ita-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10/checkpoint-9000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.072617246596067,
  "eval_steps": 500,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 4.801114463806153,
      "epoch": 0.005042864346949067,
      "grad_norm": 14.3125,
      "learning_rate": 2e-06,
      "loss": 14.3768,
      "mean_token_accuracy": 0.0,
      "num_tokens": 11122.0,
      "step": 5
    },
    {
      "entropy": 4.812651681900024,
      "epoch": 0.010085728693898134,
      "grad_norm": 15.5,
      "learning_rate": 4.5e-06,
      "loss": 14.2394,
      "mean_token_accuracy": 0.0,
      "num_tokens": 20499.0,
      "step": 10
    },
    {
      "entropy": 4.861662435531616,
      "epoch": 0.015128593040847202,
      "grad_norm": 20.375,
      "learning_rate": 7e-06,
      "loss": 14.0486,
      "mean_token_accuracy": 0.00012091898825019597,
      "num_tokens": 30664.0,
      "step": 15
    },
    {
      "entropy": 5.152417182922363,
      "epoch": 0.020171457387796268,
      "grad_norm": 32.25,
      "learning_rate": 9.5e-06,
      "loss": 13.3871,
      "mean_token_accuracy": 0.0,
      "num_tokens": 41767.0,
      "step": 20
    },
    {
      "entropy": 7.587842893600464,
      "epoch": 0.025214321734745335,
      "grad_norm": 14.25,
      "learning_rate": 1.2e-05,
      "loss": 11.6896,
      "mean_token_accuracy": 0.0,
      "num_tokens": 53233.0,
      "step": 25
    },
    {
      "entropy": 10.463200283050536,
      "epoch": 0.030257186081694403,
      "grad_norm": 3.828125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.7364,
      "mean_token_accuracy": 0.004124969011172652,
      "num_tokens": 63591.0,
      "step": 30
    },
    {
      "entropy": 10.693463897705078,
      "epoch": 0.03530005042864347,
      "grad_norm": 3.296875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.5492,
      "mean_token_accuracy": 0.013820098433643579,
      "num_tokens": 75211.0,
      "step": 35
    },
    {
      "entropy": 10.683560562133788,
      "epoch": 0.040342914775592535,
      "grad_norm": 2.640625,
      "learning_rate": 1.95e-05,
      "loss": 10.2263,
      "mean_token_accuracy": 0.015752001013606787,
      "num_tokens": 86163.0,
      "step": 40
    },
    {
      "entropy": 10.606955909729004,
      "epoch": 0.0453857791225416,
      "grad_norm": 2.609375,
      "learning_rate": 2.2e-05,
      "loss": 9.9408,
      "mean_token_accuracy": 0.02095262976363301,
      "num_tokens": 96812.0,
      "step": 45
    },
    {
      "entropy": 10.570794868469239,
      "epoch": 0.05042864346949067,
      "grad_norm": 2.015625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.767,
      "mean_token_accuracy": 0.03581012934446335,
      "num_tokens": 107246.0,
      "step": 50
    },
    {
      "entropy": 10.550295162200928,
      "epoch": 0.05547150781643974,
      "grad_norm": 2.046875,
      "learning_rate": 2.7e-05,
      "loss": 9.6982,
      "mean_token_accuracy": 0.035054446384310725,
      "num_tokens": 118169.0,
      "step": 55
    },
    {
      "entropy": 10.557003593444824,
      "epoch": 0.060514372163388806,
      "grad_norm": 1.9375,
      "learning_rate": 2.95e-05,
      "loss": 9.596,
      "mean_token_accuracy": 0.03818601593375206,
      "num_tokens": 129744.0,
      "step": 60
    },
    {
      "entropy": 10.555067825317384,
      "epoch": 0.06555723651033787,
      "grad_norm": 1.9609375,
      "learning_rate": 3.2e-05,
      "loss": 9.4879,
      "mean_token_accuracy": 0.030292158760130406,
      "num_tokens": 139826.0,
      "step": 65
    },
    {
      "entropy": 10.539035034179687,
      "epoch": 0.07060010085728693,
      "grad_norm": 1.890625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.4026,
      "mean_token_accuracy": 0.029551995545625688,
      "num_tokens": 149767.0,
      "step": 70
    },
    {
      "entropy": 10.505377006530761,
      "epoch": 0.07564296520423601,
      "grad_norm": 1.734375,
      "learning_rate": 3.7e-05,
      "loss": 9.3373,
      "mean_token_accuracy": 0.039892033115029334,
      "num_tokens": 160895.0,
      "step": 75
    },
    {
      "entropy": 10.469569301605224,
      "epoch": 0.08068582955118507,
      "grad_norm": 1.7734375,
      "learning_rate": 3.95e-05,
      "loss": 9.2884,
      "mean_token_accuracy": 0.03964783567935228,
      "num_tokens": 172080.0,
      "step": 80
    },
    {
      "entropy": 10.455696964263916,
      "epoch": 0.08572869389813415,
      "grad_norm": 1.640625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.2276,
      "mean_token_accuracy": 0.03601204827427864,
      "num_tokens": 182836.0,
      "step": 85
    },
    {
      "entropy": 10.45373363494873,
      "epoch": 0.0907715582450832,
      "grad_norm": 1.8828125,
      "learning_rate": 4.45e-05,
      "loss": 9.1377,
      "mean_token_accuracy": 0.02935194130986929,
      "num_tokens": 194781.0,
      "step": 90
    },
    {
      "entropy": 10.433523464202882,
      "epoch": 0.09581442259203228,
      "grad_norm": 1.78125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.9882,
      "mean_token_accuracy": 0.03305415883660316,
      "num_tokens": 206722.0,
      "step": 95
    },
    {
      "entropy": 10.359123516082764,
      "epoch": 0.10085728693898134,
      "grad_norm": 1.8984375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.8489,
      "mean_token_accuracy": 0.04083778336644173,
      "num_tokens": 217167.0,
      "step": 100
    },
    {
      "entropy": 10.264568710327149,
      "epoch": 0.1059001512859304,
      "grad_norm": 1.5546875,
      "learning_rate": 5.2e-05,
      "loss": 8.7978,
      "mean_token_accuracy": 0.04325296822935343,
      "num_tokens": 228427.0,
      "step": 105
    },
    {
      "entropy": 10.185066509246827,
      "epoch": 0.11094301563287948,
      "grad_norm": 1.5,
      "learning_rate": 5.45e-05,
      "loss": 8.6675,
      "mean_token_accuracy": 0.04498021528124809,
      "num_tokens": 239169.0,
      "step": 110
    },
    {
      "entropy": 10.122811222076416,
      "epoch": 0.11598587997982854,
      "grad_norm": 1.6015625,
      "learning_rate": 5.7e-05,
      "loss": 8.5549,
      "mean_token_accuracy": 0.039299645088613035,
      "num_tokens": 250393.0,
      "step": 115
    },
    {
      "entropy": 9.999498176574708,
      "epoch": 0.12102874432677761,
      "grad_norm": 1.28125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.3439,
      "mean_token_accuracy": 0.049582514539361,
      "num_tokens": 260573.0,
      "step": 120
    },
    {
      "entropy": 9.756069564819336,
      "epoch": 0.1260716086737267,
      "grad_norm": 1.4765625,
      "learning_rate": 6.2e-05,
      "loss": 8.2337,
      "mean_token_accuracy": 0.0456912275403738,
      "num_tokens": 271537.0,
      "step": 125
    },
    {
      "entropy": 9.617351722717284,
      "epoch": 0.13111447302067575,
      "grad_norm": 1.2890625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.1765,
      "mean_token_accuracy": 0.04553677849471569,
      "num_tokens": 282080.0,
      "step": 130
    },
    {
      "entropy": 9.397808265686034,
      "epoch": 0.1361573373676248,
      "grad_norm": 1.125,
      "learning_rate": 6.7e-05,
      "loss": 8.0293,
      "mean_token_accuracy": 0.04926962368190289,
      "num_tokens": 292509.0,
      "step": 135
    },
    {
      "entropy": 9.158194255828857,
      "epoch": 0.14120020171457387,
      "grad_norm": 0.92578125,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.9266,
      "mean_token_accuracy": 0.04916851259768009,
      "num_tokens": 303156.0,
      "step": 140
    },
    {
      "entropy": 8.969250679016113,
      "epoch": 0.14624306606152296,
      "grad_norm": 0.87109375,
      "learning_rate": 7.2e-05,
      "loss": 7.8665,
      "mean_token_accuracy": 0.04366149418056011,
      "num_tokens": 314398.0,
      "step": 145
    },
    {
      "entropy": 8.68967514038086,
      "epoch": 0.15128593040847202,
      "grad_norm": 0.8359375,
      "learning_rate": 7.45e-05,
      "loss": 7.894,
      "mean_token_accuracy": 0.04541506320238113,
      "num_tokens": 324774.0,
      "step": 150
    },
    {
      "entropy": 8.563091468811034,
      "epoch": 0.15632879475542108,
      "grad_norm": 0.984375,
      "learning_rate": 7.7e-05,
      "loss": 7.7359,
      "mean_token_accuracy": 0.04416811689734459,
      "num_tokens": 335154.0,
      "step": 155
    },
    {
      "entropy": 8.345014476776123,
      "epoch": 0.16137165910237014,
      "grad_norm": 1.3359375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.8209,
      "mean_token_accuracy": 0.05068969242274761,
      "num_tokens": 346180.0,
      "step": 160
    },
    {
      "entropy": 8.28213872909546,
      "epoch": 0.1664145234493192,
      "grad_norm": 0.8046875,
      "learning_rate": 8.2e-05,
      "loss": 7.7329,
      "mean_token_accuracy": 0.04978744387626648,
      "num_tokens": 356722.0,
      "step": 165
    },
    {
      "entropy": 8.216850662231446,
      "epoch": 0.1714573877962683,
      "grad_norm": 0.80859375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.6846,
      "mean_token_accuracy": 0.05312272384762764,
      "num_tokens": 367245.0,
      "step": 170
    },
    {
      "entropy": 8.235016822814941,
      "epoch": 0.17650025214321735,
      "grad_norm": 0.953125,
      "learning_rate": 8.7e-05,
      "loss": 7.6401,
      "mean_token_accuracy": 0.05696842148900032,
      "num_tokens": 377434.0,
      "step": 175
    },
    {
      "entropy": 8.149722671508789,
      "epoch": 0.1815431164901664,
      "grad_norm": 0.99609375,
      "learning_rate": 8.95e-05,
      "loss": 7.5829,
      "mean_token_accuracy": 0.06260009892284871,
      "num_tokens": 388685.0,
      "step": 180
    },
    {
      "entropy": 8.027026557922364,
      "epoch": 0.18658598083711547,
      "grad_norm": 1.0625,
      "learning_rate": 9.2e-05,
      "loss": 7.5612,
      "mean_token_accuracy": 0.05986677259206772,
      "num_tokens": 400559.0,
      "step": 185
    },
    {
      "entropy": 8.110166931152344,
      "epoch": 0.19162884518406456,
      "grad_norm": 1.3125,
      "learning_rate": 9.45e-05,
      "loss": 7.6123,
      "mean_token_accuracy": 0.06047765538096428,
      "num_tokens": 413059.0,
      "step": 190
    },
    {
      "entropy": 8.079530906677245,
      "epoch": 0.19667170953101362,
      "grad_norm": 1.2578125,
      "learning_rate": 9.7e-05,
      "loss": 7.5223,
      "mean_token_accuracy": 0.06776017844676971,
      "num_tokens": 424073.0,
      "step": 195
    },
    {
      "entropy": 7.999411535263062,
      "epoch": 0.20171457387796268,
      "grad_norm": 1.46875,
      "learning_rate": 9.95e-05,
      "loss": 7.4703,
      "mean_token_accuracy": 0.06763602308928966,
      "num_tokens": 435960.0,
      "step": 200
    },
    {
      "entropy": 7.994884729385376,
      "epoch": 0.20675743822491174,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000102,
      "loss": 7.4259,
      "mean_token_accuracy": 0.06415937095880508,
      "num_tokens": 445458.0,
      "step": 205
    },
    {
      "entropy": 7.940965747833252,
      "epoch": 0.2118003025718608,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.4427,
      "mean_token_accuracy": 0.06450643800199032,
      "num_tokens": 456237.0,
      "step": 210
    },
    {
      "entropy": 7.971080732345581,
      "epoch": 0.2168431669188099,
      "grad_norm": 1.171875,
      "learning_rate": 0.000107,
      "loss": 7.4559,
      "mean_token_accuracy": 0.06802051141858101,
      "num_tokens": 467210.0,
      "step": 215
    },
    {
      "entropy": 7.95701413154602,
      "epoch": 0.22188603126575895,
      "grad_norm": 1.734375,
      "learning_rate": 0.0001095,
      "loss": 7.4836,
      "mean_token_accuracy": 0.06283098589628935,
      "num_tokens": 478447.0,
      "step": 220
    },
    {
      "entropy": 7.983369398117065,
      "epoch": 0.22692889561270801,
      "grad_norm": 1.015625,
      "learning_rate": 0.000112,
      "loss": 7.4325,
      "mean_token_accuracy": 0.06579711437225341,
      "num_tokens": 489108.0,
      "step": 225
    },
    {
      "entropy": 7.921357250213623,
      "epoch": 0.23197175995965708,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001145,
      "loss": 7.4914,
      "mean_token_accuracy": 0.06151033192873001,
      "num_tokens": 500889.0,
      "step": 230
    },
    {
      "entropy": 7.812901639938355,
      "epoch": 0.23701462430660616,
      "grad_norm": 1.046875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.3437,
      "mean_token_accuracy": 0.06879591457545757,
      "num_tokens": 511038.0,
      "step": 235
    },
    {
      "entropy": 7.84229211807251,
      "epoch": 0.24205748865355523,
      "grad_norm": 0.984375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.4227,
      "mean_token_accuracy": 0.06682286001741886,
      "num_tokens": 522801.0,
      "step": 240
    },
    {
      "entropy": 7.874321031570434,
      "epoch": 0.24710035300050429,
      "grad_norm": 1.0,
      "learning_rate": 0.000122,
      "loss": 7.4278,
      "mean_token_accuracy": 0.06629796028137207,
      "num_tokens": 534795.0,
      "step": 245
    },
    {
      "entropy": 7.927615451812744,
      "epoch": 0.2521432173474534,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001245,
      "loss": 7.4248,
      "mean_token_accuracy": 0.06450821422040462,
      "num_tokens": 546656.0,
      "step": 250
    },
    {
      "entropy": 7.768723440170288,
      "epoch": 0.25718608169440244,
      "grad_norm": 1.015625,
      "learning_rate": 0.000127,
      "loss": 7.29,
      "mean_token_accuracy": 0.07409121245145797,
      "num_tokens": 556673.0,
      "step": 255
    },
    {
      "entropy": 7.753383922576904,
      "epoch": 0.2622289460413515,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001295,
      "loss": 7.3377,
      "mean_token_accuracy": 0.06640650182962418,
      "num_tokens": 567044.0,
      "step": 260
    },
    {
      "entropy": 7.754617547988891,
      "epoch": 0.26727181038830056,
      "grad_norm": 1.34375,
      "learning_rate": 0.000132,
      "loss": 7.2684,
      "mean_token_accuracy": 0.07174375429749488,
      "num_tokens": 577305.0,
      "step": 265
    },
    {
      "entropy": 7.622322416305542,
      "epoch": 0.2723146747352496,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3679,
      "mean_token_accuracy": 0.07188783846795559,
      "num_tokens": 589285.0,
      "step": 270
    },
    {
      "entropy": 7.7097578048706055,
      "epoch": 0.2773575390821987,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.1943,
      "mean_token_accuracy": 0.07407240830361843,
      "num_tokens": 598609.0,
      "step": 275
    },
    {
      "entropy": 7.717816877365112,
      "epoch": 0.28240040342914774,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0001395,
      "loss": 7.2602,
      "mean_token_accuracy": 0.07445880733430385,
      "num_tokens": 609872.0,
      "step": 280
    },
    {
      "entropy": 7.6907618045806885,
      "epoch": 0.2874432677760968,
      "grad_norm": 1.09375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.2711,
      "mean_token_accuracy": 0.08134151697158813,
      "num_tokens": 622583.0,
      "step": 285
    },
    {
      "entropy": 7.654936122894287,
      "epoch": 0.2924861321230459,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001445,
      "loss": 7.2412,
      "mean_token_accuracy": 0.07294923588633537,
      "num_tokens": 633592.0,
      "step": 290
    },
    {
      "entropy": 7.543200588226318,
      "epoch": 0.297528996469995,
      "grad_norm": 0.96875,
      "learning_rate": 0.000147,
      "loss": 7.169,
      "mean_token_accuracy": 0.07622305229306221,
      "num_tokens": 644858.0,
      "step": 295
    },
    {
      "entropy": 7.596815156936645,
      "epoch": 0.30257186081694404,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001495,
      "loss": 7.245,
      "mean_token_accuracy": 0.07087541855871678,
      "num_tokens": 656438.0,
      "step": 300
    },
    {
      "entropy": 7.674440145492554,
      "epoch": 0.3076147251638931,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000152,
      "loss": 7.2518,
      "mean_token_accuracy": 0.07656564638018608,
      "num_tokens": 667942.0,
      "step": 305
    },
    {
      "entropy": 7.500104904174805,
      "epoch": 0.31265758951084216,
      "grad_norm": 1.125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.1468,
      "mean_token_accuracy": 0.07846823334693909,
      "num_tokens": 678596.0,
      "step": 310
    },
    {
      "entropy": 7.610447406768799,
      "epoch": 0.3177004538577912,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000157,
      "loss": 7.3024,
      "mean_token_accuracy": 0.06737342216074467,
      "num_tokens": 691014.0,
      "step": 315
    },
    {
      "entropy": 7.612302160263061,
      "epoch": 0.3227433182047403,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001595,
      "loss": 7.2079,
      "mean_token_accuracy": 0.07754847705364228,
      "num_tokens": 701849.0,
      "step": 320
    },
    {
      "entropy": 7.523909950256348,
      "epoch": 0.32778618255168934,
      "grad_norm": 1.265625,
      "learning_rate": 0.000162,
      "loss": 7.1103,
      "mean_token_accuracy": 0.08044524192810058,
      "num_tokens": 712652.0,
      "step": 325
    },
    {
      "entropy": 7.591261148452759,
      "epoch": 0.3328290468986384,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.1597,
      "mean_token_accuracy": 0.07344280071556568,
      "num_tokens": 723704.0,
      "step": 330
    },
    {
      "entropy": 7.501336288452149,
      "epoch": 0.3378719112455875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.1081,
      "mean_token_accuracy": 0.07664402537047862,
      "num_tokens": 734715.0,
      "step": 335
    },
    {
      "entropy": 7.462927675247192,
      "epoch": 0.3429147755925366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.1461,
      "mean_token_accuracy": 0.07232181765139103,
      "num_tokens": 746772.0,
      "step": 340
    },
    {
      "entropy": 7.542045450210571,
      "epoch": 0.34795763993948564,
      "grad_norm": 1.484375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.1028,
      "mean_token_accuracy": 0.07609748989343643,
      "num_tokens": 757356.0,
      "step": 345
    },
    {
      "entropy": 7.513222789764404,
      "epoch": 0.3530005042864347,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.1177,
      "mean_token_accuracy": 0.07893777638673782,
      "num_tokens": 768269.0,
      "step": 350
    },
    {
      "entropy": 7.457395029067993,
      "epoch": 0.35804336863338376,
      "grad_norm": 1.0625,
      "learning_rate": 0.000177,
      "loss": 7.0621,
      "mean_token_accuracy": 0.07398857697844505,
      "num_tokens": 779553.0,
      "step": 355
    },
    {
      "entropy": 7.4384712219238285,
      "epoch": 0.3630862329803328,
      "grad_norm": 1.125,
      "learning_rate": 0.0001795,
      "loss": 7.1477,
      "mean_token_accuracy": 0.07859819531440734,
      "num_tokens": 791466.0,
      "step": 360
    },
    {
      "entropy": 7.519556713104248,
      "epoch": 0.3681290973272819,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000182,
      "loss": 7.0409,
      "mean_token_accuracy": 0.07564883790910244,
