1.085423197492163,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988151661472218,
      "loss": 5.9679,
      "mean_token_accuracy": 0.13455401584506035,
      "num_tokens": 2435138.0,
      "step": 1385
    },
    {
      "entropy": 6.214557266235351,
      "epoch": 1.0893416927899686,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987841382198006,
      "loss": 6.1148,
      "mean_token_accuracy": 0.13025601208209991,
      "num_tokens": 2443976.0,
      "step": 1390
    },
    {
      "entropy": 6.058465433120728,
      "epoch": 1.0932601880877744,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987527103538394,
      "loss": 6.001,
      "mean_token_accuracy": 0.1352849192917347,
      "num_tokens": 2453287.0,
      "step": 1395
    },
    {
      "entropy": 6.232900381088257,
      "epoch": 1.09717868338558,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000498720882605505,
      "loss": 6.0947,
      "mean_token_accuracy": 0.1259259968996048,
      "num_tokens": 2462054.0,
      "step": 1400
    },
    {
      "entropy": 6.098091125488281,
      "epoch": 1.1010971786833856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004986886550316788,
      "loss": 5.9975,
      "mean_token_accuracy": 0.13868628442287445,
      "num_tokens": 2470333.0,
      "step": 1405
    },
    {
      "entropy": 6.151032543182373,
      "epoch": 1.1050156739811912,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004986560276899565,
      "loss": 6.0369,
      "mean_token_accuracy": 0.13234539851546287,
      "num_tokens": 2479621.0,
      "step": 1410
    },
    {
      "entropy": 6.181175947189331,
      "epoch": 1.108934169278997,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986230006386488,
      "loss": 5.9972,
      "mean_token_accuracy": 0.1349916376173496,
      "num_tokens": 2488020.0,
      "step": 1415
    },
    {
      "entropy": 6.213946294784546,
      "epoch": 1.1128526645768024,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985895739367802,
      "loss": 6.0292,
      "mean_token_accuracy": 0.13072321489453315,
      "num_tokens": 2497779.0,
      "step": 1420
    },
    {
      "entropy": 6.117242479324341,
      "epoch": 1.1167711598746082,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985557476440895,
      "loss": 5.9767,
      "mean_token_accuracy": 0.13357633352279663,
      "num_tokens": 2506791.0,
      "step": 1425
    },
    {
      "entropy": 6.191507816314697,
      "epoch": 1.1206896551724137,
      "grad_norm": 1.171875,
      "learning_rate": 0.00049852152182103,
      "loss": 6.0892,
      "mean_token_accuracy": 0.12935334220528602,
      "num_tokens": 2515576.0,
      "step": 1430
    },
    {
      "entropy": 6.180032730102539,
      "epoch": 1.1246081504702194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004984868965287686,
      "loss": 5.9955,
      "mean_token_accuracy": 0.13097495809197426,
      "num_tokens": 2523933.0,
      "step": 1435
    },
    {
      "entropy": 6.124348115921021,
      "epoch": 1.1285266457680252,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984518718291864,
      "loss": 5.9525,
      "mean_token_accuracy": 0.12772755995392798,
      "num_tokens": 2532453.0,
      "step": 1440
    },
    {
      "entropy": 6.094198179244995,
      "epoch": 1.1324451410658307,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984164477848783,
      "loss": 6.0043,
      "mean_token_accuracy": 0.1302856981754303,
      "num_tokens": 2541149.0,
      "step": 1445
    },
    {
      "entropy": 6.125786781311035,
      "epoch": 1.1363636363636362,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983806244591528,
      "loss": 5.9822,
      "mean_token_accuracy": 0.13764647617936135,
      "num_tokens": 2550302.0,
      "step": 1450
    },
    {
      "entropy": 6.0180809020996096,
      "epoch": 1.140282131661442,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983444019160318,
      "loss": 5.8351,
      "mean_token_accuracy": 0.1406431756913662,
      "num_tokens": 2558507.0,
      "step": 1455
    },
    {
      "entropy": 6.208771991729736,
      "epoch": 1.1442006269592477,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983077802202511,
      "loss": 6.1261,
      "mean_token_accuracy": 0.13089546412229539,
      "num_tokens": 2567487.0,
      "step": 1460
    },
    {
      "entropy": 6.145636320114136,
      "epoch": 1.1481191222570533,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982707594372595,
      "loss": 5.9673,
      "mean_token_accuracy": 0.13152600824832916,
      "num_tokens": 2576232.0,
      "step": 1465
    },
    {
      "entropy": 6.04912371635437,
      "epoch": 1.152037617554859,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498233339633219,
      "loss": 5.977,
      "mean_token_accuracy": 0.13370415195822716,
      "num_tokens": 2585081.0,
      "step": 1470
    },
    {
      "entropy": 6.064982080459595,
      "epoch": 1.1559561128526645,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498195520875005,
      "loss": 5.9501,
      "mean_token_accuracy": 0.13194083124399186,
      "num_tokens": 2593890.0,
      "step": 1475
    },
    {
      "entropy": 6.1344743251800535,
      "epoch": 1.1598746081504703,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981573032302059,
      "loss": 6.0369,
      "mean_token_accuracy": 0.12466074079275132,
      "num_tokens": 2602011.0,
      "step": 1480
    },
    {
      "entropy": 6.061652421951294,
      "epoch": 1.1637931034482758,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981186867671225,
      "loss": 5.9258,
      "mean_token_accuracy": 0.13856346756219864,
      "num_tokens": 2610366.0,
      "step": 1485
    },
    {
      "entropy": 6.042865514755249,
      "epoch": 1.1677115987460815,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004980796715547691,
      "loss": 5.946,
      "mean_token_accuracy": 0.13904307335615157,
      "num_tokens": 2619737.0,
      "step": 1490
    },
    {
      "entropy": 6.136406183242798,
      "epoch": 1.171630094043887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980402576628717,
      "loss": 5.9699,
      "mean_token_accuracy": 0.1325167238712311,
      "num_tokens": 2628176.0,
      "step": 1495
    },
    {
      "entropy": 6.0512768745422365,
      "epoch": 1.1755485893416928,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980004451618697,
      "loss": 5.9832,
      "mean_token_accuracy": 0.13477908223867416,
      "num_tokens": 2636195.0,
      "step": 1500
    },
    {
      "epoch": 1.1755485893416928,
      "eval_entropy": 6.076499396516371,
      "eval_loss": 6.271363258361816,
      "eval_mean_token_accuracy": 0.12794691480176393,
      "eval_num_tokens": 2636195.0,
      "eval_runtime": 2.8465,
      "eval_samples_per_second": 1448.09,
      "eval_steps_per_second": 181.275,
      "step": 1500
    },
    {
      "entropy": 6.207622623443603,
      "epoch": 1.1794670846394983,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979602341229144,
      "loss": 6.0515,
      "mean_token_accuracy": 0.1303505390882492,
      "num_tokens": 2645035.0,
      "step": 1505
    },
    {
      "entropy": 6.107669830322266,
      "epoch": 1.183385579937304,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004979196246178694,
      "loss": 6.0111,
      "mean_token_accuracy": 0.1320280283689499,
      "num_tokens": 2653239.0,
      "step": 1510
    },
    {
      "entropy": 6.064078998565674,
      "epoch": 1.1873040752351098,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004978786167193105,
      "loss": 6.0243,
      "mean_token_accuracy": 0.1286700151860714,
      "num_tokens": 2662532.0,
      "step": 1515
    },
    {
      "entropy": 6.17049560546875,
      "epoch": 1.1912225705329154,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978372105005254,
      "loss": 6.0272,