      "num_tokens": 802064.0,
      "step": 365
    },
    {
      "entropy": 7.367311239242554,
      "epoch": 0.37317196167423095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001845,
      "loss": 7.0721,
      "mean_token_accuracy": 0.08020395338535309,
      "num_tokens": 813534.0,
      "step": 370
    },
    {
      "entropy": 7.481927680969238,
      "epoch": 0.37821482602118,
      "grad_norm": 1.140625,
      "learning_rate": 0.000187,
      "loss": 7.0958,
      "mean_token_accuracy": 0.07429472133517265,
      "num_tokens": 824783.0,
      "step": 375
    },
    {
      "entropy": 7.396484994888306,
      "epoch": 0.3832576903681291,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001895,
      "loss": 7.0813,
      "mean_token_accuracy": 0.0780322715640068,
      "num_tokens": 835019.0,
      "step": 380
    },
    {
      "entropy": 7.43559775352478,
      "epoch": 0.3883005547150782,
      "grad_norm": 0.96875,
      "learning_rate": 0.000192,
      "loss": 6.9778,
      "mean_token_accuracy": 0.07786969803273677,
      "num_tokens": 845954.0,
      "step": 385
    },
    {
      "entropy": 7.349666690826416,
      "epoch": 0.39334341906202724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001945,
      "loss": 6.9968,
      "mean_token_accuracy": 0.08358137160539628,
      "num_tokens": 856393.0,
      "step": 390
    },
    {
      "entropy": 7.377739429473877,
      "epoch": 0.3983862834089763,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0058,
      "mean_token_accuracy": 0.08096945993602275,
      "num_tokens": 868079.0,
      "step": 395
    },
    {
      "entropy": 7.420732831954956,
      "epoch": 0.40342914775592537,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.0597,
      "mean_token_accuracy": 0.08519057407975197,
      "num_tokens": 878995.0,
      "step": 400
    },
    {
      "entropy": 7.359525680541992,
      "epoch": 0.4084720121028744,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000202,
      "loss": 6.9767,
      "mean_token_accuracy": 0.08249035105109215,
      "num_tokens": 889159.0,
      "step": 405
    },
    {
      "entropy": 7.397777795791626,
      "epoch": 0.4135148764498235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.0906,
      "mean_token_accuracy": 0.07337038293480873,
      "num_tokens": 900186.0,
      "step": 410
    },
    {
      "entropy": 7.36064920425415,
      "epoch": 0.41855774079677255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000207,
      "loss": 7.0054,
      "mean_token_accuracy": 0.0781717125326395,
      "num_tokens": 910713.0,
      "step": 415
    },
    {
      "entropy": 7.512122440338135,
      "epoch": 0.4236006051437216,
      "grad_norm": 2.359375,
      "learning_rate": 0.0002095,
      "loss": 7.0956,
      "mean_token_accuracy": 0.07817071452736854,
      "num_tokens": 922020.0,
      "step": 420
    },
    {
      "entropy": 7.32919135093689,
      "epoch": 0.4286434694906707,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000212,
      "loss": 7.0302,
      "mean_token_accuracy": 0.07741615772247315,
      "num_tokens": 933033.0,
      "step": 425
    },
    {
      "entropy": 7.3811193943023685,
      "epoch": 0.4336863338376198,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002145,
      "loss": 7.0623,
      "mean_token_accuracy": 0.07646705359220504,
      "num_tokens": 944394.0,
      "step": 430
    },
    {
      "entropy": 7.382322645187378,
      "epoch": 0.43872919818456885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9714,
      "mean_token_accuracy": 0.0816558338701725,
      "num_tokens": 955337.0,
      "step": 435
    },
    {
      "entropy": 7.19914608001709,
      "epoch": 0.4437720625315179,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002195,
      "loss": 6.9435,
      "mean_token_accuracy": 0.08770947605371475,
      "num_tokens": 967020.0,
      "step": 440
    },
    {
      "entropy": 7.364274787902832,
      "epoch": 0.44881492687846697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000222,
      "loss": 6.9686,
      "mean_token_accuracy": 0.08233175501227379,
      "num_tokens": 978223.0,
      "step": 445
    },
    {
      "entropy": 7.2527648448944095,
      "epoch": 0.45385779122541603,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002245,
      "loss": 6.9751,
      "mean_token_accuracy": 0.08211390711367131,
      "num_tokens": 990239.0,
      "step": 450
    },
    {
      "entropy": 7.344982099533081,
      "epoch": 0.4589006555723651,
      "grad_norm": 1.109375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.8979,
      "mean_token_accuracy": 0.08162759393453597,
      "num_tokens": 1001834.0,
      "step": 455
    },
    {
      "entropy": 7.236715698242188,
      "epoch": 0.46394351991931415,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8744,
      "mean_token_accuracy": 0.09091370068490505,
      "num_tokens": 1012321.0,
      "step": 460
    },
    {
      "entropy": 7.2280114650726315,
      "epoch": 0.4689863842662632,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.9796,
      "mean_token_accuracy": 0.08122261688113212,
      "num_tokens": 1024500.0,
      "step": 465
    },
    {
      "entropy": 7.270635604858398,
      "epoch": 0.47402924861321233,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8926,
      "mean_token_accuracy": 0.08805500157177448,
      "num_tokens": 1034825.0,
      "step": 470
    },
    {
      "entropy": 7.1545487403869625,
      "epoch": 0.4790721129601614,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000237,
      "loss": 6.8648,
      "mean_token_accuracy": 0.08775893896818161,
      "num_tokens": 1046174.0,
      "step": 475
    },
    {
      "entropy": 7.252149486541748,
      "epoch": 0.48411497730711045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002395,
      "loss": 6.7807,
      "mean_token_accuracy": 0.09338528215885163,
      "num_tokens": 1055930.0,
      "step": 480
    },
    {
      "entropy": 7.144695091247558,
      "epoch": 0.4891578416540595,
      "grad_norm": 1.1875,
      "learning_rate": 0.000242,
      "loss": 6.8153,
      "mean_token_accuracy": 0.09350402504205704,
      "num_tokens": 1066433.0,
      "step": 485
    },
    {
      "entropy": 7.173965978622436,
      "epoch": 0.49420070600100857,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002445,
      "loss": 6.9306,
      "mean_token_accuracy": 0.08193769156932831,
      "num_tokens": 1077303.0,
      "step": 490
    },
    {
      "entropy": 7.231033754348755,
      "epoch": 0.49924357034795763,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000247,
      "loss": 6.8453,
      "mean_token_accuracy": 0.08849839866161346,
      "num_tokens": 1088053.0,
      "step": 495
    },
    {
      "entropy": 7.107387685775757,
      "epoch": 0.5042864346949067,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002495,
      "loss": 6.8499,
      "mean_token_accuracy": 0.09236783608794212,
      "num_tokens": 1098518.0,
      "step": 500
    },
    {
      "epoch": 0.5042864346949067,
      "eval_entropy": 7.061949415667973,
      "eval_loss": 6.937669277191162,
      "eval_mean_token_accuracy": 0.09019645222915341,
      "eval_num_tokens": 1098518.0,
      "eval_runtime": 2.6772,
      "eval_samples_per_second": 1172.119,
      "eval_steps_per_second": 146.795,
      "step": 500
    },
    {
      "entropy": 7.228630304336548,
      "epoch": 0.5093292990418558,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000252,
      "loss": 6.8577,
      "mean_token_accuracy": 0.08858487010002136,
      "num_tokens": 1109768.0,
      "step": 505
    },
    {
      "entropy": 7.081269025802612,
      "epoch": 0.5143721633888049,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002545,
      "loss": 6.8846,
      "mean_token_accuracy": 0.09013336524367332,
      "num_tokens": 1121969.0,
      "step": 510
    },
    {
      "entropy": 7.162287855148316,
      "epoch": 0.5194150277357539,
      "grad_norm": 1.265625,
      "learning_rate": 0.000257,
      "loss": 6.7391,
      "mean_token_accuracy": 0.09084992706775666,
      "num_tokens": 1132405.0,
      "step": 515
    },
    {
      "entropy": 7.103027248382569,
      "epoch": 0.524457892082703,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002595,
      "loss": 6.8871,
      "mean_token_accuracy": 0.07809185199439525,
      "num_tokens": 1143705.0,
      "step": 520
    },
    {
      "entropy": 7.171495628356934,
      "epoch": 0.529500756429652,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000262,
      "loss": 6.8513,
      "mean_token_accuracy": 0.08691945150494576,
      "num_tokens": 1154445.0,
      "step": 525
    },
    {
      "entropy": 7.117181491851807,
      "epoch": 0.5345436207766011,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.9407,
      "mean_token_accuracy": 0.08625975549221039,
      "num_tokens": 1166895.0,
      "step": 530
    },
    {
      "entropy": 7.119032716751098,
      "epoch": 0.5395864851235502,
      "grad_norm": 1.015625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.8261,
      "mean_token_accuracy": 0.09287582933902741,
      "num_tokens": 1176744.0,
      "step": 535
    },
    {
      "entropy": 7.1486015796661375,
      "epoch": 0.5446293494704992,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.8097,
      "mean_token_accuracy": 0.08768841177225113,
      "num_tokens": 1187024.0,
      "step": 540
    },
    {
      "entropy": 7.054621648788452,
      "epoch": 0.5496722138174484,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.8001,
      "mean_token_accuracy": 0.09048704728484154,
      "num_tokens": 1198799.0,
      "step": 545
    },
    {
      "entropy": 7.101481819152832,
      "epoch": 0.5547150781643974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002745,
      "loss": 6.7588,
      "mean_token_accuracy": 0.09443475604057312,
      "num_tokens": 1208585.0,
      "step": 550
    },
    {
      "entropy": 7.084757947921753,
      "epoch": 0.5597579425113465,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000277,
      "loss": 6.8007,
      "mean_token_accuracy": 0.09204660803079605,
      "num_tokens": 1218268.0,
      "step": 555
    },
    {
      "entropy": 7.034044981002808,
      "epoch": 0.5648008068582955,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002795,
      "loss": 6.8063,
      "mean_token_accuracy": 0.09505132585763931,
      "num_tokens": 1229106.0,
      "step": 560
    },
    {
      "entropy": 7.110476064682007,
      "epoch": 0.5698436712052446,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.7378,
      "mean_token_accuracy": 0.09246841743588448,
      "num_tokens": 1239741.0,
      "step": 565
    },
    {
      "entropy": 6.986316823959351,
      "epoch": 0.5748865355521936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002845,
      "loss": 6.7035,
      "mean_token_accuracy": 0.08977311328053475,
      "num_tokens": 1249658.0,
      "step": 570
    },
    {
      "entropy": 6.991911506652832,
      "epoch": 0.5799293998991427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000287,
      "loss": 6.7475,
      "mean_token_accuracy": 0.09191381558775902,
      "num_tokens": 1259731.0,
      "step": 575
    },
    {
      "entropy": 7.12698917388916,
      "epoch": 0.5849722642460918,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002895,
      "loss": 6.7476,
      "mean_token_accuracy": 0.09400721192359925,
      "num_tokens": 1270087.0,
      "step": 580
    },
    {
      "entropy": 6.9526652812957765,
      "epoch": 0.5900151285930408,
      "grad_norm": 1.234375,
      "learning_rate": 0.000292,
      "loss": 6.7166,
      "mean_token_accuracy": 0.09413416013121605,
      "num_tokens": 1282567.0,
      "step": 585
    },
    {
      "entropy": 6.915704536437988,
      "epoch": 0.59505799293999,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002945,
      "loss": 6.7789,
      "mean_token_accuracy": 0.09547073394060135,
      "num_tokens": 1294133.0,
      "step": 590
    },
    {
      "entropy": 6.990864229202271,
      "epoch": 0.600100857286939,
      "grad_norm": 1.0,
      "learning_rate": 0.000297,
      "loss": 6.6449,
      "mean_token_accuracy": 0.0973285436630249,
      "num_tokens": 1305306.0,
      "step": 595
    },
    {
      "entropy": 6.955234050750732,
      "epoch": 0.6051437216338881,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002995,
      "loss": 6.7173,
      "mean_token_accuracy": 0.09564844146370888,
      "num_tokens": 1316379.0,
      "step": 600
    },
    {
      "entropy": 6.972650957107544,
      "epoch": 0.6101865859808371,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000302,
      "loss": 6.6907,
      "mean_token_accuracy": 0.09237802252173424,
      "num_tokens": 1328121.0,
      "step": 605
    },
    {
      "entropy": 6.9388415813446045,
      "epoch": 0.6152294503277862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003045,
      "loss": 6.7319,
      "mean_token_accuracy": 0.08763695284724235,
      "num_tokens": 1338728.0,
      "step": 610
    },
    {
      "entropy": 6.982970094680786,
      "epoch": 0.6202723146747352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000307,
      "loss": 6.711,
      "mean_token_accuracy": 0.0885362021625042,
      "num_tokens": 1349126.0,
      "step": 615
    },
    {
      "entropy": 6.916971921920776,
      "epoch": 0.6253151790216843,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003095,
      "loss": 6.6068,
      "mean_token_accuracy": 0.09231596812605858,
      "num_tokens": 1359813.0,
      "step": 620
    },
    {
      "entropy": 6.890164804458618,
      "epoch": 0.6303580433686334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000312,
      "loss": 6.7633,
      "mean_token_accuracy": 0.09019490703940392,
      "num_tokens": 1371872.0,
      "step": 625
    },
    {
      "entropy": 6.952266883850098,
      "epoch": 0.6354009077155824,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003145,
      "loss": 6.6359,
      "mean_token_accuracy": 0.08751971200108528,
      "num_tokens": 1382789.0,
      "step": 630
    },
    {
      "entropy": 6.918084335327149,
      "epoch": 0.6404437720625316,
      "grad_norm": 1.25,
      "learning_rate": 0.000317,
      "loss": 6.7236,
      "mean_token_accuracy": 0.09168680310249329,
      "num_tokens": 1393419.0,
      "step": 635
    },
    {
      "entropy": 6.888974285125732,
      "epoch": 0.6454866364094806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195,
      "loss": 6.5702,
      "mean_token_accuracy": 0.09415488988161087,
      "num_tokens": 1404674.0,
      "step": 640
    },
    {
      "entropy": 6.903821563720703,
      "epoch": 0.6505295007564297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000322,
      "loss": 6.7776,
      "mean_token_accuracy": 0.0927918791770935,
      "num_tokens": 1416252.0,
      "step": 645
    },
    {
      "entropy": 6.834169626235962,
      "epoch": 0.6555723651033787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.584,
      "mean_token_accuracy": 0.09287177100777626,
      "num_tokens": 1426681.0,
      "step": 650
    },
    {
      "entropy": 6.861099433898926,
      "epoch": 0.6606152294503278,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.58,
      "mean_token_accuracy": 0.09400497153401374,
      "num_tokens": 1436640.0,
      "step": 655
    },
    {
      "entropy": 6.8637463569641115,
      "epoch": 0.6656580937972768,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6155,
      "mean_token_accuracy": 0.09446947798132896,
      "num_tokens": 1447183.0,
      "step": 660
    },
    {
      "entropy": 6.829385280609131,
      "epoch": 0.6707009581442259,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6385,
      "mean_token_accuracy": 0.0950157254934311,
      "num_tokens": 1458351.0,
      "step": 665
    },
    {
      "entropy": 6.804144191741943,
      "epoch": 0.675743822491175,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6369,
      "mean_token_accuracy": 0.09405813217163086,
      "num_tokens": 1469522.0,
      "step": 670
    },
    {
      "entropy": 6.76968903541565,
      "epoch": 0.680786686838124,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000337,
      "loss": 6.565,
      "mean_token_accuracy": 0.09773774519562721,
      "num_tokens": 1479947.0,
      "step": 675
    },
    {
      "entropy": 6.893829965591431,
      "epoch": 0.6858295511850732,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003395,
      "loss": 6.5879,
      "mean_token_accuracy": 0.0951867401599884,
      "num_tokens": 1490374.0,
      "step": 680
    },
    {
      "entropy": 6.818552112579345,
      "epoch": 0.6908724155320222,
      "grad_norm": 1.125,
      "learning_rate": 0.000342,
      "loss": 6.58,
      "mean_token_accuracy": 0.10172139406204224,
      "num_tokens": 1500939.0,
      "step": 685
    },
    {
      "entropy": 6.806358861923218,
      "epoch": 0.6959152798789713,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.6962,
      "mean_token_accuracy": 0.09438932090997695,
      "num_tokens": 1512803.0,
      "step": 690
    },
    {
      "entropy": 6.681111001968384,
      "epoch": 0.7009581442259203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000347,
      "loss": 6.5066,
      "mean_token_accuracy": 0.0982896327972412,
      "num_tokens": 1523679.0,
      "step": 695
    },
    {
      "entropy": 6.747470474243164,
      "epoch": 0.7060010085728694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003495,
      "loss": 6.527,
      "mean_token_accuracy": 0.09784084260463714,
      "num_tokens": 1533892.0,
      "step": 700
    },
    {
      "entropy": 6.73495717048645,
      "epoch": 0.7110438729198184,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000352,
      "loss": 6.5033,
      "mean_token_accuracy": 0.10459957495331765,
      "num_tokens": 1543935.0,
      "step": 705
    },
    {
      "entropy": 6.718096542358398,
      "epoch": 0.7160867372667675,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003545,
      "loss": 6.5985,
      "mean_token_accuracy": 0.09408540651202202,
      "num_tokens": 1555989.0,
      "step": 710
    },
    {
      "entropy": 6.870676755905151,
      "epoch": 0.7211296016137166,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000357,
      "loss": 6.5575,
      "mean_token_accuracy": 0.09637962281703949,
      "num_tokens": 1566355.0,
      "step": 715
    },
    {
      "entropy": 6.679803085327149,
      "epoch": 0.7261724659606656,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003595,
      "loss": 6.5917,
      "mean_token_accuracy": 0.09732575789093971,
      "num_tokens": 1578403.0,
      "step": 720
    },
    {
      "entropy": 6.768120145797729,
      "epoch": 0.7312153303076148,
      "grad_norm": 1.078125,
      "learning_rate": 0.000362,
      "loss": 6.6428,
      "mean_token_accuracy": 0.09719825014472008,
      "num_tokens": 1590072.0,
      "step": 725
    },
    {
      "entropy": 6.76740460395813,
      "epoch": 0.7362581946545638,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003645,
      "loss": 6.5704,
      "mean_token_accuracy": 0.10138242095708846,
      "num_tokens": 1602051.0,
      "step": 730
    },
    {
      "entropy": 6.809198188781738,
      "epoch": 0.7413010590015129,
      "grad_norm": 0.875,
      "learning_rate": 0.000367,
      "loss": 6.6497,
      "mean_token_accuracy": 0.10376588925719261,
      "num_tokens": 1615741.0,
      "step": 735
    },
    {
      "entropy": 6.637604141235352,
      "epoch": 0.7463439233484619,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003695,
      "loss": 6.4981,
      "mean_token_accuracy": 0.10018217116594315,
      "num_tokens": 1627042.0,
      "step": 740
    },
    {
      "entropy": 6.76273832321167,
      "epoch": 0.751386787695411,
      "grad_norm": 1.0,
      "learning_rate": 0.000372,
      "loss": 6.564,
      "mean_token_accuracy": 0.10186072885990143,
      "num_tokens": 1638777.0,
      "step": 745
    },
    {
      "entropy": 6.612446403503418,
      "epoch": 0.75642965204236,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003745,
      "loss": 6.4761,
      "mean_token_accuracy": 0.10184447392821312,
      "num_tokens": 1650266.0,
      "step": 750
    },
    {
      "entropy": 6.727019262313843,
      "epoch": 0.7614725163893091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000377,
      "loss": 6.5096,
      "mean_token_accuracy": 0.10364911928772927,
      "num_tokens": 1660500.0,
      "step": 755
    },
    {
      "entropy": 6.626423454284668,
      "epoch": 0.7665153807362582,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003795,
      "loss": 6.4727,
      "mean_token_accuracy": 0.10132494196295738,
      "num_tokens": 1671685.0,
      "step": 760
    },
    {
      "entropy": 6.698447322845459,
      "epoch": 0.7715582450832073,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000382,
      "loss": 6.5676,
      "mean_token_accuracy": 0.0920490249991417,
      "num_tokens": 1681954.0,
      "step": 765
    },
    {
      "entropy": 6.720503091812134,
      "epoch": 0.7766011094301564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003845,
      "loss": 6.5492,
      "mean_token_accuracy": 0.10080667063593865,
      "num_tokens": 1691761.0,
      "step": 770
    },
    {
      "entropy": 6.661979913711548,
      "epoch": 0.7816439737771054,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4826,
      "mean_token_accuracy": 0.10153618305921555,
      "num_tokens": 1701715.0,
      "step": 775
    },
    {
      "entropy": 6.6159828186035154,
      "epoch": 0.7866868381240545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4372,
      "mean_token_accuracy": 0.1015503615140915,
      "num_tokens": 1714530.0,
      "step": 780
    },
    {
      "entropy": 6.619305753707886,
      "epoch": 0.7917297024710035,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.527,
      "mean_token_accuracy": 0.10115180537104607,
      "num_tokens": 1725766.0,
      "step": 785
    },
    {
      "entropy": 6.576332902908325,
      "epoch": 0.7967725668179526,
      "grad_norm": 0.9375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4528,
      "mean_token_accuracy": 0.1076912261545658,
      "num_tokens": 1738307.0,
      "step": 790
    },
    {
      "entropy": 6.6750335693359375,
      "epoch": 0.8018154311649016,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5231,
      "mean_token_accuracy": 0.09890256971120834,
      "num_tokens": 1749129.0,
      "step": 795
    },
    {
      "entropy": 6.730860948562622,
      "epoch": 0.8068582955118507,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003995,
      "loss": 6.5414,
      "mean_token_accuracy": 0.0941914051771164,
      "num_tokens": 1759426.0,
      "step": 800
    },
    {
      "entropy": 6.552480602264405,
      "epoch": 0.8119011598587998,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000402,
      "loss": 6.3638,
      "mean_token_accuracy": 0.10434163883328437,
      "num_tokens": 1770906.0,
      "step": 805
    },
    {
      "entropy": 6.582395315170288,
      "epoch": 0.8169440242057489,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004045,
      "loss": 6.5188,
      "mean_token_accuracy": 0.10247278958559036,
      "num_tokens": 1782424.0,
      "step": 810
    },
    {
      "entropy": 6.616803169250488,
      "epoch": 0.821986888552698,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5031,
      "mean_token_accuracy": 0.1025144636631012,
      "num_tokens": 1794331.0,
      "step": 815
    },
    {
      "entropy": 6.614215946197509,
      "epoch": 0.827029752899647,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004095,
      "loss": 6.4325,
      "mean_token_accuracy": 0.09733226075768471,
      "num_tokens": 1805395.0,
      "step": 820
    },
    {
      "entropy": 6.563755941390991,
      "epoch": 0.8320726172465961,
      "grad_norm": 1.234375,
      "learning_rate": 0.000412,
      "loss": 6.4082,
      "mean_token_accuracy": 0.0999007374048233,
      "num_tokens": 1815820.0,
      "step": 825
    },
    {
      "entropy": 6.52943525314331,
      "epoch": 0.8371154815935451,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004145,
      "loss": 6.4269,
      "mean_token_accuracy": 0.10219496190547943,
      "num_tokens": 1825754.0,
      "step": 830
    },
    {
      "entropy": 6.578064441680908,
      "epoch": 0.8421583459404942,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000417,
      "loss": 6.4729,
      "mean_token_accuracy": 0.10351292341947556,
      "num_tokens": 1837899.0,
      "step": 835
    },
    {
      "entropy": 6.625820207595825,
      "epoch": 0.8472012102874432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004195,
      "loss": 6.4095,
      "mean_token_accuracy": 0.10529221072793007,
      "num_tokens": 1847659.0,
      "step": 840
    },
    {
      "entropy": 6.479169607162476,
      "epoch": 0.8522440746343923,
      "grad_norm": 0.96875,
      "learning_rate": 0.000422,
      "loss": 6.3946,
      "mean_token_accuracy": 0.1047472171485424,
      "num_tokens": 1860126.0,
      "step": 845
    },
    {
      "entropy": 6.462310934066773,
      "epoch": 0.8572869389813415,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004245,
      "loss": 6.3943,
      "mean_token_accuracy": 0.10278471186757088,
      "num_tokens": 1870567.0,
      "step": 850
    },
    {
      "entropy": 6.556047487258911,
      "epoch": 0.8623298033282905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000427,
      "loss": 6.5126,
      "mean_token_accuracy": 0.09607306495308876,
      "num_tokens": 1880274.0,
      "step": 855
    },
    {
      "entropy": 6.602808713912964,
      "epoch": 0.8673726676752396,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004295,
      "loss": 6.5002,
      "mean_token_accuracy": 0.09916604906320572,
      "num_tokens": 1891122.0,
      "step": 860
    },
    {
      "entropy": 6.556145715713501,
      "epoch": 0.8724155320221886,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000432,
      "loss": 6.4984,
      "mean_token_accuracy": 0.09922838732600212,
      "num_tokens": 1902629.0,
      "step": 865
    },
    {
      "entropy": 6.619171953201294,
      "epoch": 0.8774583963691377,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004345,
      "loss": 6.4293,
      "mean_token_accuracy": 0.10180645212531089,
      "num_tokens": 1913808.0,
      "step": 870
    },
    {
      "entropy": 6.459190368652344,
      "epoch": 0.8825012607160867,
      "grad_norm": 1.140625,
      "learning_rate": 0.000437,
      "loss": 6.4308,
      "mean_token_accuracy": 0.10478962510824204,
      "num_tokens": 1925973.0,
      "step": 875
    },
    {
      "entropy": 6.569327783584595,
      "epoch": 0.8875441250630358,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395,
      "loss": 6.4111,
      "mean_token_accuracy": 0.11385421827435493,
      "num_tokens": 1937808.0,
      "step": 880
    },
    {
      "entropy": 6.516397142410279,
      "epoch": 0.8925869894099848,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.4027,
      "mean_token_accuracy": 0.10975287705659867,
      "num_tokens": 1948581.0,
      "step": 885
    },
    {
      "entropy": 6.4545928001403805,
      "epoch": 0.8976298537569339,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445,
      "loss": 6.4461,
      "mean_token_accuracy": 0.10252013728022576,
      "num_tokens": 1959977.0,
      "step": 890
    },
    {
      "entropy": 6.630584287643432,
      "epoch": 0.902672718103883,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000447,
      "loss": 6.4874,
      "mean_token_accuracy": 0.09852444604039193,
      "num_tokens": 1971328.0,
      "step": 895
    },
    {
      "entropy": 6.492565059661866,
      "epoch": 0.9077155824508321,
      "grad_norm": 1.0,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.581,
      "mean_token_accuracy": 0.10265940502285957,
      "num_tokens": 1983805.0,
      "step": 900
    },
    {
      "entropy": 6.608131027221679,
      "epoch": 0.9127584467977812,
      "grad_norm": 1.0,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.5166,
      "mean_token_accuracy": 0.10072435140609741,
      "num_tokens": 1994773.0,
      "step": 905
    },
    {
      "entropy": 6.5182877540588375,
      "epoch": 0.9178013111447302,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3935,
      "mean_token_accuracy": 0.10560430958867073,
      "num_tokens": 2005600.0,
      "step": 910
    },
    {
      "entropy": 6.5030282497406,
      "epoch": 0.9228441754916793,
      "grad_norm": 0.890625,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3335,
      "mean_token_accuracy": 0.10612398311495781,
      "num_tokens": 2016580.0,
      "step": 915
    },
    {
      "entropy": 6.493871545791626,
      "epoch": 0.9278870398386283,
      "grad_norm": 1.0,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4296,
      "mean_token_accuracy": 0.10680199563503265,
      "num_tokens": 2030318.0,
      "step": 920
    },
    {
      "entropy": 6.425577878952026,
      "epoch": 0.9329299041855774,
      "grad_norm": 0.90625,
      "learning_rate": 0.000462,
      "loss": 6.3959,
      "mean_token_accuracy": 0.10729326754808426,
      "num_tokens": 2041744.0,
      "step": 925
    },
    {
      "entropy": 6.464505004882812,
      "epoch": 0.9379727685325264,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004645,
      "loss": 6.4385,
      "mean_token_accuracy": 0.10069913193583488,
      "num_tokens": 2053628.0,
      "step": 930
    },
    {
      "entropy": 6.5448229789733885,
      "epoch": 0.9430156328794755,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000467,
      "loss": 6.4141,
      "mean_token_accuracy": 0.10871829986572265,
      "num_tokens": 2065069.0,
      "step": 935
    },
    {
      "entropy": 6.539431858062744,
      "epoch": 0.9480584972264247,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004695,
      "loss": 6.439,
      "mean_token_accuracy": 0.10412655398249626,
      "num_tokens": 2076792.0,
      "step": 940
    },
    {
      "entropy": 6.4169426441192625,
      "epoch": 0.9531013615733737,
      "grad_norm": 1.125,
      "learning_rate": 0.000472,
      "loss": 6.2986,
      "mean_token_accuracy": 0.11213012337684632,
      "num_tokens": 2087373.0,
      "step": 945
    },
    {
      "entropy": 6.446739196777344,
      "epoch": 0.9581442259203228,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004745,
      "loss": 6.3437,
      "mean_token_accuracy": 0.10855904519557953,
      "num_tokens": 2100846.0,
      "step": 950
    },
    {
      "entropy": 6.357406377792358,
      "epoch": 0.9631870902672718,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477,
      "loss": 6.3536,
      "mean_token_accuracy": 0.11016998365521431,
      "num_tokens": 2110668.0,
      "step": 955
    },
    {
      "entropy": 6.449956846237183,
      "epoch": 0.9682299546142209,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004795,
      "loss": 6.3388,
      "mean_token_accuracy": 0.10505538731813431,
      "num_tokens": 2120917.0,
      "step": 960
    },
    {
      "entropy": 6.381419658660889,
      "epoch": 0.9732728189611699,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000482,
      "loss": 6.3337,
      "mean_token_accuracy": 0.10150417611002922,
      "num_tokens": 2132929.0,
      "step": 965
    },
    {
      "entropy": 6.349277019500732,
      "epoch": 0.978315683308119,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004845,
      "loss": 6.2908,
      "mean_token_accuracy": 0.11067101880908012,
      "num_tokens": 2145251.0,
      "step": 970
    },
    {
      "entropy": 6.365310096740723,
      "epoch": 0.983358547655068,
      "grad_norm": 1.203125,
      "learning_rate": 0.000487,
      "loss": 6.3134,
      "mean_token_accuracy": 0.11331562921404839,
      "num_tokens": 2154702.0,
      "step": 975
    },
    {
      "entropy": 6.436642789840699,
      "epoch": 0.9884014120020171,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 6.3472,
      "mean_token_accuracy": 0.1113656260073185,
      "num_tokens": 2165924.0,
      "step": 980
    },
    {
      "entropy": 6.408423995971679,
      "epoch": 0.9934442763489663,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492,
      "loss": 6.3166,
      "mean_token_accuracy": 0.11193023920059204,
      "num_tokens": 2176933.0,
      "step": 985
    },
    {
      "entropy": 6.351378726959228,
      "epoch": 0.9984871406959153,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004945,
      "loss": 6.408,
      "mean_token_accuracy": 0.10353770330548287,
      "num_tokens": 2188812.0,
      "step": 990
    },
    {
      "entropy": 6.459409925672743,
      "epoch": 1.0030257186081695,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1104872499903043,
      "num_tokens": 2199238.0,
      "step": 995
    },
    {
      "entropy": 6.32691593170166,
      "epoch": 1.0080685829551186,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995,
      "loss": 6.157,
      "mean_token_accuracy": 0.11325521096587181,
      "num_tokens": 2210530.0,
      "step": 1000
    },
    {
      "epoch": 1.0080685829551186,
      "eval_entropy": 6.25044517662689,
      "eval_loss": 6.406233310699463,
      "eval_mean_token_accuracy": 0.11158415295050951,
      "eval_num_tokens": 2210530.0,
      "eval_runtime": 2.6239,
      "eval_samples_per_second": 1195.933,
      "eval_steps_per_second": 149.777,
      "step": 1000
    },
    {
      "entropy": 6.298683118820191,
      "epoch": 1.0131114473020675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999997762223182,
      "loss": 6.1697,
      "mean_token_accuracy": 0.10936820954084396,
      "num_tokens": 2221721.0,
      "step": 1005
    },
    {
      "entropy": 6.3320824146270756,
      "epoch": 1.0181543116490166,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999988671262489,
      "loss": 6.2022,
      "mean_token_accuracy": 0.11588318571448326,
      "num_tokens": 2233571.0,
      "step": 1010
    },
    {
      "entropy": 6.387577676773072,
      "epoch": 1.0231971759959657,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999972587285105,
      "loss": 6.2302,
      "mean_token_accuracy": 0.11754814460873604,
      "num_tokens": 2243419.0,
      "step": 1015
    },
    {
      "entropy": 6.327431154251099,
      "epoch": 1.0282400403429148,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999949510341015,
      "loss": 6.1845,
      "mean_token_accuracy": 0.1209840402007103,
      "num_tokens": 2254461.0,
      "step": 1020
    },
    {
      "entropy": 6.275969505310059,
      "epoch": 1.033282904689864,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999919440501947,
      "loss": 6.1625,
      "mean_token_accuracy": 0.11422310993075371,
      "num_tokens": 2266107.0,
      "step": 1025
    },
    {
      "entropy": 6.320016622543335,
      "epoch": 1.0383257690368128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999882377861355,
      "loss": 6.1998,
      "mean_token_accuracy": 0.11143601089715957,
      "num_tokens": 2276084.0,
      "step": 1030
    },
    {
      "entropy": 6.3366124629974365,
      "epoch": 1.043368633383762,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999838322534433,
      "loss": 6.3104,
      "mean_token_accuracy": 0.10776894018054009,
      "num_tokens": 2287364.0,
      "step": 1035
    },
    {
      "entropy": 6.361055326461792,
      "epoch": 1.048411497730711,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999787274658107,
      "loss": 6.165,
      "mean_token_accuracy": 0.11821243688464164,
      "num_tokens": 2298313.0,
      "step": 1040
    },
    {
      "entropy": 6.301340389251709,
      "epoch": 1.0534543620776602,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999729234391032,
      "loss": 6.1697,
      "mean_token_accuracy": 0.11393777951598168,
      "num_tokens": 2310126.0,
      "step": 1045
    },
    {
      "entropy": 6.267262697219849,
      "epoch": 1.058497226424609,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049996642019136,
      "loss": 6.1057,
      "mean_token_accuracy": 0.11893928498029709,
      "num_tokens": 2320830.0,
      "step": 1050
    },
    {
      "entropy": 6.188592100143433,
      "epoch": 1.0635400907715582,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004999592177427933,
      "loss": 6.1674,
      "mean_token_accuracy": 0.11705956310033798,
      "num_tokens": 2332027.0,
      "step": 1055
    },
    {
      "entropy": 6.273964786529541,
      "epoch": 1.0685829551185073,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999513161157885,
      "loss": 6.1881,
      "mean_token_accuracy": 0.11757974997162819,
      "num_tokens": 2342989.0,
      "step": 1060
    },
    {
      "entropy": 6.2867076873779295,
      "epoch": 1.0736258194654564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999427153349041,
      "loss": 6.1449,
      "mean_token_accuracy": 0.11460617780685425,
      "num_tokens": 2353321.0,
      "step": 1065
    },
    {
      "entropy": 6.249440431594849,
      "epoch": 1.0786686838124055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999334154268714,
      "loss": 6.1869,
      "mean_token_accuracy": 0.11248090788722039,
      "num_tokens": 2363910.0,
      "step": 1070
    },
    {
      "entropy": 6.243444728851318,
      "epoch": 1.0837115481593544,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999234164205949,
      "loss": 6.1837,
      "mean_token_accuracy": 0.11587119996547698,
      "num_tokens": 2374589.0,
      "step": 1075
    },
    {
      "entropy": 6.413973808288574,
      "epoch": 1.0887544125063036,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999127183471517,
      "loss": 6.2527,
      "mean_token_accuracy": 0.11490818783640862,
      "num_tokens": 2384684.0,
      "step": 1080
    },
    {
      "entropy": 6.217401218414307,
      "epoch": 1.0937972768532527,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999013212397915,
      "loss": 6.2538,
      "mean_token_accuracy": 0.1154061272740364,
      "num_tokens": 2396649.0,
      "step": 1085
    },
    {
      "entropy": 6.351813507080078,
      "epoch": 1.0988401412002018,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998892251339368,
      "loss": 6.1245,
      "mean_token_accuracy": 0.12135570496320724,
      "num_tokens": 2407696.0,
      "step": 1090
    },
    {
      "entropy": 6.272814846038818,