      "mean_token_accuracy": 0.13095418065786363,
      "num_tokens": 2671867.0,
      "step": 1520
    },
    {
      "entropy": 6.024587631225586,
      "epoch": 1.1951410658307209,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004977954060355139,
      "loss": 5.9538,
      "mean_token_accuracy": 0.13054575994610787,
      "num_tokens": 2680693.0,
      "step": 1525
    },
    {
      "entropy": 6.192902612686157,
      "epoch": 1.1990595611285266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977532033989871,
      "loss": 6.0462,
      "mean_token_accuracy": 0.13150823041796683,
      "num_tokens": 2689228.0,
      "step": 1530
    },
    {
      "entropy": 6.0371331691741945,
      "epoch": 1.2029780564263324,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004977106026663681,
      "loss": 5.9103,
      "mean_token_accuracy": 0.13035874292254448,
      "num_tokens": 2698483.0,
      "step": 1535
    },
    {
      "entropy": 6.142786264419556,
      "epoch": 1.206896551724138,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004976676039137914,
      "loss": 6.0491,
      "mean_token_accuracy": 0.13021435514092444,
      "num_tokens": 2706920.0,
      "step": 1540
    },
    {
      "entropy": 6.042557334899902,
      "epoch": 1.2108150470219436,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976242072181026,
      "loss": 5.9328,
      "mean_token_accuracy": 0.13973603770136833,
      "num_tokens": 2715954.0,
      "step": 1545
    },
    {
      "entropy": 6.062022352218628,
      "epoch": 1.2147335423197492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975804126568587,
      "loss": 5.9571,
      "mean_token_accuracy": 0.1364055834710598,
      "num_tokens": 2724890.0,
      "step": 1550
    },
    {
      "entropy": 6.17275595664978,
      "epoch": 1.218652037617555,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975362203083277,
      "loss": 5.9721,
      "mean_token_accuracy": 0.13264910951256753,
      "num_tokens": 2733688.0,
      "step": 1555
    },
    {
      "entropy": 6.082532548904419,
      "epoch": 1.2225705329153604,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004974916302514886,
      "loss": 5.979,
      "mean_token_accuracy": 0.13056719824671745,
      "num_tokens": 2743364.0,
      "step": 1560
    },
    {
      "entropy": 6.061155843734741,
      "epoch": 1.2264890282131662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004974466425660307,
      "loss": 6.0256,
      "mean_token_accuracy": 0.13040423914790153,
      "num_tokens": 2754276.0,
      "step": 1565
    },
    {
      "entropy": 6.098849248886109,
      "epoch": 1.2304075235109717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974012573323545,
      "loss": 5.8312,
      "mean_token_accuracy": 0.1432400055229664,
      "num_tokens": 2762680.0,
      "step": 1570
    },
    {
      "entropy": 5.957777738571167,
      "epoch": 1.2343260188087775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973554746315709,
      "loss": 5.9289,
      "mean_token_accuracy": 0.14215950816869735,
      "num_tokens": 2771161.0,
      "step": 1575
    },
    {
      "entropy": 6.126145744323731,
      "epoch": 1.238244514106583,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973092945455009,
      "loss": 5.9042,
      "mean_token_accuracy": 0.141022390127182,
      "num_tokens": 2779321.0,
      "step": 1580
    },
    {
      "entropy": 5.973110008239746,
      "epoch": 1.2421630094043887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004972627171566757,
      "loss": 6.0625,
      "mean_token_accuracy": 0.13462768569588662,
      "num_tokens": 2788425.0,
      "step": 1585
    },
    {
      "entropy": 6.077723455429077,
      "epoch": 1.2460815047021945,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972157425483368,
      "loss": 5.9129,
      "mean_token_accuracy": 0.13926904946565627,
      "num_tokens": 2796747.0,
      "step": 1590
    },
    {
      "entropy": 5.98823504447937,
      "epoch": 1.25,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004971683708044353,
      "loss": 5.9614,
      "mean_token_accuracy": 0.1371827580034733,
      "num_tokens": 2805335.0,
      "step": 1595
    },
    {
      "entropy": 6.1075457572937015,
      "epoch": 1.2539184952978055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971206020096323,
      "loss": 5.8454,
      "mean_token_accuracy": 0.14424244835972785,
      "num_tokens": 2813350.0,
      "step": 1600
    },
    {
      "entropy": 6.131724071502686,
      "epoch": 1.2578369905956113,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970724362492984,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1399139069020748,
      "num_tokens": 2821922.0,
      "step": 1605
    },
    {
      "entropy": 6.015088748931885,
      "epoch": 1.261755485893417,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004970238736095135,
      "loss": 5.952,
      "mean_token_accuracy": 0.13358094021677971,
      "num_tokens": 2830832.0,
      "step": 1610
    },
    {
      "entropy": 6.085702276229858,
      "epoch": 1.2656739811912225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969749141770671,
      "loss": 5.9374,
      "mean_token_accuracy": 0.13813378065824508,
      "num_tokens": 2839699.0,
      "step": 1615
    },
    {
      "entropy": 6.056669616699219,
      "epoch": 1.2695924764890283,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969255580394575,
      "loss": 6.0123,
      "mean_token_accuracy": 0.13840878382325172,
      "num_tokens": 2848293.0,
      "step": 1620
    },
    {
      "entropy": 6.114861154556275,
      "epoch": 1.2735109717868338,
      "grad_norm": 1.171875,
      "learning_rate": 0.000496875805284892,
      "loss": 5.9607,
      "mean_token_accuracy": 0.13838078081607819,
      "num_tokens": 2856468.0,
      "step": 1625
    },
    {
      "entropy": 6.044304227828979,
      "epoch": 1.2774294670846396,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968256560022871,
      "loss": 5.9018,
      "mean_token_accuracy": 0.13837780505418779,
      "num_tokens": 2865366.0,
      "step": 1630
    },
    {
      "entropy": 5.9802772998809814,
      "epoch": 1.281347962382445,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004967751102812676,
      "loss": 5.9539,
      "mean_token_accuracy": 0.14084723740816116,
      "num_tokens": 2873520.0,
      "step": 1635
    },
    {
      "entropy": 6.120439767837524,
      "epoch": 1.2852664576802508,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004967241682121669,
      "loss": 6.0169,
      "mean_token_accuracy": 0.13527923449873924,
      "num_tokens": 2883324.0,
      "step": 1640
    },
    {
      "entropy": 6.048051452636718,
      "epoch": 1.2891849529780564,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004966728298860267,
      "loss": 5.948,
      "mean_token_accuracy": 0.13382584750652313,
      "num_tokens": 2892200.0,
      "step": 1645
    },
    {
      "entropy": 6.069919157028198,
      "epoch": 1.293103448275862,
      "grad_norm": 1.25,
      "learning_rate": 0.0004966210953945969,
      "loss": 5.9192,
      "mean_token_accuracy": 0.13777885362505912,
      "num_tokens": 2900597.0,
      "step": 1650
    },
    {
      "entropy": 6.0597516059875485,
      "epoch": 1.2970219435736676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965689648303355,
      "loss": 5.9559,
      "mean_token_accuracy": 0.13197760209441184,
      "num_tokens": 2909869.0,
      "step": 1655
    },
    {
      "entropy": 5.956328582763672,
      "epoch": 1.3009404388714734,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004965164382864083,
      "loss": 5.9761,
      "mean_token_accuracy": 0.13753246441483497,
      "num_tokens": 2919166.0,
      "step": 1660
    },
    {
      "entropy": 6.117870473861695,