      "epoch": 1.1038830055471507,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998764300671827,
      "loss": 6.2113,
      "mean_token_accuracy": 0.11851643472909927,
      "num_tokens": 2418333.0,
      "step": 1095
    },
    {
      "entropy": 6.2627276420593265,
      "epoch": 1.1089258698940998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998629360792965,
      "loss": 6.2299,
      "mean_token_accuracy": 0.11417210474610329,
      "num_tokens": 2431081.0,
      "step": 1100
    },
    {
      "entropy": 6.3452956199646,
      "epoch": 1.113968734241049,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998487432122179,
      "loss": 6.1657,
      "mean_token_accuracy": 0.12272334843873978,
      "num_tokens": 2441055.0,
      "step": 1105
    },
    {
      "entropy": 6.271652364730835,
      "epoch": 1.119011598587998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998338515100585,
      "loss": 6.2095,
      "mean_token_accuracy": 0.11820752769708634,
      "num_tokens": 2451933.0,
      "step": 1110
    },
    {
      "entropy": 6.253003406524658,
      "epoch": 1.1240544629349472,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998182610191022,
      "loss": 6.2196,
      "mean_token_accuracy": 0.11624273434281349,
      "num_tokens": 2463062.0,
      "step": 1115
    },
    {
      "entropy": 6.2633141040802,
      "epoch": 1.129097327281896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998019717878047,
      "loss": 6.2218,
      "mean_token_accuracy": 0.11095208078622817,
      "num_tokens": 2473650.0,
      "step": 1120
    },
    {
      "entropy": 6.33440580368042,
      "epoch": 1.1341401916288452,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499784983866793,
      "loss": 6.2787,
      "mean_token_accuracy": 0.11275165677070617,
      "num_tokens": 2484510.0,
      "step": 1125
    },
    {
      "entropy": 6.183946895599365,
      "epoch": 1.1391830559757943,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997672973088661,
      "loss": 6.1521,
      "mean_token_accuracy": 0.11300064027309417,
      "num_tokens": 2495881.0,
      "step": 1130
    },
    {
      "entropy": 6.259727573394775,
      "epoch": 1.1442259203227434,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997489121689944,
      "loss": 6.1078,
      "mean_token_accuracy": 0.11746827661991119,
      "num_tokens": 2506657.0,
      "step": 1135
    },
    {
      "entropy": 6.106284093856812,
      "epoch": 1.1492687846696923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997298285043189,
      "loss": 6.121,
      "mean_token_accuracy": 0.12150057107210159,
      "num_tokens": 2518082.0,
      "step": 1140
    },
    {
      "entropy": 6.241512632369995,
      "epoch": 1.1543116490166414,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997100463741527,
      "loss": 6.0919,
      "mean_token_accuracy": 0.12284635081887245,
      "num_tokens": 2528722.0,
      "step": 1145
    },
    {
      "entropy": 6.2802904605865475,
      "epoch": 1.1593545133635905,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996895658399785,
      "loss": 6.0979,
      "mean_token_accuracy": 0.1160362608730793,
      "num_tokens": 2539008.0,
      "step": 1150
    },
    {
      "entropy": 6.154987907409668,
      "epoch": 1.1643973777105396,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996683869654507,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1161638431251049,
      "num_tokens": 2549445.0,
      "step": 1155
    },
    {
      "entropy": 6.135028886795044,
      "epoch": 1.1694402420574885,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996465098163934,
      "loss": 6.1583,
      "mean_token_accuracy": 0.11396335363388062,
      "num_tokens": 2561079.0,
      "step": 1160
    },
    {
      "entropy": 6.272411060333252,
      "epoch": 1.1744831064044376,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996239344608015,
      "loss": 6.0888,
      "mean_token_accuracy": 0.12005198225378991,
      "num_tokens": 2571540.0,
      "step": 1165
    },
    {
      "entropy": 6.181358337402344,
      "epoch": 1.1795259707513868,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996006609688398,
      "loss": 6.0589,
      "mean_token_accuracy": 0.11994132474064827,
      "num_tokens": 2582184.0,
      "step": 1170
    },
    {
      "entropy": 6.147908496856689,
      "epoch": 1.1845688350983359,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995766894128426,
      "loss": 6.1837,
      "mean_token_accuracy": 0.12015488147735595,
      "num_tokens": 2592771.0,
      "step": 1175
    },
    {
      "entropy": 6.2181120872497555,
      "epoch": 1.189611699445285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995520198673143,
      "loss": 6.0737,
      "mean_token_accuracy": 0.11993672326207161,
      "num_tokens": 2602974.0,
      "step": 1180
    },
    {
      "entropy": 6.233900117874145,
      "epoch": 1.194654563792234,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995266524089283,
      "loss": 6.06,
      "mean_token_accuracy": 0.12310326918959617,
      "num_tokens": 2613915.0,
      "step": 1185
    },
    {
      "entropy": 6.168328189849854,
      "epoch": 1.199697428139183,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004995005871165271,
      "loss": 6.1496,
      "mean_token_accuracy": 0.1177591860294342,
      "num_tokens": 2626038.0,
      "step": 1190
    },
    {
      "entropy": 6.194040536880493,
      "epoch": 1.2047402924861321,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994738240711225,
      "loss": 6.1042,
      "mean_token_accuracy": 0.12066976353526115,
      "num_tokens": 2636448.0,
      "step": 1195
    },
    {
      "entropy": 6.251908779144287,
      "epoch": 1.2097831568330812,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994463633558945,
      "loss": 6.1974,
      "mean_token_accuracy": 0.11828442364931106,
      "num_tokens": 2648803.0,
      "step": 1200
    },
    {
      "entropy": 6.1560954570770265,
      "epoch": 1.2148260211800301,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994182050561917,
      "loss": 6.0962,
      "mean_token_accuracy": 0.1210496187210083,
      "num_tokens": 2659981.0,
      "step": 1205
    },
    {
      "entropy": 6.110389518737793,
      "epoch": 1.2198688855269793,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993893492595306,
      "loss": 6.0856,
      "mean_token_accuracy": 0.12283521145582199,
      "num_tokens": 2670340.0,
      "step": 1210
    },
    {
      "entropy": 6.237795495986939,
      "epoch": 1.2249117498739284,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993597960555959,
      "loss": 6.1174,
      "mean_token_accuracy": 0.12508749216794968,
      "num_tokens": 2681147.0,
      "step": 1215
    },
    {
      "entropy": 6.172670507431031,
      "epoch": 1.2299546142208775,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993295455362395,
      "loss": 6.0388,
      "mean_token_accuracy": 0.1264190822839737,
      "num_tokens": 2693877.0,
      "step": 1220
    },
    {
      "entropy": 6.102581548690796,
      "epoch": 1.2349974785678266,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992985977954808,
      "loss": 6.0757,
      "mean_token_accuracy": 0.11901459470391273,
      "num_tokens": 2704371.0,
      "step": 1225
    },
    {
      "entropy": 6.239574670791626,
      "epoch": 1.2400403429147757,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499266952929506,
      "loss": 6.1372,
      "mean_token_accuracy": 0.11878894343972206,
      "num_tokens": 2714549.0,
      "step": 1230
    },
    {
      "entropy": 6.153753709793091,
      "epoch": 1.2450832072617246,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992346110366682,
      "loss": 6.0684,
      "mean_token_accuracy": 0.12529382407665252,
      "num_tokens": 2723750.0,
      "step": 1235
    },
    {
      "entropy": 6.140466403961182,
      "epoch": 1.2501260716086737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992015722174865,
      "loss": 6.1064,
      "mean_token_accuracy": 0.11989799737930298,
      "num_tokens": 2734557.0,
      "step": 1240
    },
    {
      "entropy": 6.191011524200439,
      "epoch": 1.2551689359556228,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991678365746466,
      "loss": 6.1197,
      "mean_token_accuracy": 0.12287942469120025,
      "num_tokens": 2746521.0,
      "step": 1245
    },
    {
      "entropy": 6.118342781066895,
      "epoch": 1.2602118003025717,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991334042129996,
      "loss": 6.1396,
      "mean_token_accuracy": 0.12459613531827926,
      "num_tokens": 2758074.0,
      "step": 1250
    },
    {
      "entropy": 6.138072919845581,
      "epoch": 1.2652546646495209,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990982752395618,
      "loss": 5.9861,
      "mean_token_accuracy": 0.12695645466446875,
      "num_tokens": 2770059.0,
      "step": 1255
    },
    {
      "entropy": 6.090990257263184,
      "epoch": 1.27029752899647,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990624497635152,
      "loss": 6.0559,
      "mean_token_accuracy": 0.12386075705289841,
      "num_tokens": 2780802.0,
      "step": 1260
    },
    {
      "entropy": 6.221565437316895,
      "epoch": 1.275340393343419,
      "grad_norm": 0.890625,
      "learning_rate": 0.000499025927896206,
      "loss": 6.1523,
      "mean_token_accuracy": 0.12033093869686126,
      "num_tokens": 2793693.0,
      "step": 1265
    },
    {
      "entropy": 6.183721590042114,
      "epoch": 1.2803832576903682,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989887097511451,
      "loss": 6.1682,
      "mean_token_accuracy": 0.11709739565849304,
      "num_tokens": 2805547.0,
      "step": 1270
    },
    {
      "entropy": 6.118980169296265,
      "epoch": 1.2854261220373173,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989507954440074,
      "loss": 6.0879,
      "mean_token_accuracy": 0.117892437428236,
      "num_tokens": 2816207.0,
      "step": 1275
    },
    {
      "entropy": 6.165093946456909,
      "epoch": 1.2904689863842662,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989121850926313,
      "loss": 6.0669,
      "mean_token_accuracy": 0.11663686856627464,
      "num_tokens": 2827272.0,
      "step": 1280
    },
    {
      "entropy": 6.1378819942474365,
      "epoch": 1.2955118507312153,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004988728788170188,
      "loss": 6.0369,
      "mean_token_accuracy": 0.12149282097816468,
      "num_tokens": 2839814.0,
      "step": 1285
    },
    {
      "entropy": 6.1398155212402346,
      "epoch": 1.3005547150781644,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988328767393346,
      "loss": 6.1763,
      "mean_token_accuracy": 0.11872457265853882,
      "num_tokens": 2851629.0,
      "step": 1290
    },
    {
      "entropy": 6.199816131591797,
      "epoch": 1.3055975794251133,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987921789839061,
      "loss": 6.1082,
      "mean_token_accuracy": 0.12454739883542061,
      "num_tokens": 2863518.0,
      "step": 1295
    },
    {
      "entropy": 6.043096208572388,
      "epoch": 1.3106404437720625,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987507856772229,
      "loss": 6.0624,
      "mean_token_accuracy": 0.1212753787636757,
      "num_tokens": 2874803.0,
      "step": 1300
    },
    {
      "entropy": 6.212449645996093,
      "epoch": 1.3156833081190116,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987086969479364,
      "loss": 6.05,
      "mean_token_accuracy": 0.12336436361074447,
      "num_tokens": 2886052.0,
      "step": 1305
    },
    {
      "entropy": 6.058530807495117,
      "epoch": 1.3207261724659607,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986659129268592,
      "loss": 6.0717,
      "mean_token_accuracy": 0.12124264538288117,
      "num_tokens": 2897797.0,
      "step": 1310
    },
    {
      "entropy": 6.182499027252197,
      "epoch": 1.3257690368129098,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986224337469652,
      "loss": 6.0392,
      "mean_token_accuracy": 0.11823606640100479,
      "num_tokens": 2908752.0,
      "step": 1315
    },
    {
      "entropy": 6.041653299331665,
      "epoch": 1.330811901159859,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985782595433886,
      "loss": 6.1067,
      "mean_token_accuracy": 0.11841807141900063,
      "num_tokens": 2919351.0,
      "step": 1320
    },
    {
      "entropy": 6.149489307403565,
      "epoch": 1.3358547655068078,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985333904534237,
      "loss": 6.0661,
      "mean_token_accuracy": 0.12007037401199341,
      "num_tokens": 2930198.0,
      "step": 1325
    },
    {
      "entropy": 6.100072622299194,
      "epoch": 1.340897629853757,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000498487826616525,
      "loss": 6.0666,
      "mean_token_accuracy": 0.12396523952484131,
      "num_tokens": 2941349.0,
      "step": 1330
    },
    {
      "entropy": 6.110146522521973,
      "epoch": 1.345940494200706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984415681743056,
      "loss": 5.9878,
      "mean_token_accuracy": 0.12134144976735114,
      "num_tokens": 2951740.0,
      "step": 1335
    },
    {
      "entropy": 6.1147205352783205,
      "epoch": 1.350983358547655,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983946152705382,
      "loss": 6.0311,
      "mean_token_accuracy": 0.12353800013661384,
      "num_tokens": 2963900.0,
      "step": 1340
    },
    {
      "entropy": 6.1047674179077145,
      "epoch": 1.356026222894604,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983469680511531,
      "loss": 6.0708,
      "mean_token_accuracy": 0.11983567550778389,
      "num_tokens": 2975385.0,
      "step": 1345
    },
    {
      "entropy": 6.116179466247559,
      "epoch": 1.3610690872415532,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004982986266642393,
      "loss": 6.0633,
      "mean_token_accuracy": 0.12223733887076378,
      "num_tokens": 2986340.0,
      "step": 1350
    },
    {
      "entropy": 6.080137825012207,
      "epoch": 1.3661119515885023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004982495912600426,
      "loss": 5.9413,
      "mean_token_accuracy": 0.128122878074646,
      "num_tokens": 2996582.0,
      "step": 1355
    },
    {
      "entropy": 6.022544431686401,
      "epoch": 1.3711548159354514,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004981998619909664,
      "loss": 5.9682,
      "mean_token_accuracy": 0.12417263835668564,
      "num_tokens": 3007620.0,
      "step": 1360
    },
    {
      "entropy": 6.088971042633057,
      "epoch": 1.3761976802824005,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004981494390115704,
      "loss": 6.0683,
      "mean_token_accuracy": 0.12460343763232232,
      "num_tokens": 3019520.0,
      "step": 1365
    },
    {
      "entropy": 6.027232551574707,
      "epoch": 1.3812405446293494,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980983224785704,
      "loss": 6.0585,
      "mean_token_accuracy": 0.11659184172749519,
      "num_tokens": 3030616.0,
      "step": 1370
    },
    {
      "entropy": 6.165927600860596,
      "epoch": 1.3862834089762985,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004980465125508376,
      "loss": 6.1292,
      "mean_token_accuracy": 0.11850353851914405,
      "num_tokens": 3042383.0,
      "step": 1375
    },
    {
      "entropy": 6.158199501037598,
      "epoch": 1.3913262733232477,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979940093893986,
      "loss": 6.0277,
      "mean_token_accuracy": 0.1222493290901184,
      "num_tokens": 3052833.0,
      "step": 1380
    },
    {
      "entropy": 6.009306812286377,
      "epoch": 1.3963691376701965,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979408131574345,
      "loss": 6.0017,
      "mean_token_accuracy": 0.12259870544075965,
      "num_tokens": 3063800.0,
      "step": 1385
    },
    {
      "entropy": 6.128568363189697,
      "epoch": 1.4014120020171457,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004978869240202802,
      "loss": 6.1616,
      "mean_token_accuracy": 0.12267429605126381,
      "num_tokens": 3075243.0,
      "step": 1390
    },
    {
      "entropy": 6.081125736236572,
      "epoch": 1.4064548663640948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978323421454246,
      "loss": 6.001,
      "mean_token_accuracy": 0.1273365207016468,
      "num_tokens": 3086590.0,
      "step": 1395
    },
    {
      "entropy": 6.090214252471924,
      "epoch": 1.411497730711044,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977770677025092,
      "loss": 6.0373,
      "mean_token_accuracy": 0.1295775778591633,
      "num_tokens": 3097554.0,
      "step": 1400
    },
    {
      "entropy": 6.092472696304322,
      "epoch": 1.416540595057993,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004977211008633284,
      "loss": 6.057,
      "mean_token_accuracy": 0.12078494280576706,
      "num_tokens": 3110051.0,
      "step": 1405
    },
    {
      "entropy": 6.035629320144653,
      "epoch": 1.4215834594049421,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004976644418018285,
      "loss": 5.9504,
      "mean_token_accuracy": 0.12921829596161843,
      "num_tokens": 3120640.0,
      "step": 1410
    },
    {
      "entropy": 6.042394065856934,
      "epoch": 1.426626323751891,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497607090694107,
      "loss": 5.9583,
      "mean_token_accuracy": 0.12378115579485893,
      "num_tokens": 3131293.0,
      "step": 1415
    },
    {
      "entropy": 5.968035554885864,
      "epoch": 1.4316691880988401,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004975490477184126,
      "loss": 6.0001,
      "mean_token_accuracy": 0.12400781959295273,
      "num_tokens": 3141272.0,
      "step": 1420
    },
    {
      "entropy": 6.159467649459839,
      "epoch": 1.4367120524457893,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497490313055144,
      "loss": 5.9489,
      "mean_token_accuracy": 0.12632449343800545,
      "num_tokens": 3151349.0,
      "step": 1425
    },
    {
      "entropy": 5.933459615707397,
      "epoch": 1.4417549167927382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004974308868868501,
      "loss": 6.0528,
      "mean_token_accuracy": 0.12241479903459548,
      "num_tokens": 3162258.0,
      "step": 1430
    },
    {
      "entropy": 6.177392387390137,
      "epoch": 1.4467977811396873,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973707693982287,
      "loss": 6.0687,
      "mean_token_accuracy": 0.12557331547141076,
      "num_tokens": 3172821.0,
      "step": 1435
    },
    {
      "entropy": 6.031765031814575,
      "epoch": 1.4518406454866364,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973099607761264,
      "loss": 5.9993,
      "mean_token_accuracy": 0.12058438062667846,
      "num_tokens": 3183720.0,
      "step": 1440
    },
    {
      "entropy": 6.232817029953003,
      "epoch": 1.4568835098335855,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972484612095378,
      "loss": 6.0611,
      "mean_token_accuracy": 0.12563046663999558,
      "num_tokens": 3192882.0,
      "step": 1445
    },
    {
      "entropy": 6.00401120185852,
      "epoch": 1.4619263741805346,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497186270889605,
      "loss": 6.0029,
      "mean_token_accuracy": 0.12171068787574768,
      "num_tokens": 3203908.0,
      "step": 1450
    },
    {
      "entropy": 6.086980676651001,
      "epoch": 1.4669692385274837,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004971233900096169,
      "loss": 6.0922,
      "mean_token_accuracy": 0.11998873427510262,
      "num_tokens": 3215145.0,
      "step": 1455
    },
    {
      "entropy": 6.075078773498535,
      "epoch": 1.4720121028744326,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004970598187650085,
      "loss": 5.9639,
      "mean_token_accuracy": 0.13046645373106003,
      "num_tokens": 3225714.0,
      "step": 1460
    },
    {
      "entropy": 6.057655906677246,
      "epoch": 1.4770549672213817,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000496995557353361,
      "loss": 6.0777,
      "mean_token_accuracy": 0.12587454319000244,
      "num_tokens": 3238541.0,
      "step": 1465
    },
    {
      "entropy": 5.956527376174927,
      "epoch": 1.4820978315683309,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969306059744004,
      "loss": 5.8903,
      "mean_token_accuracy": 0.1326131209731102,
      "num_tokens": 3250219.0,
      "step": 1470
    },
    {
      "entropy": 5.961444568634033,
      "epoch": 1.4871406959152798,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004968649648299968,
      "loss": 5.9488,
      "mean_token_accuracy": 0.12518008425831795,
      "num_tokens": 3261676.0,
      "step": 1475
    },
    {
      "entropy": 6.040499210357666,
      "epoch": 1.4921835602622289,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004967986341241646,
      "loss": 6.0356,
      "mean_token_accuracy": 0.11670267432928086,
      "num_tokens": 3272862.0,
      "step": 1480
    },
    {
      "entropy": 5.975739526748657,
      "epoch": 1.497226424609178,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004967316140630611,
      "loss": 5.9459,
      "mean_token_accuracy": 0.12727636694908143,
      "num_tokens": 3282891.0,
      "step": 1485
    },
    {
      "entropy": 5.996143436431884,
      "epoch": 1.502269288956127,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004966639048549862,
      "loss": 5.8887,
      "mean_token_accuracy": 0.12851305529475213,
      "num_tokens": 3293105.0,
      "step": 1490
    },
    {
      "entropy": 6.014690017700195,
      "epoch": 1.5073121533030762,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004965955067103816,
      "loss": 5.959,
      "mean_token_accuracy": 0.13141607046127318,
      "num_tokens": 3304431.0,
      "step": 1495
    },
    {
      "entropy": 6.129389524459839,
      "epoch": 1.5123550176500253,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965264198418305,
      "loss": 6.0494,
      "mean_token_accuracy": 0.12193582206964493,
      "num_tokens": 3314391.0,
      "step": 1500
    },
    {
      "epoch": 1.5123550176500253,
      "eval_entropy": 5.9836117936151325,
      "eval_loss": 6.151969909667969,
      "eval_mean_token_accuracy": 0.12492486130228177,
      "eval_num_tokens": 3314391.0,
      "eval_runtime": 2.6419,
      "eval_samples_per_second": 1187.797,
      "eval_steps_per_second": 148.759,
      "step": 1500
    },
    {
      "entropy": 6.002770566940308,
      "epoch": 1.5173978819969742,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964566444640564,
      "loss": 5.9062,
      "mean_token_accuracy": 0.1287309117615223,
      "num_tokens": 3324578.0,
      "step": 1505
    },
    {
      "entropy": 5.993435716629028,
      "epoch": 1.5224407463439233,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004963861807939227,
      "loss": 5.9685,
      "mean_token_accuracy": 0.13316882848739625,
      "num_tokens": 3334724.0,
      "step": 1510
    },
    {
      "entropy": 6.05625205039978,
      "epoch": 1.5274836106908725,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004963150290504321,
      "loss": 6.0687,
      "mean_token_accuracy": 0.1278290867805481,
      "num_tokens": 3345371.0,
      "step": 1515
    },
    {
      "entropy": 6.040426445007324,
      "epoch": 1.5325264750378214,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004962431894547259,
      "loss": 5.9856,
      "mean_token_accuracy": 0.12599704489111901,
      "num_tokens": 3357176.0,
      "step": 1520
    },
    {
      "entropy": 5.967979764938354,
      "epoch": 1.5375693393847705,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004961706622300834,
      "loss": 6.0129,
      "mean_token_accuracy": 0.11759765893220901,
      "num_tokens": 3368213.0,
      "step": 1525
    },
    {
      "entropy": 6.0974396705627445,
      "epoch": 1.5426122037317196,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960974476019206,
      "loss": 5.9772,
      "mean_token_accuracy": 0.12633119896054268,
      "num_tokens": 3379676.0,
      "step": 1530
    },
    {
      "entropy": 5.973466157913208,
      "epoch": 1.5476550680786687,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004960235457977905,
      "loss": 5.9103,
      "mean_token_accuracy": 0.12467504814267158,
      "num_tokens": 3390765.0,
      "step": 1535
    },
    {
      "entropy": 6.052669954299927,
      "epoch": 1.5526979324256178,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004959489570473814,
      "loss": 5.9436,
      "mean_token_accuracy": 0.1279728040099144,
      "num_tokens": 3401804.0,
      "step": 1540
    },
    {
      "entropy": 6.094581937789917,
      "epoch": 1.557740796772567,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000495873681582517,
      "loss": 6.0648,
      "mean_token_accuracy": 0.1282322771847248,
      "num_tokens": 3413509.0,
      "step": 1545
    },
    {
      "entropy": 5.933273983001709,
      "epoch": 1.5627836611195158,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957977196371552,
      "loss": 5.9076,
      "mean_token_accuracy": 0.13429276049137115,
      "num_tokens": 3423807.0,
      "step": 1550
    },
    {
      "entropy": 5.946268749237061,
      "epoch": 1.567826525466465,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957210714473873,
      "loss": 5.88,
      "mean_token_accuracy": 0.12932645455002784,
      "num_tokens": 3434214.0,
      "step": 1555
    },
    {
      "entropy": 6.094040393829346,
      "epoch": 1.572869389813414,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004956437372514378,
      "loss": 5.9502,
      "mean_token_accuracy": 0.12896995097398758,
      "num_tokens": 3445081.0,
      "step": 1560
    },
    {
      "entropy": 5.898585033416748,
      "epoch": 1.577912254160363,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955657172896631,
      "loss": 5.9275,
      "mean_token_accuracy": 0.13340998068451881,
      "num_tokens": 3454992.0,
      "step": 1565
    },
    {
      "entropy": 5.992181396484375,
      "epoch": 1.582955118507312,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954870118045511,
      "loss": 6.0237,
      "mean_token_accuracy": 0.11989426463842393,
      "num_tokens": 3465635.0,
      "step": 1570
    },
    {
      "entropy": 5.947319221496582,
      "epoch": 1.5879979828542612,
      "grad_norm": 0.875,
      "learning_rate": 0.0004954076210407203,
      "loss": 5.9283,
      "mean_token_accuracy": 0.13070602416992189,
      "num_tokens": 3477746.0,
      "step": 1575
    },
    {
      "entropy": 6.0351625919342045,
      "epoch": 1.5930408472012103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953275452449188,
      "loss": 5.9775,
      "mean_token_accuracy": 0.12918647080659867,
      "num_tokens": 3488915.0,
      "step": 1580
    },
    {
      "entropy": 5.954270458221435,
      "epoch": 1.5980837115481594,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004952467846660243,
      "loss": 5.9517,
      "mean_token_accuracy": 0.12838053107261657,
      "num_tokens": 3499096.0,
      "step": 1585
    },
    {
      "entropy": 5.958280849456787,
      "epoch": 1.6031265758951085,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004951653395550423,
      "loss": 5.8424,
      "mean_token_accuracy": 0.12435204163193703,
      "num_tokens": 3512544.0,
      "step": 1590
    },
    {
      "entropy": 5.980319499969482,
      "epoch": 1.6081694402420574,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004950832101651062,
      "loss": 5.9792,
      "mean_token_accuracy": 0.1283943310379982,
      "num_tokens": 3522814.0,
      "step": 1595
    },
    {
      "entropy": 5.901864099502563,
      "epoch": 1.6132123045890066,
      "grad_norm": 0.875,
      "learning_rate": 0.0004950003967514762,
      "loss": 5.9033,
      "mean_token_accuracy": 0.13048650473356246,
      "num_tokens": 3533724.0,
      "step": 1600
    },
    {
      "entropy": 6.041346263885498,
      "epoch": 1.6182551689359557,
      "grad_norm": 0.984375,
      "learning_rate": 0.000494916899571538,
      "loss": 5.9635,
      "mean_token_accuracy": 0.1256171390414238,
      "num_tokens": 3544777.0,
      "step": 1605
    },
    {
      "entropy": 6.030082273483276,
      "epoch": 1.6232980332829046,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004948327188848029,
      "loss": 5.9863,
      "mean_token_accuracy": 0.1326717607676983,
      "num_tokens": 3556397.0,
      "step": 1610
    },
    {
      "entropy": 5.949783515930176,
      "epoch": 1.6283408976298537,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004947478549529065,
      "loss": 5.9397,
      "mean_token_accuracy": 0.12490741759538651,
      "num_tokens": 3568320.0,
      "step": 1615
    },
    {
      "entropy": 5.998614501953125,
      "epoch": 1.6333837619768028,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004946623080396077,
      "loss": 5.9348,
      "mean_token_accuracy": 0.12845998406410217,
      "num_tokens": 3578289.0,
      "step": 1620
    },
    {
      "entropy": 5.963990306854248,
      "epoch": 1.638426626323752,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004945760784107884,
      "loss": 5.9822,
      "mean_token_accuracy": 0.12893664538860322,
      "num_tokens": 3589710.0,
      "step": 1625
    },
    {
      "entropy": 6.021360588073731,
      "epoch": 1.643469490670701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004944891663344522,
      "loss": 5.9729,
      "mean_token_accuracy": 0.12624122723937034,
      "num_tokens": 3601618.0,
      "step": 1630
    },
    {
      "entropy": 6.045843553543091,
      "epoch": 1.6485123550176501,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004944015720807239,
      "loss": 6.0503,
      "mean_token_accuracy": 0.1247016966342926,
      "num_tokens": 3612329.0,
      "step": 1635
    },
    {
      "entropy": 6.013518476486206,
      "epoch": 1.653555219364599,
      "grad_norm": 0.875,
      "learning_rate": 0.0004943132959218484,
      "loss": 6.0217,
      "mean_token_accuracy": 0.12285120636224747,
      "num_tokens": 3623887.0,
      "step": 1640
    },
    {
      "entropy": 5.96614294052124,
      "epoch": 1.6585980837115482,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00049422433813219,
      "loss": 5.9486,
      "mean_token_accuracy": 0.12361961752176284,
      "num_tokens": 3635484.0,
      "step": 1645
    },
    {
      "entropy": 5.967590427398681,
      "epoch": 1.6636409480584973,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941346989882318,
      "loss": 5.8444,
      "mean_token_accuracy": 0.12881785854697228,
      "num_tokens": 3645802.0,
      "step": 1650
    },
    {
      "entropy": 5.977982473373413,
      "epoch": 1.6686838124054462,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004940443787685739,
      "loss": 5.9296,
      "mean_token_accuracy": 0.12940604165196418,
      "num_tokens": 3655599.0,
      "step": 1655
    },
    {
      "entropy": 6.026707506179809,
      "epoch": 1.6737266767523953,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004939533777539341,
      "loss": 5.9475,
      "mean_token_accuracy": 0.12480253130197524,
      "num_tokens": 3667019.0,
      "step": 1660
    },
    {
      "entropy": 5.816624021530151,
      "epoch": 1.6787695410993444,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004938616962271453,
      "loss": 5.8174,
      "mean_token_accuracy": 0.13443900346755983,
      "num_tokens": 3678004.0,
      "step": 1665
    },
    {
      "entropy": 5.990376710891724,
      "epoch": 1.6838124054462935,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004937693344731561,
      "loss": 5.8891,
      "mean_token_accuracy": 0.1256047248840332,
      "num_tokens": 3688177.0,
      "step": 1670
    },
    {
      "entropy": 5.9272315979003904,
      "epoch": 1.6888552697932426,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004936762927790286,
      "loss": 5.923,
      "mean_token_accuracy": 0.1307574525475502,
      "num_tokens": 3700375.0,
      "step": 1675
    },
    {
      "entropy": 5.980884552001953,
      "epoch": 1.6938981341401917,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004935825714339389,
      "loss": 5.9813,
      "mean_token_accuracy": 0.1269807793200016,
      "num_tokens": 3711629.0,
      "step": 1680
    },
    {
      "entropy": 5.915404605865478,
      "epoch": 1.6989409984871406,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004934881707291748,
      "loss": 5.8829,
      "mean_token_accuracy": 0.12866655737161636,
      "num_tokens": 3723570.0,
      "step": 1685
    },
    {
      "entropy": 5.964293813705444,
      "epoch": 1.7039838628340898,
      "grad_norm": 1.0,
      "learning_rate": 0.0004933930909581362,
      "loss": 5.8167,
      "mean_token_accuracy": 0.13202881217002868,
      "num_tokens": 3733687.0,
      "step": 1690
    },
    {
      "entropy": 5.902900743484497,
      "epoch": 1.7090267271810389,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004932973324163328,
      "loss": 5.8501,
      "mean_token_accuracy": 0.13000352382659913,
      "num_tokens": 3744455.0,
      "step": 1695
    },
    {
      "entropy": 5.912310600280762,
      "epoch": 1.7140695915279878,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004932008954013846,
      "loss": 5.8379,
      "mean_token_accuracy": 0.12738796770572663,
      "num_tokens": 3756079.0,
      "step": 1700
    },
    {
      "entropy": 5.903970384597779,
      "epoch": 1.7191124558749369,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004931037802130201,
      "loss": 5.8167,
      "mean_token_accuracy": 0.13375495001673698,
      "num_tokens": 3766548.0,
      "step": 1705
    },
    {
      "entropy": 5.9737762928009035,
      "epoch": 1.724155320221886,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004930059871530755,
      "loss": 5.9092,
      "mean_token_accuracy": 0.13129813969135284,
      "num_tokens": 3776482.0,
      "step": 1710
    },
    {
      "entropy": 5.92631573677063,
      "epoch": 1.7291981845688351,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004929075165254938,
      "loss": 5.8892,
      "mean_token_accuracy": 0.1338859997689724,
      "num_tokens": 3787030.0,
      "step": 1715
    },
    {
      "entropy": 5.911603832244873,
      "epoch": 1.7342410489157842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004928083686363241,
      "loss": 5.8502,
      "mean_token_accuracy": 0.12917665019631386,
      "num_tokens": 3797575.0,
      "step": 1720
    },
    {
      "entropy": 5.90240626335144,
      "epoch": 1.7392839132627333,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004927085437937204,
      "loss": 5.8342,
      "mean_token_accuracy": 0.13536562025547028,
      "num_tokens": 3807634.0,
      "step": 1725
    },
    {
      "entropy": 5.942410898208618,
      "epoch": 1.7443267776096822,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004926080423079404,
      "loss": 6.0215,
      "mean_token_accuracy": 0.12834933772683144,
      "num_tokens": 3820691.0,
      "step": 1730
    },
    {
      "entropy": 5.9908922672271725,
      "epoch": 1.7493696419566314,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000492506864491345,
      "loss": 6.0174,
      "mean_token_accuracy": 0.12165129706263542,
      "num_tokens": 3832434.0,
      "step": 1735
    },
    {
      "entropy": 5.99032940864563,
      "epoch": 1.7544125063035805,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004924050106583976,
      "loss": 5.9201,
      "mean_token_accuracy": 0.13164895102381707,
      "num_tokens": 3842547.0,
      "step": 1740
    },
    {
      "entropy": 5.9721756935119625,
      "epoch": 1.7594553706505294,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004923024811256618,
      "loss": 5.9257,
      "mean_token_accuracy": 0.12628312036395073,
      "num_tokens": 3852830.0,
      "step": 1745
    },
    {
      "entropy": 5.966777801513672,
      "epoch": 1.7644982349974785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492199276211802,
      "loss": 5.9681,
      "mean_token_accuracy": 0.12151766121387482,
      "num_tokens": 3863396.0,
      "step": 1750
    },
    {
      "entropy": 5.879304456710815,
      "epoch": 1.7695410993444276,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004920953962375813,
      "loss": 5.8456,
      "mean_token_accuracy": 0.12973029911518097,
      "num_tokens": 3874380.0,
      "step": 1755
    },
    {
      "entropy": 5.922623920440674,
      "epoch": 1.7745839636913767,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919908415258612,
      "loss": 5.8587,
      "mean_token_accuracy": 0.1308340273797512,
      "num_tokens": 3883454.0,
      "step": 1760
    },
    {
      "entropy": 5.874415349960327,
      "epoch": 1.7796268280383258,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004918856124016002,
      "loss": 5.8596,
      "mean_token_accuracy": 0.1265338569879532,
      "num_tokens": 3893419.0,
      "step": 1765
    },
    {
      "entropy": 5.949326276779175,
      "epoch": 1.784669692385275,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004917797091918526,
      "loss": 5.9122,
      "mean_token_accuracy": 0.12785057127475738,
      "num_tokens": 3903760.0,
      "step": 1770
    },
    {
      "entropy": 5.9434820175170895,
      "epoch": 1.789712556732224,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004916731322257683,
      "loss": 5.9561,
      "mean_token_accuracy": 0.13075582906603814,
      "num_tokens": 3914881.0,
      "step": 1775
    },
    {
      "entropy": 6.076316213607788,
      "epoch": 1.794755421079173,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004915658818345908,
      "loss": 5.9866,
      "mean_token_accuracy": 0.12632244527339936,
      "num_tokens": 3925774.0,
      "step": 1780
    },
    {
      "entropy": 5.857404661178589,
      "epoch": 1.799798285426122,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004914579583516568,
      "loss": 5.8635,
      "mean_token_accuracy": 0.13342424854636192,
      "num_tokens": 3936357.0,
      "step": 1785
    },
    {
      "entropy": 6.05221209526062,