      "epoch": 1.3048589341692791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004964635158566886,
      "loss": 5.9859,
      "mean_token_accuracy": 0.13755642771720886,
      "num_tokens": 2927884.0,
      "step": 1665
    },
    {
      "entropy": 6.016185617446899,
      "epoch": 1.3087774294670846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004964101976357574,
      "loss": 5.915,
      "mean_token_accuracy": 0.14404550418257714,
      "num_tokens": 2936737.0,
      "step": 1670
    },
    {
      "entropy": 6.071380281448365,
      "epoch": 1.3126959247648902,
      "grad_norm": 1.265625,
      "learning_rate": 0.000496356483718903,
      "loss": 6.0118,
      "mean_token_accuracy": 0.13691077008843422,
      "num_tokens": 2944203.0,
      "step": 1675
    },
    {
      "entropy": 6.0966150760650635,
      "epoch": 1.316614420062696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004963023742021211,
      "loss": 6.025,
      "mean_token_accuracy": 0.13278019577264785,
      "num_tokens": 2952639.0,
      "step": 1680
    },
    {
      "entropy": 6.0501056671142575,
      "epoch": 1.3205329153605017,
      "grad_norm": 1.125,
      "learning_rate": 0.000496247869182114,
      "loss": 5.9512,
      "mean_token_accuracy": 0.13901304081082344,
      "num_tokens": 2961482.0,
      "step": 1685
    },
    {
      "entropy": 6.046927785873413,
      "epoch": 1.3244514106583072,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004961929687562909,
      "loss": 5.9634,
      "mean_token_accuracy": 0.13337364569306373,
      "num_tokens": 2970927.0,
      "step": 1690
    },
    {
      "entropy": 5.964669847488404,
      "epoch": 1.328369905956113,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961376730227681,
      "loss": 5.9123,
      "mean_token_accuracy": 0.1409181497991085,
      "num_tokens": 2979329.0,
      "step": 1695
    },
    {
      "entropy": 6.0310773849487305,
      "epoch": 1.3322884012539185,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004960819820803675,
      "loss": 5.8666,
      "mean_token_accuracy": 0.14356955140829086,
      "num_tokens": 2986894.0,
      "step": 1700
    },
    {
      "entropy": 6.004372310638428,
      "epoch": 1.3362068965517242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004960258960286185,
      "loss": 5.8687,
      "mean_token_accuracy": 0.139854796230793,
      "num_tokens": 2994721.0,
      "step": 1705
    },
    {
      "entropy": 6.086730909347534,
      "epoch": 1.3401253918495297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004959694149677553,
      "loss": 6.0317,
      "mean_token_accuracy": 0.12823668047785758,
      "num_tokens": 3003165.0,
      "step": 1710
    },
    {
      "entropy": 6.0770214080810545,
      "epoch": 1.3440438871473355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004959125389987193,
      "loss": 6.008,
      "mean_token_accuracy": 0.13203698322176932,
      "num_tokens": 3012178.0,
      "step": 1715
    },
    {
      "entropy": 6.01055006980896,
      "epoch": 1.347962382445141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004958552682231567,
      "loss": 5.9027,
      "mean_token_accuracy": 0.14051611423492433,
      "num_tokens": 3021685.0,
      "step": 1720
    },
    {
      "entropy": 6.005949354171753,
      "epoch": 1.3518808777429467,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004957976027434199,
      "loss": 5.956,
      "mean_token_accuracy": 0.12991197034716606,
      "num_tokens": 3030667.0,
      "step": 1725
    },
    {
      "entropy": 6.097392654418945,
      "epoch": 1.3557993730407523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004957395426625663,
      "loss": 6.0049,
      "mean_token_accuracy": 0.1332964815199375,
      "num_tokens": 3039501.0,
      "step": 1730
    },
    {
      "entropy": 5.981141185760498,
      "epoch": 1.359717868338558,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004956810880843587,
      "loss": 5.9382,
      "mean_token_accuracy": 0.13364407643675805,
      "num_tokens": 3048547.0,
      "step": 1735
    },
    {
      "entropy": 6.120581865310669,
      "epoch": 1.3636363636363638,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495622239113265,
      "loss": 5.9044,
      "mean_token_accuracy": 0.14041278436779975,
      "num_tokens": 3056521.0,
      "step": 1740
    },
    {
      "entropy": 5.939314699172973,
      "epoch": 1.3675548589341693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004955629958544577,
      "loss": 5.8949,
      "mean_token_accuracy": 0.13747348114848137,
      "num_tokens": 3066007.0,
      "step": 1745
    },
    {
      "entropy": 6.10064115524292,
      "epoch": 1.3714733542319748,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004955033584138143,
      "loss": 6.0698,
      "mean_token_accuracy": 0.12523617148399352,
      "num_tokens": 3075111.0,
      "step": 1750
    },
    {
      "entropy": 6.022913122177124,
      "epoch": 1.3753918495297806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004954433268979164,
      "loss": 5.8818,
      "mean_token_accuracy": 0.14991883113980292,
      "num_tokens": 3083766.0,
      "step": 1755
    },
    {
      "entropy": 5.973852968215942,
      "epoch": 1.3793103448275863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953829014140502,
      "loss": 5.9353,
      "mean_token_accuracy": 0.13615802749991418,
      "num_tokens": 3092103.0,
      "step": 1760
    },
    {
      "entropy": 5.990107297897339,
      "epoch": 1.3832288401253918,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004953220820702057,
      "loss": 5.9593,
      "mean_token_accuracy": 0.13366186618804932,
      "num_tokens": 3101286.0,
      "step": 1765
    },
    {
      "entropy": 6.151833629608154,
      "epoch": 1.3871473354231976,
      "grad_norm": 1.125,
      "learning_rate": 0.0004952608689750771,
      "loss": 5.9611,
      "mean_token_accuracy": 0.13463746160268783,
      "num_tokens": 3110554.0,
      "step": 1770
    },
    {
      "entropy": 6.00332202911377,
      "epoch": 1.391065830721003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004951992622380619,
      "loss": 5.8163,
      "mean_token_accuracy": 0.14524587914347648,
      "num_tokens": 3119265.0,
      "step": 1775
    },
    {
      "entropy": 5.971736478805542,
      "epoch": 1.3949843260188088,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951372619692615,
      "loss": 5.8515,
      "mean_token_accuracy": 0.14273120015859603,
      "num_tokens": 3127962.0,
      "step": 1780
    },
    {
      "entropy": 6.01099796295166,
      "epoch": 1.3989028213166144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950748682794804,
      "loss": 5.9917,
      "mean_token_accuracy": 0.13492952212691306,
      "num_tokens": 3137265.0,
      "step": 1785
    },
    {
      "entropy": 5.9283558368682865,
      "epoch": 1.40282131661442,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004950120812802262,
      "loss": 5.8128,
      "mean_token_accuracy": 0.1431027092039585,
      "num_tokens": 3146458.0,
      "step": 1790
    },
    {
      "entropy": 6.07517991065979,
      "epoch": 1.4067398119122256,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004949489010837095,
      "loss": 5.9588,
      "mean_token_accuracy": 0.13590797632932664,
      "num_tokens": 3155557.0,
      "step": 1795
    },
    {
      "entropy": 5.9258420944213865,
      "epoch": 1.4106583072100314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004948853278028436,
      "loss": 5.8368,
      "mean_token_accuracy": 0.14588673859834672,
      "num_tokens": 3163478.0,
      "step": 1800
    },
    {
      "entropy": 5.979496049880981,
      "epoch": 1.414576802507837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000494821361551244,
      "loss": 5.9323,
      "mean_token_accuracy": 0.1362486593425274,