      "epoch": 1.804841149773071,
      "grad_norm": 0.984375,
      "learning_rate": 0.000491349362112395,
      "loss": 5.9406,
      "mean_token_accuracy": 0.130137200653553,
      "num_tokens": 3946493.0,
      "step": 1790
    },
    {
      "entropy": 5.943054580688477,
      "epoch": 1.80988401412002,
      "grad_norm": 1.234375,
      "learning_rate": 0.000491240093454325,
      "loss": 5.918,
      "mean_token_accuracy": 0.12943230643868447,
      "num_tokens": 3958022.0,
      "step": 1795
    },
    {
      "entropy": 5.880387878417968,
      "epoch": 1.8149268784669692,
      "grad_norm": 1.0,
      "learning_rate": 0.0004911301527170563,
      "loss": 5.74,
      "mean_token_accuracy": 0.13886643648147584,
      "num_tokens": 3968033.0,
      "step": 1800
    },
    {
      "entropy": 5.889808988571167,
      "epoch": 1.8199697428139183,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004910195402422873,
      "loss": 5.9123,
      "mean_token_accuracy": 0.1269322395324707,
      "num_tokens": 3978319.0,
      "step": 1805
    },
    {
      "entropy": 5.946580696105957,
      "epoch": 1.8250126071608674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909082563738042,
      "loss": 5.8572,
      "mean_token_accuracy": 0.1318647786974907,
      "num_tokens": 3989519.0,
      "step": 1810
    },
    {
      "entropy": 5.929449892044067,
      "epoch": 1.8300554715078166,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004907963014574797,
      "loss": 5.943,
      "mean_token_accuracy": 0.13093527406454086,
      "num_tokens": 4000680.0,
      "step": 1815
    },
    {
      "entropy": 5.897009038925171,
      "epoch": 1.8350983358547657,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004906836758412722,
      "loss": 5.8782,
      "mean_token_accuracy": 0.13037897050380706,
      "num_tokens": 4012886.0,
      "step": 1820
    },
    {
      "entropy": 5.903923892974854,
      "epoch": 1.8401412002017146,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004905703798752248,
      "loss": 5.928,
      "mean_token_accuracy": 0.1296511970460415,
      "num_tokens": 4023690.0,
      "step": 1825
    },
    {
      "entropy": 5.838856601715088,
      "epoch": 1.8451840645486637,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000490456413911464,
      "loss": 5.7606,
      "mean_token_accuracy": 0.13645405173301697,
      "num_tokens": 4033338.0,
      "step": 1830
    },
    {
      "entropy": 5.848793077468872,
      "epoch": 1.8502269288956126,
      "grad_norm": 1.0,
      "learning_rate": 0.0004903417783041987,
      "loss": 5.7763,
      "mean_token_accuracy": 0.1365301750600338,
      "num_tokens": 4042873.0,
      "step": 1835
    },
    {
      "entropy": 5.928408908843994,
      "epoch": 1.8552697932425617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004902264734097188,
      "loss": 5.9228,
      "mean_token_accuracy": 0.13415781110525132,
      "num_tokens": 4054049.0,
      "step": 1840
    },
    {
      "entropy": 5.961584663391113,
      "epoch": 1.8603126575895108,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004901104995863948,
      "loss": 5.9105,
      "mean_token_accuracy": 0.12920038104057313,
      "num_tokens": 4064254.0,
      "step": 1845
    },
    {
      "entropy": 5.959888362884522,
      "epoch": 1.86535552193646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004899938571946759,
      "loss": 5.9088,
      "mean_token_accuracy": 0.13008437752723695,
      "num_tokens": 4075235.0,
      "step": 1850
    },
    {
      "entropy": 6.01555323600769,
      "epoch": 1.870398386283409,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004898765465970894,
      "loss": 6.0332,
      "mean_token_accuracy": 0.12419131100177765,
      "num_tokens": 4087434.0,
      "step": 1855
    },
    {
      "entropy": 5.92230715751648,
      "epoch": 1.8754412506303582,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897585681582393,
      "loss": 5.8621,
      "mean_token_accuracy": 0.1299331970512867,
      "num_tokens": 4099223.0,
      "step": 1860
    },
    {
      "entropy": 5.86011061668396,
      "epoch": 1.8804841149773073,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004896399222448053,
      "loss": 5.8249,
      "mean_token_accuracy": 0.1367386259138584,
      "num_tokens": 4110706.0,
      "step": 1865
    },
    {
      "entropy": 6.009511661529541,
      "epoch": 1.8855269793242562,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895206092255417,
      "loss": 5.9138,
      "mean_token_accuracy": 0.12570383846759797,
      "num_tokens": 4122358.0,
      "step": 1870
    },
    {
      "entropy": 5.795359373092651,
      "epoch": 1.8905698436712053,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004894006294712761,
      "loss": 5.6944,
      "mean_token_accuracy": 0.1380500726401806,
      "num_tokens": 4132603.0,
      "step": 1875
    },
    {
      "entropy": 5.757789611816406,
      "epoch": 1.8956127080181542,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004892799833549083,
      "loss": 5.8502,
      "mean_token_accuracy": 0.13927022740244865,
      "num_tokens": 4144708.0,
      "step": 1880
    },
    {
      "entropy": 5.990426540374756,
      "epoch": 1.9006555723651033,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004891586712514091,
      "loss": 5.8796,
      "mean_token_accuracy": 0.13205204233527185,
      "num_tokens": 4155751.0,
      "step": 1885
    },
    {
      "entropy": 5.84541425704956,
      "epoch": 1.9056984367120524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004890366935378196,
      "loss": 5.8613,
      "mean_token_accuracy": 0.13200437948107718,
      "num_tokens": 4167368.0,
      "step": 1890
    },
    {
      "entropy": 5.916485404968261,
      "epoch": 1.9107413010590015,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004889140505932492,
      "loss": 5.7987,
      "mean_token_accuracy": 0.1378105029463768,
      "num_tokens": 4178601.0,
      "step": 1895
    },
    {
      "entropy": 5.882160186767578,
      "epoch": 1.9157841654059506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004887907427988749,
      "loss": 5.862,
      "mean_token_accuracy": 0.12706216275691987,
      "num_tokens": 4190421.0,
      "step": 1900
    },
    {
      "entropy": 5.8674720287322994,
      "epoch": 1.9208270297528998,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886667705379403,
      "loss": 5.8175,
      "mean_token_accuracy": 0.13355472385883332,
      "num_tokens": 4201916.0,
      "step": 1905
    },
    {
      "entropy": 5.874330472946167,
      "epoch": 1.9258698940998489,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004885421341957541,
      "loss": 5.7994,
      "mean_token_accuracy": 0.12998703494668007,
      "num_tokens": 4213742.0,
      "step": 1910
    },
    {
      "entropy": 5.91501522064209,
      "epoch": 1.9309127584467978,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004884168341596888,
      "loss": 5.9826,
      "mean_token_accuracy": 0.1286608800292015,
      "num_tokens": 4226743.0,
      "step": 1915
    },
    {
      "entropy": 5.861303329467773,
      "epoch": 1.9359556227937469,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004882908708191797,
      "loss": 5.7969,
      "mean_token_accuracy": 0.13505506440997123,
      "num_tokens": 4237458.0,
      "step": 1920
    },
    {
      "entropy": 5.8801346778869625,
      "epoch": 1.9409984871406958,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048816424456572403,
      "loss": 5.853,
      "mean_token_accuracy": 0.13410338014364243,
      "num_tokens": 4248466.0,
      "step": 1925
    },
    {
      "entropy": 5.8861486434936525,
      "epoch": 1.946041351487645,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048803695579287905,
      "loss": 5.9155,
      "mean_token_accuracy": 0.1292600877583027,
      "num_tokens": 4260186.0,
      "step": 1930
    },
    {
      "entropy": 5.9890905857086185,
      "epoch": 1.951084215834594,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048790900489626104,
      "loss": 5.9427,
      "mean_token_accuracy": 0.12938498109579086,
      "num_tokens": 4272357.0,
      "step": 1935
    },
    {
      "entropy": 5.975081062316894,
      "epoch": 1.9561270801815431,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004877803922735445,
      "loss": 5.9479,
      "mean_token_accuracy": 0.13041204810142518,
      "num_tokens": 4285583.0,
      "step": 1940
    },
    {
      "entropy": 5.870516920089722,
      "epoch": 1.9611699445284922,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004876511183244603,
      "loss": 5.8324,
      "mean_token_accuracy": 0.13399678468704224,
      "num_tokens": 4296396.0,
      "step": 1945
    },
    {
      "entropy": 5.805631160736084,
      "epoch": 1.9662128088754414,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048752118345079496,
      "loss": 5.7974,
      "mean_token_accuracy": 0.13942713886499405,
      "num_tokens": 4306630.0,
      "step": 1950
    },
    {
      "entropy": 6.00920295715332,
      "epoch": 1.9712556732223905,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048739058805638905,
      "loss": 5.923,
      "mean_token_accuracy": 0.1364424578845501,
      "num_tokens": 4318488.0,
      "step": 1955
    },
    {
      "entropy": 5.812536239624023,
      "epoch": 1.9762985375693394,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000487259332547136,
      "loss": 5.7415,
      "mean_token_accuracy": 0.13686809912323952,
      "num_tokens": 4329604.0,
      "step": 1960
    },
    {
      "entropy": 5.856548261642456,
      "epoch": 1.9813414019162885,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000487127417330981,
      "loss": 5.7621,
      "mean_token_accuracy": 0.13868005126714705,
      "num_tokens": 4341325.0,
      "step": 1965
    },
    {
      "entropy": 5.849880933761597,
      "epoch": 1.9863842662632374,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004869948428179196,
      "loss": 5.8356,
      "mean_token_accuracy": 0.1344212606549263,
      "num_tokens": 4353308.0,
      "step": 1970
    },
    {
      "entropy": 5.881313610076904,
      "epoch": 1.9914271306101865,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004868616094199964,
      "loss": 5.7938,
      "mean_token_accuracy": 0.13609838709235192,
      "num_tokens": 4364937.0,
      "step": 1975
    },
    {
      "entropy": 5.780176305770874,
      "epoch": 1.9964699949571356,
      "grad_norm": 0.984375,
      "learning_rate": 0.000486727717551304,
      "loss": 5.7866,
      "mean_token_accuracy": 0.13767975196242332,
      "num_tokens": 4376477.0,
      "step": 1980
    },
    {
      "entropy": 5.966682434082031,
      "epoch": 2.00100857286939,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004865931676279813,
      "loss": 6.1222,
      "mean_token_accuracy": 0.12126517958111233,
      "num_tokens": 4386192.0,
      "step": 1985
    },
    {
      "entropy": 5.920079565048217,
      "epoch": 2.006051437216339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004864579600682125,
      "loss": 5.5192,
      "mean_token_accuracy": 0.14558632522821427,
      "num_tokens": 4397170.0,
      "step": 1990
    },
    {
      "entropy": 5.766311836242676,
      "epoch": 2.011094301563288,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048632209529222607,
      "loss": 5.5758,
      "mean_token_accuracy": 0.14178299009799958,
      "num_tokens": 4407523.0,
      "step": 1995
    },
    {
      "entropy": 5.935072803497315,
      "epoch": 2.016137165910237,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048618557372229253,
      "loss": 5.5745,
      "mean_token_accuracy": 0.13952894136309624,
      "num_tokens": 4418947.0,
      "step": 2000
    },
    {
      "epoch": 2.016137165910237,
      "eval_entropy": 5.7165008052311475,
      "eval_loss": 5.974915027618408,
      "eval_mean_token_accuracy": 0.13348384690648726,
      "eval_num_tokens": 4418947.0,
      "eval_runtime": 2.6409,
      "eval_samples_per_second": 1188.245,
      "eval_steps_per_second": 148.815,
      "step": 2000
    },
    {
      "entropy": 5.688140535354615,
      "epoch": 2.0211800302571863,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004860483957827243,
      "loss": 5.4905,
      "mean_token_accuracy": 0.14712394177913665,
      "num_tokens": 4429923.0,
      "step": 2005
    },
    {
      "entropy": 5.770442533493042,
      "epoch": 2.026222894604135,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004859105618998735,
      "loss": 5.4643,
      "mean_token_accuracy": 0.14797610491514207,
      "num_tokens": 4441490.0,
      "step": 2010
    },
    {
      "entropy": 5.791120147705078,
      "epoch": 2.031265758951084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004857720725021311,
      "loss": 5.5229,
      "mean_token_accuracy": 0.14146862626075746,
      "num_tokens": 4453302.0,
      "step": 2015
    },
    {
      "entropy": 5.897823286056519,
      "epoch": 2.036308623298033,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048563292801992534,
      "loss": 5.4987,
      "mean_token_accuracy": 0.1451796017587185,
      "num_tokens": 4465393.0,
      "step": 2020
    },
    {
      "entropy": 5.624887275695801,
      "epoch": 2.0413514876449823,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004854931288857203,
      "loss": 5.4281,
      "mean_token_accuracy": 0.15317499190568923,
      "num_tokens": 4475269.0,
      "step": 2025
    },
    {
      "entropy": 5.740149974822998,
      "epoch": 2.0463943519919314,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048535267553401505,
      "loss": 5.4506,
      "mean_token_accuracy": 0.15706779807806015,
      "num_tokens": 4486275.0,
      "step": 2030
    },
    {
      "entropy": 5.748965501785278,
      "epoch": 2.0514372163388805,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048521156840134193,
      "loss": 5.4887,
      "mean_token_accuracy": 0.14535945057868957,
      "num_tokens": 4498684.0,
      "step": 2035
    },
    {
      "entropy": 5.789137077331543,
      "epoch": 2.0564800806858297,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000485069807926265,
      "loss": 5.5163,
      "mean_token_accuracy": 0.14781803041696548,
      "num_tokens": 4509482.0,
      "step": 2040
    },
    {
      "entropy": 5.7220855236053465,
      "epoch": 2.0615229450327788,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004849273945493792,
      "loss": 5.5873,
      "mean_token_accuracy": 0.13608732670545579,
      "num_tokens": 4520638.0,
      "step": 2045
    },
    {
      "entropy": 5.799525737762451,
      "epoch": 2.066565809379728,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004847843287133084,
      "loss": 5.596,
      "mean_token_accuracy": 0.14232194870710374,
      "num_tokens": 4531236.0,
      "step": 2050
    },
    {
      "entropy": 5.817762947082519,
      "epoch": 2.0716086737266766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004846406108627048,
      "loss": 5.5019,
      "mean_token_accuracy": 0.14913840889930724,
      "num_tokens": 4542590.0,
      "step": 2055
    },
    {
      "entropy": 5.653972434997558,
      "epoch": 2.0766515380736257,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048449624144424665,
      "loss": 5.4833,
      "mean_token_accuracy": 0.15156522691249846,
      "num_tokens": 4553592.0,
      "step": 2060
    },
    {
      "entropy": 5.776318883895874,
      "epoch": 2.081694402420575,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048435122090663744,
      "loss": 5.5904,
      "mean_token_accuracy": 0.1468003585934639,
      "num_tokens": 4564735.0,
      "step": 2065
    },
    {
      "entropy": 5.825178718566894,
      "epoch": 2.086737266767524,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048420554970060424,
      "loss": 5.6439,
      "mean_token_accuracy": 0.14915079921483992,
      "num_tokens": 4576515.0,
      "step": 2070
    },
    {
      "entropy": 5.780494499206543,
      "epoch": 2.091780131114473,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048405922827889677,
      "loss": 5.5677,
      "mean_token_accuracy": 0.1428247794508934,
      "num_tokens": 4588212.0,
      "step": 2075
    },
    {
      "entropy": 5.887700366973877,
      "epoch": 2.096822995461422,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048391225709628515,
      "loss": 5.6617,
      "mean_token_accuracy": 0.14353824853897096,
      "num_tokens": 4599482.0,
      "step": 2080
    },
    {
      "entropy": 5.65536413192749,
      "epoch": 2.1018658598083713,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004837646366095593,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1518516346812248,
      "num_tokens": 4609920.0,
      "step": 2085
    },
    {
      "entropy": 5.8641563892364506,
      "epoch": 2.1069087241553204,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048361636727752716,
      "loss": 5.6943,
      "mean_token_accuracy": 0.13604221642017364,
      "num_tokens": 4621944.0,
      "step": 2090
    },
    {
      "entropy": 5.724246883392334,
      "epoch": 2.1119515885022695,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004834674495610131,
      "loss": 5.564,
      "mean_token_accuracy": 0.14400226026773452,
      "num_tokens": 4634191.0,
      "step": 2095
    },
    {
      "entropy": 5.79607253074646,
      "epoch": 2.116994452849218,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048331788392285693,
      "loss": 5.5153,
      "mean_token_accuracy": 0.14646561592817306,
      "num_tokens": 4645655.0,
      "step": 2100
    },
    {
      "entropy": 5.688280439376831,
      "epoch": 2.1220373171961673,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048316767082791205,
      "loss": 5.532,
      "mean_token_accuracy": 0.14386310204863548,
      "num_tokens": 4656922.0,
      "step": 2105
    },
    {
      "entropy": 5.762145566940307,
      "epoch": 2.1270801815431164,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048301681074304427,
      "loss": 5.5514,
      "mean_token_accuracy": 0.14774558395147325,
      "num_tokens": 4668647.0,
      "step": 2110
    },
    {
      "entropy": 5.7129593849182125,
      "epoch": 2.1321230458900655,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048286530413713016,
      "loss": 5.4646,
      "mean_token_accuracy": 0.1492694079875946,
      "num_tokens": 4678708.0,
      "step": 2115
    },
    {
      "entropy": 5.693505477905274,
      "epoch": 2.1371659102370146,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048271315148105574,
      "loss": 5.5788,
      "mean_token_accuracy": 0.146945621073246,
      "num_tokens": 4690962.0,
      "step": 2120
    },
    {
      "entropy": 5.746316862106323,
      "epoch": 2.1422087745839637,
      "grad_norm": 1.109375,
      "learning_rate": 0.000482560353247715,
      "loss": 5.6015,
      "mean_token_accuracy": 0.14061713814735413,
      "num_tokens": 4701833.0,
      "step": 2125
    },
    {
      "entropy": 5.674240255355835,
      "epoch": 2.147251638930913,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004824069099120084,
      "loss": 5.4593,
      "mean_token_accuracy": 0.147596974670887,
      "num_tokens": 4712756.0,
      "step": 2130
    },
    {
      "entropy": 5.7342997074127195,
      "epoch": 2.152294503277862,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004822528219508413,
      "loss": 5.5309,
      "mean_token_accuracy": 0.14017585515975953,
      "num_tokens": 4723226.0,
      "step": 2135
    },
    {
      "entropy": 5.725196743011475,
      "epoch": 2.157337367624811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004820980898431227,
      "loss": 5.5562,
      "mean_token_accuracy": 0.1423388034105301,
      "num_tokens": 4734703.0,
      "step": 2140
    },
    {
      "entropy": 5.743356704711914,
      "epoch": 2.1623802319717598,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004819427140697635,
      "loss": 5.5509,
      "mean_token_accuracy": 0.14523996859788896,
      "num_tokens": 4745356.0,
      "step": 2145
    },
    {
      "entropy": 5.659573221206665,
      "epoch": 2.167423096318709,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004817866951136753,
      "loss": 5.5178,
      "mean_token_accuracy": 0.1489861160516739,
      "num_tokens": 4756342.0,
      "step": 2150
    },
    {
      "entropy": 5.754667043685913,
      "epoch": 2.172465960665658,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048163003345976847,
      "loss": 5.6071,
      "mean_token_accuracy": 0.14500229954719543,
      "num_tokens": 4768782.0,
      "step": 2155
    },
    {
      "entropy": 5.7073558330535885,
      "epoch": 2.177508825012607,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004814727295949512,
      "loss": 5.5012,
      "mean_token_accuracy": 0.1480948284268379,
      "num_tokens": 4779807.0,
      "step": 2160
    },
    {
      "entropy": 5.67652153968811,
      "epoch": 2.1825516893595562,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048131478400812735,
      "loss": 5.5534,
      "mean_token_accuracy": 0.1416294828057289,
      "num_tokens": 4789924.0,
      "step": 2165
    },
    {
      "entropy": 5.771294641494751,
      "epoch": 2.1875945537065054,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004811561971901956,
      "loss": 5.5901,
      "mean_token_accuracy": 0.1433164209127426,
      "num_tokens": 4801112.0,
      "step": 2170
    },
    {
      "entropy": 5.665855121612549,
      "epoch": 2.1926374180534545,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004809969696340474,
      "loss": 5.5258,
      "mean_token_accuracy": 0.14302971139550208,
      "num_tokens": 4811777.0,
      "step": 2175
    },
    {
      "entropy": 5.628685474395752,
      "epoch": 2.1976802824004036,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048083710183456554,
      "loss": 5.4384,
      "mean_token_accuracy": 0.15123602449893953,
      "num_tokens": 4822666.0,
      "step": 2180
    },
    {
      "entropy": 5.6726518154144285,
      "epoch": 2.2027231467473527,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048067659428862306,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1448014609515667,
      "num_tokens": 4833315.0,
      "step": 2185
    },
    {
      "entropy": 5.700483512878418,
      "epoch": 2.2077660110943014,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048051544749508096,
      "loss": 5.552,
      "mean_token_accuracy": 0.14341717287898065,
      "num_tokens": 4844606.0,
      "step": 2190
    },
    {
      "entropy": 5.748077821731568,
      "epoch": 2.2128088754412505,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004803536619547872,
      "loss": 5.6252,
      "mean_token_accuracy": 0.14131659865379334,
      "num_tokens": 4857460.0,
      "step": 2195
    },
    {
      "entropy": 5.683910083770752,
      "epoch": 2.2178517397881996,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000480191238170575,
      "loss": 5.5277,
      "mean_token_accuracy": 0.14796703606843947,
      "num_tokens": 4869289.0,
      "step": 2200
    },
    {
      "entropy": 5.621657609939575,
      "epoch": 2.2228946041351487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004800281766472612,
      "loss": 5.5012,
      "mean_token_accuracy": 0.14826309382915498,
      "num_tokens": 4878912.0,
      "step": 2205
    },
    {
      "entropy": 5.725436496734619,
      "epoch": 2.227937468482098,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047986447789164486,
      "loss": 5.5777,
      "mean_token_accuracy": 0.1396354019641876,
      "num_tokens": 4891312.0,
      "step": 2210
    },
    {
      "entropy": 5.737303686141968,
      "epoch": 2.232980332829047,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004797001424125054,
      "loss": 5.585,
      "mean_token_accuracy": 0.1401967354118824,
      "num_tokens": 4902877.0,
      "step": 2215
    },
    {
      "entropy": 5.702682971954346,
      "epoch": 2.238023197175996,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047953517072060124,
      "loss": 5.4982,
      "mean_token_accuracy": 0.1487305074930191,
      "num_tokens": 4914145.0,
      "step": 2220
    },
    {
      "entropy": 5.738278722763061,
      "epoch": 2.243066061522945,
      "grad_norm": 1.0,
      "learning_rate": 0.0004793695633286684,
      "loss": 5.6342,
      "mean_token_accuracy": 0.13958853334188462,
      "num_tokens": 4924647.0,
      "step": 2225
    },
    {
      "entropy": 5.702555847167969,
      "epoch": 2.2481089258698943,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047920332075141825,
      "loss": 5.4958,
      "mean_token_accuracy": 0.14522965103387833,
      "num_tokens": 4934425.0,
      "step": 2230
    },
    {
      "entropy": 5.58786678314209,
      "epoch": 2.253151790216843,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004790364435055367,
      "loss": 5.4836,
      "mean_token_accuracy": 0.15151820629835128,
      "num_tokens": 4945045.0,
      "step": 2235
    },
    {
      "entropy": 5.675472831726074,
      "epoch": 2.258194654563792,
      "grad_norm": 0.984375,
      "learning_rate": 0.000478868932109682,
      "loss": 5.4835,
      "mean_token_accuracy": 0.1467411383986473,
      "num_tokens": 4956225.0,
      "step": 2240
    },
    {
      "entropy": 5.6820001125335695,
      "epoch": 2.263237518910741,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004787007870844835,
      "loss": 5.4893,
      "mean_token_accuracy": 0.14946691393852235,
      "num_tokens": 4968591.0,
      "step": 2245
    },
    {
      "entropy": 5.58260931968689,
      "epoch": 2.2682803832576903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047853200895253976,
      "loss": 5.5583,
      "mean_token_accuracy": 0.14588147550821304,
      "num_tokens": 4978612.0,
      "step": 2250
    },
    {
      "entropy": 5.651156854629517,
      "epoch": 2.2733232476046394,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.4441,
      "mean_token_accuracy": 0.15017334818840028,
      "num_tokens": 4990039.0,
      "step": 2255
    },
    {
      "entropy": 5.674918603897095,
      "epoch": 2.2783661119515886,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047819255546864797,
      "loss": 5.5703,
      "mean_token_accuracy": 0.14660640731453894,
      "num_tokens": 5001978.0,
      "step": 2260
    },
    {
      "entropy": 5.604819250106812,
      "epoch": 2.2834089762985377,
      "grad_norm": 1.0,
      "learning_rate": 0.0004780218811717293,
      "loss": 5.4722,
      "mean_token_accuracy": 0.15239026844501496,
      "num_tokens": 5012559.0,
      "step": 2265
    },
    {
      "entropy": 5.747910261154175,
      "epoch": 2.288451840645487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047785057587812057,
      "loss": 5.5238,
      "mean_token_accuracy": 0.14815680906176568,
      "num_tokens": 5022944.0,
      "step": 2270
    },
    {
      "entropy": 5.67052116394043,
      "epoch": 2.293494704992436,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047767864012024266,
      "loss": 5.5407,
      "mean_token_accuracy": 0.14561946988105773,
      "num_tokens": 5033771.0,
      "step": 2275
    },
    {
      "entropy": 5.625165319442749,
      "epoch": 2.2985375693393846,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047750607443247585,
      "loss": 5.4905,
      "mean_token_accuracy": 0.15097652673721312,
      "num_tokens": 5045631.0,
      "step": 2280
    },
    {
      "entropy": 5.718620777130127,
      "epoch": 2.3035804336863337,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047733287935115837,
      "loss": 5.4859,
      "mean_token_accuracy": 0.14836325496435165,
      "num_tokens": 5056392.0,
      "step": 2285
    },
    {
      "entropy": 5.577760028839111,
      "epoch": 2.308623298033283,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004771590554145844,
      "loss": 5.5731,
      "mean_token_accuracy": 0.15519963204860687,
      "num_tokens": 5066420.0,
      "step": 2290
    },
    {
      "entropy": 5.729710531234741,
      "epoch": 2.313666162380232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047698460316300285,
      "loss": 5.5464,
      "mean_token_accuracy": 0.14488720446825026,
      "num_tokens": 5075608.0,
      "step": 2295
    },
    {
      "entropy": 5.6740076541900635,
      "epoch": 2.318709026727181,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004768095231386153,
      "loss": 5.5265,
      "mean_token_accuracy": 0.14924899190664292,
      "num_tokens": 5086304.0,
      "step": 2300
    },
    {
      "entropy": 5.702730703353882,
      "epoch": 2.32375189107413,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004766338158855744,
      "loss": 5.5951,
      "mean_token_accuracy": 0.14297810941934586,
      "num_tokens": 5096811.0,
      "step": 2305
    },
    {
      "entropy": 5.710735511779785,
      "epoch": 2.3287947554210793,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004764574819499826,
      "loss": 5.5755,
      "mean_token_accuracy": 0.14123538881540298,
      "num_tokens": 5107118.0,
      "step": 2310
    },
    {
      "entropy": 5.738034582138061,
      "epoch": 2.3338376197680284,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047628052187988955,
      "loss": 5.5514,
      "mean_token_accuracy": 0.15025267228484154,
      "num_tokens": 5118362.0,
      "step": 2315
    },
    {
      "entropy": 5.643430137634278,
      "epoch": 2.338880484114977,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004761029362252914,
      "loss": 5.5419,
      "mean_token_accuracy": 0.1514013096690178,
      "num_tokens": 5128705.0,
      "step": 2320
    },
    {
      "entropy": 5.639211702346802,
      "epoch": 2.343923348461926,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047592472553812856,
      "loss": 5.4308,
      "mean_token_accuracy": 0.15441965758800508,
      "num_tokens": 5139952.0,
      "step": 2325
    },
    {
      "entropy": 5.707053899765015,
      "epoch": 2.3489662128088753,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004757458903722837,
      "loss": 5.6183,
      "mean_token_accuracy": 0.13712134584784508,
      "num_tokens": 5149453.0,
      "step": 2330
    },
    {
      "entropy": 5.68634819984436,
      "epoch": 2.3540090771558244,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047556643128358075,
      "loss": 5.6103,
      "mean_token_accuracy": 0.14707150533795357,
      "num_tokens": 5160951.0,
      "step": 2335
    },
    {
      "entropy": 5.704366636276245,
      "epoch": 2.3590519415027735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047538634882978277,
      "loss": 5.5608,
      "mean_token_accuracy": 0.14414046555757523,
      "num_tokens": 5171172.0,
      "step": 2340
    },
    {
      "entropy": 5.769184732437134,
      "epoch": 2.3640948058497226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047520564357059013,
      "loss": 5.5922,
      "mean_token_accuracy": 0.14316453635692597,
      "num_tokens": 5183178.0,
      "step": 2345
    },
    {
      "entropy": 5.586590814590454,
      "epoch": 2.3691376701966718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004750243160676389,
      "loss": 5.5244,
      "mean_token_accuracy": 0.14793571829795837,
      "num_tokens": 5193719.0,
      "step": 2350
    },
    {
      "entropy": 5.6622710704803465,
      "epoch": 2.374180534543621,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004748423668844992,
      "loss": 5.4408,
      "mean_token_accuracy": 0.15502848774194716,
      "num_tokens": 5203873.0,
      "step": 2355
    },
    {
      "entropy": 5.62037353515625,
      "epoch": 2.37922339889057,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004746597965866733,
      "loss": 5.5127,
      "mean_token_accuracy": 0.1461717367172241,
      "num_tokens": 5213682.0,
      "step": 2360
    },
    {
      "entropy": 5.600476837158203,
      "epoch": 2.384266263237519,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004744766057415938,
      "loss": 5.4398,
      "mean_token_accuracy": 0.14920284003019332,
      "num_tokens": 5224850.0,
      "step": 2365
    },
    {
      "entropy": 5.665679979324341,
      "epoch": 2.389309127584468,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004742927949186222,
      "loss": 5.5412,
      "mean_token_accuracy": 0.15057540088891982,
      "num_tokens": 5237256.0,
      "step": 2370
    },
    {
      "entropy": 5.6559959888458256,
      "epoch": 2.394351991931417,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047410836468904664,
      "loss": 5.5618,
      "mean_token_accuracy": 0.14627649784088134,
      "num_tokens": 5247638.0,
      "step": 2375
    },
    {
      "entropy": 5.632528686523438,
      "epoch": 2.399394856278366,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047392331562608073,
      "loss": 5.4998,
      "mean_token_accuracy": 0.14702658206224442,
      "num_tokens": 5259438.0,
      "step": 2380
    },
    {
      "entropy": 5.69215202331543,
      "epoch": 2.404437720625315,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737376483048609,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1548854447901249,
      "num_tokens": 5269809.0,
      "step": 2385
    },
    {
      "entropy": 5.587312126159668,
      "epoch": 2.4094805849722642,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004735513633024458,
      "loss": 5.5387,
      "mean_token_accuracy": 0.1431642770767212,
      "num_tokens": 5280710.0,
      "step": 2390
    },
    {
      "entropy": 5.765651845932007,
      "epoch": 2.4145234493192134,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004733644611978134,
      "loss": 5.5453,
      "mean_token_accuracy": 0.14382847920060157,
      "num_tokens": 5293162.0,
      "step": 2395
    },
    {
      "entropy": 5.609010028839111,
      "epoch": 2.4195663136661625,
      "grad_norm": 1.0,
      "learning_rate": 0.0004731769425718597,
      "loss": 5.5804,
      "mean_token_accuracy": 0.14515646547079086,
      "num_tokens": 5304510.0,
      "step": 2400
    },
    {
      "entropy": 5.6429084777832035,
      "epoch": 2.4246091780131116,
      "grad_norm": 0.90625,
      "learning_rate": 0.000472988808007397,
      "loss": 5.4373,
      "mean_token_accuracy": 0.14908331930637359,
      "num_tokens": 5314962.0,
      "step": 2405
    },
    {
      "entropy": 5.6114805221557615,
      "epoch": 2.4296520423600603,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047280005808915176,
      "loss": 5.5188,
      "mean_token_accuracy": 0.14984937012195587,
      "num_tokens": 5325314.0,
      "step": 2410
    },
    {
      "entropy": 5.680434942245483,
      "epoch": 2.4346949067070094,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004726106934037631,
      "loss": 5.5899,
      "mean_token_accuracy": 0.14161864668130875,
      "num_tokens": 5336408.0,
      "step": 2415
    },
    {
      "entropy": 5.650278377532959,
      "epoch": 2.4397377710539585,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.4762,
      "mean_token_accuracy": 0.15070291012525558,
      "num_tokens": 5348014.0,
      "step": 2420
    },
    {
      "entropy": 5.612211799621582,
      "epoch": 2.4447806354009076,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047223012208766384,
      "loss": 5.4997,
      "mean_token_accuracy": 0.15422628223896026,
      "num_tokens": 5357275.0,
      "step": 2425
    },
    {
      "entropy": 5.655633640289307,
      "epoch": 2.4498234997478567,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004720389166397777,
      "loss": 5.4646,
      "mean_token_accuracy": 0.14888779520988465,
      "num_tokens": 5367822.0,
      "step": 2430
    },
    {
      "entropy": 5.659993982315063,
      "epoch": 2.454866364094806,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047184709879039325,
      "loss": 5.5216,
      "mean_token_accuracy": 0.14660426527261733,
      "num_tokens": 5379193.0,
      "step": 2435
    },
    {
      "entropy": 5.592276620864868,
      "epoch": 2.459909228441755,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047165466913568497,
      "loss": 5.4914,
      "mean_token_accuracy": 0.1498927131295204,
      "num_tokens": 5389825.0,
      "step": 2440
    },
    {
      "entropy": 5.517462110519409,
      "epoch": 2.464952092788704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004714616282737285,
      "loss": 5.3926,
      "mean_token_accuracy": 0.15678471475839614,
      "num_tokens": 5400318.0,
      "step": 2445
    },
    {
      "entropy": 5.629178714752197,
      "epoch": 2.469994957135653,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004712679768044995,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1479016587138176,
      "num_tokens": 5411432.0,
      "step": 2450
    },
    {
      "entropy": 5.653389549255371,
      "epoch": 2.4750378214826023,
      "grad_norm": 0.9375,
      "learning_rate": 0.000471073715329871,
      "loss": 5.5909,
      "mean_token_accuracy": 0.14415307492017745,
      "num_tokens": 5422662.0,
      "step": 2455
    },
    {
      "entropy": 5.606546401977539,
      "epoch": 2.4800806858295514,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004708788444536123,
      "loss": 5.6083,
      "mean_token_accuracy": 0.14451165348291398,
      "num_tokens": 5434976.0,
      "step": 2460
    },
    {
      "entropy": 5.709273767471314,
      "epoch": 2.4851235501765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004706833647813865,
      "loss": 5.4878,
      "mean_token_accuracy": 0.14486440569162368,
      "num_tokens": 5445607.0,
      "step": 2465
    },
    {
      "entropy": 5.710936784744263,
      "epoch": 2.4901664145234492,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047048727692074893,
      "loss": 5.6434,
      "mean_token_accuracy": 0.14222367703914643,
      "num_tokens": 5455793.0,
      "step": 2470
    },
    {
      "entropy": 5.652678918838501,
      "epoch": 2.4952092788703983,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004702905814811452,
      "loss": 5.5881,
      "mean_token_accuracy": 0.14239798486232758,
      "num_tokens": 5466800.0,
      "step": 2475
    },
    {
      "entropy": 5.7010115623474125,
      "epoch": 2.5002521432173475,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047009327907390926,
      "loss": 5.5406,
      "mean_token_accuracy": 0.14696398004889488,