      "num_tokens": 3172632.0,
      "step": 1805
    },
    {
      "entropy": 6.013187265396118,
      "epoch": 1.4184952978056427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004947570024432291,
      "loss": 5.9629,
      "mean_token_accuracy": 0.14079801365733147,
      "num_tokens": 3181325.0,
      "step": 1810
    },
    {
      "entropy": 6.030901575088501,
      "epoch": 1.4224137931034484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004946922505938189,
      "loss": 6.0188,
      "mean_token_accuracy": 0.14003786072134972,
      "num_tokens": 3189783.0,
      "step": 1815
    },
    {
      "entropy": 6.039680242538452,
      "epoch": 1.426332288401254,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004946271061187354,
      "loss": 5.8608,
      "mean_token_accuracy": 0.13449943587183952,
      "num_tokens": 3198605.0,
      "step": 1820
    },
    {
      "entropy": 5.975720500946045,
      "epoch": 1.4302507836990594,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004945615691344025,
      "loss": 5.9606,
      "mean_token_accuracy": 0.12840349078178406,
      "num_tokens": 3208061.0,
      "step": 1825
    },
    {
      "entropy": 5.996256589889526,
      "epoch": 1.4341692789968652,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004944956397579453,
      "loss": 5.865,
      "mean_token_accuracy": 0.14466958418488501,
      "num_tokens": 3216093.0,
      "step": 1830
    },
    {
      "entropy": 5.9624964714050295,
      "epoch": 1.438087774294671,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004944293181071902,
      "loss": 5.8584,
      "mean_token_accuracy": 0.14238173440098761,
      "num_tokens": 3224446.0,
      "step": 1835
    },
    {
      "entropy": 6.023101758956909,
      "epoch": 1.4420062695924765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004943626043006649,
      "loss": 5.9836,
      "mean_token_accuracy": 0.1356622613966465,
      "num_tokens": 3233398.0,
      "step": 1840
    },
    {
      "entropy": 6.034554052352905,
      "epoch": 1.4459247648902822,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000494295498457598,
      "loss": 5.9682,
      "mean_token_accuracy": 0.13090459257364273,
      "num_tokens": 3243391.0,
      "step": 1845
    },
    {
      "entropy": 5.992100238800049,
      "epoch": 1.4498432601880877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004942280006979179,
      "loss": 5.9107,
      "mean_token_accuracy": 0.1431814417243004,
      "num_tokens": 3251954.0,
      "step": 1850
    },
    {
      "entropy": 6.005795526504516,
      "epoch": 1.4537617554858935,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941601111422546,
      "loss": 5.9356,
      "mean_token_accuracy": 0.13904605209827423,
      "num_tokens": 3261121.0,
      "step": 1855
    },
    {
      "entropy": 5.875429582595825,
      "epoch": 1.457680250783699,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004940918299119375,
      "loss": 5.851,
      "mean_token_accuracy": 0.14492034763097764,
      "num_tokens": 3270179.0,
      "step": 1860
    },
    {
      "entropy": 5.878690958023071,
      "epoch": 1.4615987460815048,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004940231571289962,
      "loss": 5.8587,
      "mean_token_accuracy": 0.13866079300642015,
      "num_tokens": 3279749.0,
      "step": 1865
    },
    {
      "entropy": 6.03740701675415,
      "epoch": 1.4655172413793103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004939540929161603,
      "loss": 5.8712,
      "mean_token_accuracy": 0.14061601012945174,
      "num_tokens": 3288458.0,
      "step": 1870
    },
    {
      "entropy": 5.949235296249389,
      "epoch": 1.469435736677116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004938846373968586,
      "loss": 5.8497,
      "mean_token_accuracy": 0.14182733744382858,
      "num_tokens": 3297231.0,
      "step": 1875
    },
    {
      "entropy": 6.013594055175782,
      "epoch": 1.4733542319749215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004938147906952194,
      "loss": 6.0087,
      "mean_token_accuracy": 0.1348317749798298,
      "num_tokens": 3306915.0,
      "step": 1880
    },
    {
      "entropy": 5.975127410888672,
      "epoch": 1.4772727272727273,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049374455293607,
      "loss": 5.9349,
      "mean_token_accuracy": 0.13975367248058318,
      "num_tokens": 3315985.0,
      "step": 1885
    },
    {
      "entropy": 6.011495923995971,
      "epoch": 1.481191222570533,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004936739242449369,
      "loss": 5.9395,
      "mean_token_accuracy": 0.13877131789922714,
      "num_tokens": 3324924.0,
      "step": 1890
    },
    {
      "entropy": 5.972328424453735,
      "epoch": 1.4851097178683386,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004936029047480447,
      "loss": 5.923,
      "mean_token_accuracy": 0.1383238084614277,
      "num_tokens": 3333464.0,
      "step": 1895
    },
    {
      "entropy": 6.008746147155762,
      "epoch": 1.489028213166144,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004935314945723171,
      "loss": 5.8358,
      "mean_token_accuracy": 0.14501626715064048,
      "num_tokens": 3342441.0,
      "step": 1900
    },
    {
      "entropy": 5.930044031143188,
      "epoch": 1.4929467084639498,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004934596938453754,
      "loss": 5.8856,
      "mean_token_accuracy": 0.1392125241458416,
      "num_tokens": 3351338.0,
      "step": 1905
    },
    {
      "entropy": 5.987050580978393,
      "epoch": 1.4968652037617556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004933875026955391,
      "loss": 5.8414,
      "mean_token_accuracy": 0.13805665150284768,
      "num_tokens": 3360356.0,
      "step": 1910
    },
    {
      "entropy": 6.016159772872925,
      "epoch": 1.500783699059561,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004933149212518258,
      "loss": 5.933,
      "mean_token_accuracy": 0.13254097029566764,
      "num_tokens": 3369485.0,
      "step": 1915
    },
    {
      "entropy": 5.981318569183349,
      "epoch": 1.5047021943573666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004932419496439501,
      "loss": 5.884,
      "mean_token_accuracy": 0.14330490529537201,
      "num_tokens": 3378355.0,
      "step": 1920
    },
    {
      "entropy": 5.914515161514283,
      "epoch": 1.5086206896551724,
      "grad_norm": 1.140625,
      "learning_rate": 0.000493168588002324,
      "loss": 5.8032,
      "mean_token_accuracy": 0.14702262133359909,
      "num_tokens": 3386729.0,
      "step": 1925
    },
    {
      "entropy": 6.070053672790527,
      "epoch": 1.5125391849529781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004930948364580569,
      "loss": 6.0245,
      "mean_token_accuracy": 0.13127490803599356,
      "num_tokens": 3395822.0,
      "step": 1930
    },
    {
      "entropy": 5.980561304092407,
      "epoch": 1.5164576802507836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004930206951429546,
      "loss": 5.9416,
      "mean_token_accuracy": 0.13797413259744645,
      "num_tokens": 3404671.0,
      "step": 1935
    },
    {
      "entropy": 5.985497379302979,
      "epoch": 1.5203761755485894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004929461641895197,
      "loss": 5.8679,
      "mean_token_accuracy": 0.14024683982133865,
      "num_tokens": 3413289.0,
      "step": 1940
    },
    {
      "entropy": 5.936700534820557,
      "epoch": 1.5242946708463951,
      "grad_norm": 1.1875,
      "learning_rate": 0.000492871243730951,
      "loss": 5.917,
      "mean_token_accuracy": 0.13965860977768899,
      "num_tokens": 3421960.0,
      "step": 1945
    },
    {
      "entropy": 5.997755765914917,
      "epoch": 1.5282131661442007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004927959339011437,
      "loss": 5.9149,
      "mean_token_accuracy": 0.1362414240837097,
      "num_tokens": 3431958.0,
      "step": 1950
    },
    {
      "entropy": 5.822151041030883,
      "epoch": 1.5321316614420062,
      "grad_norm": 1.125,
      "learning_rate": 0.0004927202348346885,
      "loss": 5.7435,
      "mean_token_accuracy": 0.1449103720486164,
      "num_tokens": 3441438.0,
      "step": 1955
    },
    {
      "entropy": 6.006295967102051,
      "epoch": 1.536050156739812,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492644146666872,
      "loss": 5.8473,
      "mean_token_accuracy": 0.1476273812353611,
      "num_tokens": 3450176.0,
      "step": 1960
    },
    {
      "entropy": 5.828231477737427,
      "epoch": 1.5399686520376177,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004925676695336761,
      "loss": 5.8462,
      "mean_token_accuracy": 0.14223939031362534,
      "num_tokens": 3458685.0,
      "step": 1965
    },
    {
      "entropy": 6.036013793945313,
      "epoch": 1.5438871473354232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004924908035717777,
      "loss": 5.9932,
      "mean_token_accuracy": 0.1358640395104885,
      "num_tokens": 3468362.0,
      "step": 1970
    },
    {
      "entropy": 5.99300446510315,
      "epoch": 1.5478056426332287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000492413548918549,
      "loss": 5.9158,
      "mean_token_accuracy": 0.13908401131629944,
      "num_tokens": 3476791.0,
      "step": 1975
    },
    {
      "entropy": 5.858194875717163,
      "epoch": 1.5517241379310345,
      "grad_norm": 1.25,
      "learning_rate": 0.0004923359057120563,
      "loss": 5.8306,
      "mean_token_accuracy": 0.1353352040052414,
      "num_tokens": 3486486.0,
      "step": 1980
    },
    {
      "entropy": 5.9135054588317875,
      "epoch": 1.5556426332288402,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004922578740910608,
      "loss": 5.7897,
      "mean_token_accuracy": 0.147303768992424,
      "num_tokens": 3496145.0,
      "step": 1985
    },
    {
      "entropy": 5.895607662200928,
      "epoch": 1.5595611285266457,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004921794541950177,
      "loss": 5.7769,
      "mean_token_accuracy": 0.1445461630821228,
      "num_tokens": 3504644.0,
      "step": 1990
    },
    {
      "entropy": 5.8936514377594,
      "epoch": 1.5634796238244513,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004921006461640758,
      "loss": 5.8104,
      "mean_token_accuracy": 0.13581008464097977,
      "num_tokens": 3513126.0,
      "step": 1995
    },
    {
      "entropy": 5.810939693450928,
      "epoch": 1.567398119122257,
      "grad_norm": 1.21875,
      "learning_rate": 0.000492021450139078,
      "loss": 5.749,
      "mean_token_accuracy": 0.14536509588360785,
      "num_tokens": 3521505.0,
      "step": 2000
    },
    {
      "epoch": 1.567398119122257,
      "eval_entropy": 5.800363567448402,
      "eval_loss": 6.0877366065979,
      "eval_mean_token_accuracy": 0.13831741765661296,
      "eval_num_tokens": 3521505.0,
      "eval_runtime": 2.8597,
      "eval_samples_per_second": 1441.409,
      "eval_steps_per_second": 180.438,
      "step": 2000
    },
    {
      "entropy": 6.01237120628357,
      "epoch": 1.5713166144200628,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004919418662615605,
      "loss": 5.9987,
      "mean_token_accuracy": 0.13824992179870604,
      "num_tokens": 3530008.0,
      "step": 2005
    },
    {
      "entropy": 6.025429964065552,
      "epoch": 1.5752351097178683,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004918618946737525,
      "loss": 5.8827,
      "mean_token_accuracy": 0.1387758158147335,
      "num_tokens": 3538778.0,
      "step": 2010
    },
    {
      "entropy": 5.967436027526856,
      "epoch": 1.579153605015674,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917815355185762,
      "loss": 5.8638,
      "mean_token_accuracy": 0.13804899901151657,
      "num_tokens": 3547108.0,
      "step": 2015
    },
    {
      "entropy": 5.896082782745362,
      "epoch": 1.5830721003134798,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004917007889396464,
      "loss": 5.9114,
      "mean_token_accuracy": 0.13484615311026574,
      "num_tokens": 3556094.0,
      "step": 2020
    },
    {
      "entropy": 5.933627510070801,
      "epoch": 1.5869905956112853,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004916196550812706,
      "loss": 5.7628,
      "mean_token_accuracy": 0.14669432789087294,
      "num_tokens": 3564222.0,
      "step": 2025
    },
    {
      "entropy": 5.776369333267212,
      "epoch": 1.5909090909090908,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004915381340884477,
      "loss": 5.773,
      "mean_token_accuracy": 0.14765530303120614,
      "num_tokens": 3573123.0,
      "step": 2030
    },
    {
      "entropy": 5.964944887161255,
      "epoch": 1.5948275862068966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004914562261068693,
      "loss": 5.8091,
      "mean_token_accuracy": 0.14593618959188462,
      "num_tokens": 3581462.0,
      "step": 2035
    },
    {
      "entropy": 5.8525710105896,
      "epoch": 1.5987460815047023,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004913739312829181,
      "loss": 5.8172,
      "mean_token_accuracy": 0.14506494328379632,
      "num_tokens": 3590718.0,
      "step": 2040
    },
    {
      "entropy": 5.894586277008057,
      "epoch": 1.6026645768025078,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912912497636683,
      "loss": 5.8484,
      "mean_token_accuracy": 0.13913847804069518,
      "num_tokens": 3599890.0,
      "step": 2045
    },
    {
      "entropy": 6.029708480834961,
      "epoch": 1.6065830721003134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004912081816968853,
      "loss": 5.9327,
      "mean_token_accuracy": 0.13791020438075066,
      "num_tokens": 3607927.0,
      "step": 2050
    },
    {
      "entropy": 5.966132402420044,
      "epoch": 1.6105015673981191,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000491124727231025,
      "loss": 5.7923,
      "mean_token_accuracy": 0.147890043258667,
      "num_tokens": 3615935.0,
      "step": 2055
    },
    {
      "entropy": 5.831100559234619,
      "epoch": 1.6144200626959249,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004910408865152343,
      "loss": 5.8251,
      "mean_token_accuracy": 0.13951457142829896,
      "num_tokens": 3624475.0,
      "step": 2060
    },
    {
      "entropy": 6.039763879776001,
      "epoch": 1.6183385579937304,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004909566596993498,
      "loss": 6.0219,
      "mean_token_accuracy": 0.1320653535425663,
      "num_tokens": 3633995.0,
      "step": 2065
    },
    {
      "entropy": 5.985185670852661,
      "epoch": 1.622257053291536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004908720469338988,
      "loss": 5.8851,
      "mean_token_accuracy": 0.14443439915776252,
      "num_tokens": 3643121.0,
      "step": 2070
    },
    {
      "entropy": 5.913325405120849,
      "epoch": 1.6261755485893417,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004907870483700979,
      "loss": 5.8127,
      "mean_token_accuracy": 0.14168807715177537,
      "num_tokens": 3652285.0,
      "step": 2075
    },
    {
      "entropy": 5.871179533004761,
      "epoch": 1.6300940438871474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004907016641598534,
      "loss": 5.8916,
      "mean_token_accuracy": 0.13586366698145866,
      "num_tokens": 3661441.0,
      "step": 2080
    },
    {
      "entropy": 5.870393753051758,
      "epoch": 1.634012539184953,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004906158944557607,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1525380626320839,
      "num_tokens": 3669364.0,
      "step": 2085
    },
    {
      "entropy": 5.979267978668213,
      "epoch": 1.6379310344827587,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000490529739411104,
      "loss": 5.8639,
      "mean_token_accuracy": 0.14333550184965133,
      "num_tokens": 3677848.0,
      "step": 2090
    },
    {
      "entropy": 5.851955509185791,
      "epoch": 1.6418495297805644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004904431991798565,
      "loss": 5.7564,
      "mean_token_accuracy": 0.1478363499045372,
      "num_tokens": 3686617.0,
      "step": 2095
    },
    {
      "entropy": 5.895144653320313,
      "epoch": 1.64576802507837,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004903562739166797,
      "loss": 5.7719,
      "mean_token_accuracy": 0.14103165566921233,
      "num_tokens": 3695106.0,
      "step": 2100
    },
    {
      "entropy": 5.809584045410157,
      "epoch": 1.6496865203761755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004902689637769229,
      "loss": 5.7045,
      "mean_token_accuracy": 0.1489320859313011,
      "num_tokens": 3703167.0,
      "step": 2105
    },
    {
      "entropy": 5.91784520149231,
      "epoch": 1.6536050156739812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004901812689166237,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1423416718840599,
      "num_tokens": 3711597.0,
      "step": 2110
    },
    {
      "entropy": 5.876578712463379,
      "epoch": 1.657523510971787,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004900931894925069,
      "loss": 5.9171,
      "mean_token_accuracy": 0.1454581469297409,
      "num_tokens": 3720277.0,
      "step": 2115
    },
    {
      "entropy": 5.875493669509888,
      "epoch": 1.6614420062695925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004900047256619849,
      "loss": 5.8751,
      "mean_token_accuracy": 0.14267786517739295,
      "num_tokens": 3729832.0,
      "step": 2120
    },
    {
      "entropy": 5.92410454750061,
      "epoch": 1.665360501567398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004899158775831566,
      "loss": 5.8171,
      "mean_token_accuracy": 0.1460487052798271,
      "num_tokens": 3738390.0,
      "step": 2125
    },
    {
      "entropy": 5.860368490219116,
      "epoch": 1.6692789968652038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004898266454148081,
      "loss": 5.8842,
      "mean_token_accuracy": 0.14248219579458238,
      "num_tokens": 3747046.0,
      "step": 2130
    },
    {
      "entropy": 5.8945173740386965,
      "epoch": 1.6731974921630095,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004897370293164119,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1434356912970543,
      "num_tokens": 3755818.0,
      "step": 2135
    },
    {
      "entropy": 6.046916913986206,
      "epoch": 1.677115987460815,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004896470294481262,
      "loss": 5.9463,
      "mean_token_accuracy": 0.14069818705320358,
      "num_tokens": 3764123.0,
      "step": 2140
    },
    {
      "entropy": 5.899928379058838,
      "epoch": 1.6810344827586206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004895566459707954,
      "loss": 6.0044,
      "mean_token_accuracy": 0.13193708062171935,
      "num_tokens": 3773105.0,
      "step": 2145
    },
    {
      "entropy": 5.964847612380981,
      "epoch": 1.6849529780564263,
      "grad_norm": 1.25,
      "learning_rate": 0.0004894658790459498,
      "loss": 5.8433,
      "mean_token_accuracy": 0.1431967169046402,
      "num_tokens": 3781417.0,
      "step": 2150
    },
    {
      "entropy": 6.018510866165161,
      "epoch": 1.688871473354232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004893747288358041,
      "loss": 5.9208,
      "mean_token_accuracy": 0.1428132861852646,
      "num_tokens": 3789657.0,
      "step": 2155
    },
    {
      "entropy": 5.967458343505859,
      "epoch": 1.6927899686520376,
      "grad_norm": 1.140625,
      "learning_rate": 0.000489283195503259,
      "loss": 5.9488,
      "mean_token_accuracy": 0.13716802671551703,
      "num_tokens": 3798700.0,
      "step": 2160
    },
    {
      "entropy": 5.913800382614136,
      "epoch": 1.6967084639498433,
      "grad_norm": 1.15625,
      "learning_rate": 0.000489191279211899,
      "loss": 5.7975,
      "mean_token_accuracy": 0.14700483977794648,
      "num_tokens": 3807299.0,
      "step": 2165
    },
    {
      "entropy": 5.914860534667969,
      "epoch": 1.700626959247649,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004890989801259935,
      "loss": 5.8909,
      "mean_token_accuracy": 0.14413824826478958,
      "num_tokens": 3816292.0,
      "step": 2170
    },
    {
      "entropy": 5.9630790710449215,
      "epoch": 1.7045454545454546,
      "grad_norm": 1.140625,
      "learning_rate": 0.000489006298410496,
      "loss": 5.805,
      "mean_token_accuracy": 0.140317203104496,
      "num_tokens": 3825070.0,
      "step": 2175
    },
    {
      "entropy": 5.899825859069824,
      "epoch": 1.70846394984326,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004889132342310438,
      "loss": 5.7943,
      "mean_token_accuracy": 0.14539300575852393,
      "num_tokens": 3834016.0,
      "step": 2180
    },
    {
      "entropy": 5.92696852684021,
      "epoch": 1.7123824451410659,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004888197877539577,
      "loss": 5.8552,
      "mean_token_accuracy": 0.14206420630216599,
      "num_tokens": 3842172.0,
      "step": 2185
    },
    {
      "entropy": 5.870498418807983,
      "epoch": 1.7163009404388716,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004887259591462417,
      "loss": 5.7546,
      "mean_token_accuracy": 0.14559593349695205,
      "num_tokens": 3851109.0,
      "step": 2190
    },
    {
      "entropy": 5.859376382827759,
      "epoch": 1.7202194357366771,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004886317485755825,
      "loss": 5.8524,
      "mean_token_accuracy": 0.14535145387053489,
      "num_tokens": 3859656.0,
      "step": 2195
    },
    {
      "entropy": 5.879422616958618,
      "epoch": 1.7241379310344827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004885371562103498,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1510668709874153,
      "num_tokens": 3868698.0,
      "step": 2200
    },
    {
      "entropy": 5.836897468566894,
      "epoch": 1.7280564263322884,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004884421822195957,
      "loss": 5.8164,
      "mean_token_accuracy": 0.14483954310417174,
      "num_tokens": 3877474.0,
      "step": 2205
    },
    {
      "entropy": 5.9123077392578125,
      "epoch": 1.7319749216300941,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004883468267730538,
      "loss": 5.7831,
      "mean_token_accuracy": 0.1473929390311241,
      "num_tokens": 3886328.0,
      "step": 2210
    },
    {
      "entropy": 5.849182271957398,
      "epoch": 1.7358934169278997,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048825109004114006,
      "loss": 5.7698,
      "mean_token_accuracy": 0.1499294877052307,
      "num_tokens": 3894424.0,
      "step": 2215
    },
    {
      "entropy": 5.945364856719971,
      "epoch": 1.7398119122257052,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004881549721949513,
      "loss": 5.8649,
      "mean_token_accuracy": 0.1370491735637188,
      "num_tokens": 3903746.0,
      "step": 2220
    },
    {
      "entropy": 5.7993536472320555,
      "epoch": 1.743730407523511,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004880584734062658,
      "loss": 5.8347,
      "mean_token_accuracy": 0.14640080332756042,
      "num_tokens": 3912033.0,
      "step": 2225
    },
    {
      "entropy": 5.84619402885437,
      "epoch": 1.7476489028213167,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004879615938475425,
      "loss": 5.7356,
      "mean_token_accuracy": 0.14485909938812255,
      "num_tokens": 3920388.0,
      "step": 2230
    },
    {
      "entropy": 5.835551357269287,
      "epoch": 1.7515673981191222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004878643336919209,
      "loss": 5.9415,
      "mean_token_accuracy": 0.1401818074285984,
      "num_tokens": 3929926.0,
      "step": 2235
    },
    {
      "entropy": 6.145181274414062,
      "epoch": 1.7554858934169277,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004877666931132206,
      "loss": 5.9229,
      "mean_token_accuracy": 0.1354038953781128,
      "num_tokens": 3938696.0,
      "step": 2240
    },
    {
      "entropy": 5.85701961517334,
      "epoch": 1.7594043887147337,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004876686722859413,
      "loss": 5.7488,
      "mean_token_accuracy": 0.15484919100999833,
      "num_tokens": 3947537.0,
      "step": 2245
    },
    {
      "entropy": 5.902706909179687,
      "epoch": 1.7633228840125392,
      "grad_norm": 1.171875,
      "learning_rate": 0.000487570271385262,
      "loss": 5.7587,
      "mean_token_accuracy": 0.14329107031226157,
      "num_tokens": 3956539.0,
      "step": 2250
    },
    {
      "entropy": 5.800901651382446,
      "epoch": 1.7672413793103448,
      "grad_norm": 1.125,
      "learning_rate": 0.0004874714905870411,
      "loss": 5.729,
      "mean_token_accuracy": 0.14343830123543738,
      "num_tokens": 3964927.0,
      "step": 2255
    },
    {
      "entropy": 5.866149950027466,
      "epoch": 1.7711598746081505,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004873723300678159,
      "loss": 5.8463,
      "mean_token_accuracy": 0.1442209616303444,
      "num_tokens": 3973565.0,
      "step": 2260
    },
    {
      "entropy": 5.898542833328247,
      "epoch": 1.7750783699059562,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048727279000480226,
      "loss": 5.893,
      "mean_token_accuracy": 0.1396360144019127,
      "num_tokens": 3982683.0,
      "step": 2265
    },
    {
      "entropy": 5.953893518447876,
      "epoch": 1.7789968652037618,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048717287057589454,
      "loss": 5.7974,
      "mean_token_accuracy": 0.14212608486413955,
      "num_tokens": 3991462.0,
      "step": 2270
    },
    {
      "entropy": 5.688703346252441,
      "epoch": 1.7829153605015673,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004870725719596648,
      "loss": 5.5537,
      "mean_token_accuracy": 0.1619800105690956,
      "num_tokens": 3999821.0,
      "step": 2275
    },
    {
      "entropy": 5.895339012145996,
      "epoch": 1.786833855799373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004869718943353631,
      "loss": 5.8178,
      "mean_token_accuracy": 0.14287081807851792,
      "num_tokens": 4008655.0,
      "step": 2280
    },
    {
      "entropy": 5.915091705322266,
      "epoch": 1.7907523510971788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048687083788291656,
      "loss": 5.8518,
      "mean_token_accuracy": 0.14822934269905091,
      "num_tokens": 4017317.0,
      "step": 2285
    },
    {
      "entropy": 5.805630207061768,
      "epoch": 1.7946708463949843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048676940278292953,
      "loss": 5.6807,
      "mean_token_accuracy": 0.1457840844988823,
      "num_tokens": 4025407.0,
      "step": 2290
    },
    {
      "entropy": 5.82935643196106,
      "epoch": 1.7985893416927898,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048666758921668286,
      "loss": 5.7005,
      "mean_token_accuracy": 0.1469753384590149,
      "num_tokens": 4034310.0,
      "step": 2295
    },
    {
      "entropy": 5.86659049987793,
      "epoch": 1.8025078369905956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048656539736613403,
      "loss": 5.8567,
      "mean_token_accuracy": 0.14453689455986024,
      "num_tokens": 4042326.0,
      "step": 2300
    },
    {
      "entropy": 5.759845781326294,
      "epoch": 1.8064263322884013,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004864628274139164,
      "loss": 5.7187,
      "mean_token_accuracy": 0.15020606070756912,
      "num_tokens": 4051248.0,
      "step": 2305
    },
    {
      "entropy": 5.875640344619751,
      "epoch": 1.8103448275862069,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004863598795433391,
      "loss": 5.842,
      "mean_token_accuracy": 0.1422995701432228,
      "num_tokens": 4060227.0,
      "step": 2310
    },
    {
      "entropy": 6.000100374221802,
      "epoch": 1.8142633228840124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048625655393838666,
      "loss": 5.9161,
      "mean_token_accuracy": 0.1404888704419136,
      "num_tokens": 4069700.0,
      "step": 2315
    },
    {
      "entropy": 5.8732929706573485,
      "epoch": 1.8181818181818183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004861528507837186,
      "loss": 5.7869,
      "mean_token_accuracy": 0.14729702919721605,
      "num_tokens": 4079363.0,
      "step": 2320
    },
    {
      "entropy": 5.989973688125611,
      "epoch": 1.8221003134796239,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004860487702646695,
      "loss": 5.8711,
      "mean_token_accuracy": 0.14336605072021485,
      "num_tokens": 4087944.0,
      "step": 2325
    },
    {
      "entropy": 5.748229169845581,
      "epoch": 1.8260188087774294,
      "grad_norm": 1.375,
      "learning_rate": 0.0004859443125672479,
      "loss": 5.5825,
      "mean_token_accuracy": 0.16451934427022935,
      "num_tokens": 4096431.0,
      "step": 2330
    },
    {
      "entropy": 5.809984350204468,
      "epoch": 1.8299373040752351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004858394778781368,
      "loss": 5.7304,
      "mean_token_accuracy": 0.14536977857351302,
      "num_tokens": 4104894.0,
      "step": 2335
    },
    {
      "entropy": 5.85737190246582,
      "epoch": 1.8338557993730409,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004857342663846927,
      "loss": 5.7793,
      "mean_token_accuracy": 0.14765906557440758,
      "num_tokens": 4112878.0,
      "step": 2340
    },
    {
      "entropy": 5.807412147521973,
      "epoch": 1.8377742946708464,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004856286782749456,
      "loss": 5.7972,
      "mean_token_accuracy": 0.15013805031776428,
      "num_tokens": 4122501.0,
      "step": 2345
    },
    {
      "entropy": 5.830691576004028,
      "epoch": 1.841692789968652,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004855227137375986,
      "loss": 5.7605,
      "mean_token_accuracy": 0.14317431300878525,
      "num_tokens": 4131301.0,
      "step": 2350
    },
    {
      "entropy": 5.865339088439941,
      "epoch": 1.8456112852664577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004854163729620275,
      "loss": 5.8114,
      "mean_token_accuracy": 0.1458986446261406,
      "num_tokens": 4140076.0,
      "step": 2355
    },
    {
      "entropy": 5.822086429595947,
      "epoch": 1.8495297805642634,
      "grad_norm": 1.125,
      "learning_rate": 0.0004853096561382805,
      "loss": 5.7286,
      "mean_token_accuracy": 0.15770585536956788,
      "num_tokens": 4148728.0,
      "step": 2360
    },
    {
      "entropy": 5.795872402191162,
      "epoch": 1.853448275862069,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004852025634570779,
      "loss": 5.8682,
      "mean_token_accuracy": 0.14420996382832527,
      "num_tokens": 4157520.0,
      "step": 2365
    },
    {
      "entropy": 5.847042655944824,
      "epoch": 1.8573667711598745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004850950951098116,
      "loss": 5.6724,
      "mean_token_accuracy": 0.15193334817886353,
      "num_tokens": 4165547.0,
      "step": 2370
    },
    {
      "entropy": 5.845370531082153,
      "epoch": 1.8612852664576802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004849872512885451,
      "loss": 5.8217,
      "mean_token_accuracy": 0.1410464257001877,
      "num_tokens": 4174656.0,
      "step": 2375
    },
    {
      "entropy": 5.869585704803467,
      "epoch": 1.865203761755486,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004848790321860127,
      "loss": 5.8582,
      "mean_token_accuracy": 0.1439307987689972,
      "num_tokens": 4183452.0,
      "step": 2380
    },
    {
      "entropy": 5.851327800750733,
      "epoch": 1.8691222570532915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048477043799561946,
      "loss": 5.7456,
      "mean_token_accuracy": 0.14440877065062524,
      "num_tokens": 4192306.0,
      "step": 2385
    },
    {
      "entropy": 5.840966320037841,
      "epoch": 1.873040752351097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004846614689114409,
      "loss": 5.7992,
      "mean_token_accuracy": 0.14650781750679015,
      "num_tokens": 4201201.0,
      "step": 2390
    },
    {
      "entropy": 5.8397688388824465,
      "epoch": 1.876959247648903,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004845521251282223,
      "loss": 5.7082,
      "mean_token_accuracy": 0.15190000385046004,
      "num_tokens": 4209374.0,
      "step": 2395
    },
    {
      "entropy": 5.744638442993164,
      "epoch": 1.8808777429467085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004844424068413789,
      "loss": 5.6826,
      "mean_token_accuracy": 0.1490471452474594,
      "num_tokens": 4217258.0,
      "step": 2400
    },
    {
      "entropy": 5.77668948173523,
      "epoch": 1.884796238244514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048433231424699504,
      "loss": 5.7063,
      "mean_token_accuracy": 0.15233246684074403,
      "num_tokens": 4226484.0,
      "step": 2405
    },
    {
      "entropy": 5.866843271255493,
      "epoch": 1.8887147335423198,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048422184754182384,
      "loss": 5.7461,
      "mean_token_accuracy": 0.14515922889113425,
      "num_tokens": 4235546.0,
      "step": 2410
    },
    {
      "entropy": 5.902421998977661,
      "epoch": 1.8926332288401255,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004841110069232875,
      "loss": 5.9077,
      "mean_token_accuracy": 0.14326248839497566,
      "num_tokens": 4244549.0,
      "step": 2415
    },
    {
      "entropy": 5.853962659835815,
      "epoch": 1.896551724137931,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048399979258947597,
      "loss": 5.7864,
      "mean_token_accuracy": 0.15010540187358856,
      "num_tokens": 4252918.0,
      "step": 2420
    },
    {
      "entropy": 5.853643131256104,
      "epoch": 1.9004702194357366,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004838882047391474,
      "loss": 5.7754,
      "mean_token_accuracy": 0.14980721175670625,
      "num_tokens": 4261674.0,
      "step": 2425
    },
    {
      "entropy": 5.83618278503418,
      "epoch": 1.9043887147335423,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048377624357172724,
      "loss": 5.682,
      "mean_token_accuracy": 0.1527015097439289,
      "num_tokens": 4269986.0,
      "step": 2430
    },
    {
      "entropy": 5.860264682769776,
      "epoch": 1.908307210031348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048366390928730843,
      "loss": 5.7785,
      "mean_token_accuracy": 0.14807360768318176,
      "num_tokens": 4278131.0,
      "step": 2435
    },
    {
      "entropy": 5.86714735031128,
      "epoch": 1.9122257053291536,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004835512020866504,
      "loss": 5.765,
      "mean_token_accuracy": 0.14669661298394204,
      "num_tokens": 4287214.0,
      "step": 2440
    },
    {
      "entropy": 5.832046842575073,
      "epoch": 1.9161442006269591,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048343812217117925,
      "loss": 5.718,
      "mean_token_accuracy": 0.15012799203395844,
      "num_tokens": 4295949.0,
      "step": 2445
    },
    {
      "entropy": 5.764661169052124,
      "epoch": 1.9200626959247649,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048332466974298723,
      "loss": 5.6965,
      "mean_token_accuracy": 0.14930392503738404,
      "num_tokens": 4305014.0,
      "step": 2450
    },
    {
      "entropy": 5.798896789550781,
      "epoch": 1.9239811912225706,
      "grad_norm": 1.125,
      "learning_rate": 0.00048321084500483203,
      "loss": 5.6813,
      "mean_token_accuracy": 0.15102455914020538,
      "num_tokens": 4313560.0,
      "step": 2455
    },
    {
      "entropy": 5.798026752471924,
      "epoch": 1.9278996865203761,
      "grad_norm": 1.140625,
      "learning_rate": 0.000483096648160137,
      "loss": 5.7752,
      "mean_token_accuracy": 0.14778882637619972,
      "num_tokens": 4323280.0,
      "step": 2460
    },
    {
      "entropy": 5.883830404281616,
      "epoch": 1.9318181818181817,
      "grad_norm": 1.25,
      "learning_rate": 0.00048298207941299047,
      "loss": 5.8151,
      "mean_token_accuracy": 0.1466996341943741,
      "num_tokens": 4331757.0,
      "step": 2465
    },
    {
      "entropy": 5.7072854995727536,
      "epoch": 1.9357366771159876,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048286713896814536,
      "loss": 5.7525,
      "mean_token_accuracy": 0.1452985815703869,
      "num_tokens": 4341453.0,
      "step": 2470
    },
    {
      "entropy": 5.942684984207153,
      "epoch": 1.9396551724137931,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048275182703101877,
      "loss": 5.6894,
      "mean_token_accuracy": 0.14936211407184602,
      "num_tokens": 4349095.0,
      "step": 2475
    },
    {
      "entropy": 5.746797847747803,
      "epoch": 1.9435736677115987,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048263614380769193,
      "loss": 5.7959,
      "mean_token_accuracy": 0.1456053450703621,
      "num_tokens": 4357913.0,
      "step": 2480
    },
    {
      "entropy": 5.870775651931763,
      "epoch": 1.9474921630094044,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048252008950490957,
      "loss": 5.7708,
      "mean_token_accuracy": 0.14770165383815764,
      "num_tokens": 4366586.0,
      "step": 2485
    },
    {
      "entropy": 5.838080787658692,
      "epoch": 1.9514106583072102,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048240366433007935,
      "loss": 5.7125,
      "mean_token_accuracy": 0.15084166526794435,
      "num_tokens": 4375393.0,
      "step": 2490
    },
    {
      "entropy": 5.78338623046875,
      "epoch": 1.9553291536050157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048228686849127213,
      "loss": 5.7729,
      "mean_token_accuracy": 0.15526207983493806,
      "num_tokens": 4383747.0,
      "step": 2495
    },
    {
      "entropy": 5.892977046966553,
      "epoch": 1.9592476489028212,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004821697021972209,
      "loss": 5.7593,
      "mean_token_accuracy": 0.14306001290678977,
      "num_tokens": 4392085.0,
      "step": 2500
    },
    {
      "epoch": 1.9592476489028212,
      "eval_entropy": 5.763626485831978,
      "eval_loss": 5.954261302947998,
      "eval_mean_token_accuracy": 0.143982818169707,
      "eval_num_tokens": 4392085.0,
      "eval_runtime": 3.0476,
      "eval_samples_per_second": 1352.559,
      "eval_steps_per_second": 169.316,
      "step": 2500
    }
  ],
  "logging_steps": 5,
  "max_steps": 12750,
  "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": 5913023374080000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}