      "num_tokens": 5477875.0,
      "step": 2480
    },
    {
      "entropy": 5.590509605407715,
      "epoch": 2.5052950075642966,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004698953703122616,
      "loss": 5.5611,
      "mean_token_accuracy": 0.14322107136249543,
      "num_tokens": 5488193.0,
      "step": 2485
    },
    {
      "entropy": 5.782047271728516,
      "epoch": 2.5103378719112457,
      "grad_norm": 0.84375,
      "learning_rate": 0.00046969685581130715,
      "loss": 5.7244,
      "mean_token_accuracy": 0.1373824082314968,
      "num_tokens": 5500179.0,
      "step": 2490
    },
    {
      "entropy": 5.704003095626831,
      "epoch": 2.515380736258195,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046949773618803353,
      "loss": 5.5792,
      "mean_token_accuracy": 0.13811678811907768,
      "num_tokens": 5513413.0,
      "step": 2495
    },
    {
      "entropy": 5.694434595108032,
      "epoch": 2.5204236006051435,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046929801206130925,
      "loss": 5.5885,
      "mean_token_accuracy": 0.14615954756736754,
      "num_tokens": 5524697.0,
      "step": 2500
    },
    {
      "epoch": 2.5204236006051435,
      "eval_entropy": 5.569981584718816,
      "eval_loss": 5.916189670562744,
      "eval_mean_token_accuracy": 0.13634676808543483,
      "eval_num_tokens": 5524697.0,
      "eval_runtime": 2.6358,
      "eval_samples_per_second": 1190.523,
      "eval_steps_per_second": 149.1,
      "step": 2500
    },
    {
      "entropy": 5.715534448623657,
      "epoch": 2.5254664649520926,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046909768405188135,
      "loss": 5.6376,
      "mean_token_accuracy": 0.14564819484949112,
      "num_tokens": 5536816.0,
      "step": 2505
    },
    {
      "entropy": 5.663525581359863,
      "epoch": 2.5305093292990417,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046889675278237404,
      "loss": 5.6021,
      "mean_token_accuracy": 0.14897046610713005,
      "num_tokens": 5549329.0,
      "step": 2510
    },
    {
      "entropy": 5.640772485733033,
      "epoch": 2.535552193645991,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004686952188772863,
      "loss": 5.4746,
      "mean_token_accuracy": 0.15271224081516266,
      "num_tokens": 5560015.0,
      "step": 2515
    },
    {
      "entropy": 5.686579275131225,
      "epoch": 2.54059505799294,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046849308296299005,
      "loss": 5.5935,
      "mean_token_accuracy": 0.14233456403017045,
      "num_tokens": 5571031.0,
      "step": 2520
    },
    {
      "entropy": 5.681400966644287,
      "epoch": 2.545637922339889,
      "grad_norm": 1.0,
      "learning_rate": 0.0004682903456677285,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1482929654419422,
      "num_tokens": 5582283.0,
      "step": 2525
    },
    {
      "entropy": 5.7204197406768795,
      "epoch": 2.550680786686838,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046808700762161385,
      "loss": 5.594,
      "mean_token_accuracy": 0.14391592592000962,
      "num_tokens": 5593185.0,
      "step": 2530
    },
    {
      "entropy": 5.6744321346282955,
      "epoch": 2.5557236510337873,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004678830694566253,
      "loss": 5.4978,
      "mean_token_accuracy": 0.14866942316293716,
      "num_tokens": 5604147.0,
      "step": 2535
    },
    {
      "entropy": 5.506406497955322,
      "epoch": 2.5607665153807364,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004676785318066075,
      "loss": 5.4899,
      "mean_token_accuracy": 0.14699023813009263,
      "num_tokens": 5615074.0,
      "step": 2540
    },
    {
      "entropy": 5.619188117980957,
      "epoch": 2.5658093797276855,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004674733953072682,
      "loss": 5.5366,
      "mean_token_accuracy": 0.15007463097572327,
      "num_tokens": 5626599.0,
      "step": 2545
    },
    {
      "entropy": 5.606451225280762,
      "epoch": 2.5708522440746346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004672676605961763,
      "loss": 5.4654,
      "mean_token_accuracy": 0.15293785482645034,
      "num_tokens": 5637378.0,
      "step": 2550
    },
    {
      "entropy": 5.608837270736695,
      "epoch": 2.5758951084215833,
      "grad_norm": 1.0,
      "learning_rate": 0.00046706132831276007,
      "loss": 5.4957,
      "mean_token_accuracy": 0.15240262746810912,
      "num_tokens": 5649152.0,
      "step": 2555
    },
    {
      "entropy": 5.674561071395874,
      "epoch": 2.5809379727685324,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046685439909830486,
      "loss": 5.591,
      "mean_token_accuracy": 0.1499859184026718,
      "num_tokens": 5660476.0,
      "step": 2560
    },
    {
      "entropy": 5.571696519851685,
      "epoch": 2.5859808371154815,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004666468735959516,
      "loss": 5.4435,
      "mean_token_accuracy": 0.152605539560318,
      "num_tokens": 5672178.0,
      "step": 2565
    },
    {
      "entropy": 5.64791316986084,
      "epoch": 2.5910237014624307,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004664387524506943,
      "loss": 5.5385,
      "mean_token_accuracy": 0.1452690616250038,
      "num_tokens": 5683988.0,
      "step": 2570
    },
    {
      "entropy": 5.638162469863891,
      "epoch": 2.59606656580938,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004662300363093784,
      "loss": 5.566,
      "mean_token_accuracy": 0.14803780764341354,
      "num_tokens": 5695009.0,
      "step": 2575
    },
    {
      "entropy": 5.633869409561157,
      "epoch": 2.601109430156329,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004660207258206983,
      "loss": 5.556,
      "mean_token_accuracy": 0.14445877969264984,
      "num_tokens": 5706838.0,
      "step": 2580
    },
    {
      "entropy": 5.617272233963012,
      "epoch": 2.606152294503278,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.448,
      "mean_token_accuracy": 0.15211595892906188,
      "num_tokens": 5717846.0,
      "step": 2585
    },
    {
      "entropy": 5.605248832702637,
      "epoch": 2.6111951588502267,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004656003244052582,
      "loss": 5.4754,
      "mean_token_accuracy": 0.14720592200756072,
      "num_tokens": 5727519.0,
      "step": 2590
    },
    {
      "entropy": 5.626052570343018,
      "epoch": 2.616238023197176,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004653892347851155,
      "loss": 5.5436,
      "mean_token_accuracy": 0.1400951400399208,
      "num_tokens": 5738717.0,
      "step": 2595
    },
    {
      "entropy": 5.694786405563354,
      "epoch": 2.621280887544125,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004651775534308392,
      "loss": 5.6016,
      "mean_token_accuracy": 0.14594041481614112,
      "num_tokens": 5750095.0,
      "step": 2600
    },
    {
      "entropy": 5.621736240386963,
      "epoch": 2.626323751891074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004649652810003398,
      "loss": 5.4696,
      "mean_token_accuracy": 0.15165551304817199,
      "num_tokens": 5760282.0,
      "step": 2605
    },
    {
      "entropy": 5.6037664890289305,
      "epoch": 2.631366616238023,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046475241815336504,
      "loss": 5.5187,
      "mean_token_accuracy": 0.14478351324796676,
      "num_tokens": 5770547.0,
      "step": 2610
    },
    {
      "entropy": 5.631533622741699,
      "epoch": 2.6364094805849723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004645389655514974,
      "loss": 5.4969,
      "mean_token_accuracy": 0.14975763708353043,
      "num_tokens": 5780801.0,
      "step": 2615
    },
    {
      "entropy": 5.684841108322144,
      "epoch": 2.6414523449319214,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004643249238581525,
      "loss": 5.5415,
      "mean_token_accuracy": 0.14400465935468673,
      "num_tokens": 5791851.0,
      "step": 2620
    },
    {
      "entropy": 5.636694002151489,
      "epoch": 2.6464952092788705,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004641102937385769,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1403758630156517,
      "num_tokens": 5802551.0,
      "step": 2625
    },
    {
      "entropy": 5.620252275466919,
      "epoch": 2.6515380736258196,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004638950758598461,
      "loss": 5.4711,
      "mean_token_accuracy": 0.1502656802535057,
      "num_tokens": 5813376.0,
      "step": 2630
    },
    {
      "entropy": 5.625906467437744,
      "epoch": 2.6565809379727687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046367927089086203,
      "loss": 5.5303,
      "mean_token_accuracy": 0.14435117393732072,
      "num_tokens": 5824634.0,
      "step": 2635
    },
    {
      "entropy": 5.620067501068116,
      "epoch": 2.661623802319718,
      "grad_norm": 0.875,
      "learning_rate": 0.00046346287950235163,
      "loss": 5.4227,
      "mean_token_accuracy": 0.15278890430927278,
      "num_tokens": 5836771.0,
      "step": 2640
    },
    {
      "entropy": 5.5916036605834964,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004632459023668642,
      "loss": 5.4494,
      "mean_token_accuracy": 0.14611073359847068,
      "num_tokens": 5847695.0,
      "step": 2645
    },
    {
      "entropy": 5.585438299179077,
      "epoch": 2.6717095310136156,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004630283401587699,
      "loss": 5.5338,
      "mean_token_accuracy": 0.1470185749232769,
      "num_tokens": 5859843.0,
      "step": 2650
    },
    {
      "entropy": 5.619991207122803,
      "epoch": 2.6767523953605648,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004628101935542568,
      "loss": 5.4991,
      "mean_token_accuracy": 0.1494264855980873,
      "num_tokens": 5871093.0,
      "step": 2655
    },
    {
      "entropy": 5.59155764579773,
      "epoch": 2.681795259707514,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046259146323132986,
      "loss": 5.4893,
      "mean_token_accuracy": 0.15731683522462844,
      "num_tokens": 5882235.0,
      "step": 2660
    },
    {
      "entropy": 5.544424819946289,
      "epoch": 2.686838124054463,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046237214986980776,
      "loss": 5.4143,
      "mean_token_accuracy": 0.15443341434001923,
      "num_tokens": 5892567.0,
      "step": 2665
    },
    {
      "entropy": 5.669776868820191,
      "epoch": 2.691880988401412,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004621522541513216,
      "loss": 5.5736,
      "mean_token_accuracy": 0.14649525731801988,
      "num_tokens": 5904849.0,
      "step": 2670
    },
    {
      "entropy": 5.6247350692749025,
      "epoch": 2.696923852748361,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004619317767593123,
      "loss": 5.5149,
      "mean_token_accuracy": 0.14898416474461557,
      "num_tokens": 5914873.0,
      "step": 2675
    },
    {
      "entropy": 5.647739696502685,
      "epoch": 2.70196671709531,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046171071837902873,
      "loss": 5.5413,
      "mean_token_accuracy": 0.14528264254331588,
      "num_tokens": 5926312.0,
      "step": 2680
    },
    {
      "entropy": 5.601176309585571,
      "epoch": 2.707009581442259,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004614890796975254,
      "loss": 5.5962,
      "mean_token_accuracy": 0.14780106246471406,
      "num_tokens": 5939665.0,
      "step": 2685
    },
    {
      "entropy": 5.626812934875488,
      "epoch": 2.712052445789208,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004612668614036605,
      "loss": 5.5046,
      "mean_token_accuracy": 0.1570080816745758,
      "num_tokens": 5949999.0,
      "step": 2690
    },
    {
      "entropy": 5.612789058685303,
      "epoch": 2.7170953101361572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004610440641880934,
      "loss": 5.4787,
      "mean_token_accuracy": 0.15323931947350503,
      "num_tokens": 5960026.0,
      "step": 2695
    },
    {
      "entropy": 5.532758378982544,
      "epoch": 2.7221381744831064,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046082068874328324,
      "loss": 5.4822,
      "mean_token_accuracy": 0.14454658031463624,
      "num_tokens": 5971505.0,
      "step": 2700
    },
    {
      "entropy": 5.682231378555298,
      "epoch": 2.7271810388300555,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046059673576348596,
      "loss": 5.4588,
      "mean_token_accuracy": 0.15356901586055755,
      "num_tokens": 5982919.0,
      "step": 2705
    },
    {
      "entropy": 5.490727090835572,
      "epoch": 2.7322239031770046,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046037220594475264,
      "loss": 5.4787,
      "mean_token_accuracy": 0.14964330568909645,
      "num_tokens": 5993146.0,
      "step": 2710
    },
    {
      "entropy": 5.609921073913574,
      "epoch": 2.7372667675239537,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000460147099984927,
      "loss": 5.4831,
      "mean_token_accuracy": 0.14875664785504342,
      "num_tokens": 6004864.0,
      "step": 2715
    },
    {
      "entropy": 5.591419696807861,
      "epoch": 2.742309631870903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004599214185836438,
      "loss": 5.4795,
      "mean_token_accuracy": 0.14928718283772469,
      "num_tokens": 6016440.0,
      "step": 2720
    },
    {
      "entropy": 5.523524332046509,
      "epoch": 2.747352496217852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045969516244232574,
      "loss": 5.4372,
      "mean_token_accuracy": 0.15902109295129777,
      "num_tokens": 6026227.0,
      "step": 2725
    },
    {
      "entropy": 5.582226324081421,
      "epoch": 2.752395360564801,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045946833226418235,
      "loss": 5.499,
      "mean_token_accuracy": 0.1450572945177555,
      "num_tokens": 6038290.0,
      "step": 2730
    },
    {
      "entropy": 5.630072546005249,
      "epoch": 2.7574382249117497,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004592409287542069,
      "loss": 5.5257,
      "mean_token_accuracy": 0.14881781116127968,
      "num_tokens": 6049611.0,
      "step": 2735
    },
    {
      "entropy": 5.658566808700561,
      "epoch": 2.762481089258699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004590129526191748,
      "loss": 5.5283,
      "mean_token_accuracy": 0.15084161460399628,
      "num_tokens": 6060270.0,
      "step": 2740
    },
    {
      "entropy": 5.582420682907104,
      "epoch": 2.767523953605648,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004587844045676413,
      "loss": 5.5306,
      "mean_token_accuracy": 0.15279269218444824,
      "num_tokens": 6071380.0,
      "step": 2745
    },
    {
      "entropy": 5.655093336105347,
      "epoch": 2.772566817952597,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.564,
      "mean_token_accuracy": 0.14139169007539748,
      "num_tokens": 6082655.0,
      "step": 2750
    },
    {
      "entropy": 5.594664430618286,
      "epoch": 2.777609682299546,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045832559555817525,
      "loss": 5.4424,
      "mean_token_accuracy": 0.1508428856730461,
      "num_tokens": 6094446.0,
      "step": 2755
    },
    {
      "entropy": 5.65729866027832,
      "epoch": 2.7826525466464953,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045809533602623185,
      "loss": 5.5457,
      "mean_token_accuracy": 0.14784732162952424,
      "num_tokens": 6105606.0,
      "step": 2760
    },
    {
      "entropy": 5.543753385543823,
      "epoch": 2.7876954109934444,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004578645074297603,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1504577398300171,
      "num_tokens": 6117743.0,
      "step": 2765
    },
    {
      "entropy": 5.611225271224976,
      "epoch": 2.792738275340393,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045763311048618144,
      "loss": 5.5584,
      "mean_token_accuracy": 0.15163731351494789,
      "num_tokens": 6128324.0,
      "step": 2770
    },
    {
      "entropy": 5.623370504379272,
      "epoch": 2.797781139687342,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004574011459146822,
      "loss": 5.5027,
      "mean_token_accuracy": 0.14757978320121765,
      "num_tokens": 6138742.0,
      "step": 2775
    },
    {
      "entropy": 5.694330739974975,
      "epoch": 2.8028240040342913,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045716861443621386,
      "loss": 5.5109,
      "mean_token_accuracy": 0.14752026051282882,
      "num_tokens": 6149990.0,
      "step": 2780
    },
    {
      "entropy": 5.572966194152832,
      "epoch": 2.8078668683812404,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045693551677348967,
      "loss": 5.5465,
      "mean_token_accuracy": 0.1497933492064476,
      "num_tokens": 6160281.0,
      "step": 2785
    },
    {
      "entropy": 5.601632356643677,
      "epoch": 2.8129097327281896,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004567018536509823,
      "loss": 5.5768,
      "mean_token_accuracy": 0.1480606362223625,
      "num_tokens": 6171501.0,
      "step": 2790
    },
    {
      "entropy": 5.7564770698547365,
      "epoch": 2.8179525970751387,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045646762579492245,
      "loss": 5.6462,
      "mean_token_accuracy": 0.14339930117130278,
      "num_tokens": 6183401.0,
      "step": 2795
    },
    {
      "entropy": 5.63318281173706,
      "epoch": 2.822995461422088,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00045623283393329546,
      "loss": 5.4648,
      "mean_token_accuracy": 0.15423868894577025,
      "num_tokens": 6194451.0,
      "step": 2800
    },
    {
      "entropy": 5.533782958984375,
      "epoch": 2.828038325769037,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045599747879584,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1552419275045395,
      "num_tokens": 6204410.0,
      "step": 2805
    },
    {
      "entropy": 5.581860542297363,
      "epoch": 2.833081190115986,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004557615611140451,
      "loss": 5.4663,
      "mean_token_accuracy": 0.14584519416093827,
      "num_tokens": 6215091.0,
      "step": 2810
    },
    {
      "entropy": 5.628701496124267,
      "epoch": 2.838124054462935,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004555250816211485,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1537807837128639,
      "num_tokens": 6227201.0,
      "step": 2815
    },
    {
      "entropy": 5.635453462600708,
      "epoch": 2.8431669188098843,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045528804105213395,
      "loss": 5.5297,
      "mean_token_accuracy": 0.1468923158943653,
      "num_tokens": 6237067.0,
      "step": 2820
    },
    {
      "entropy": 5.607160139083862,
      "epoch": 2.848209783156833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004550504401437289,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1566188246011734,
      "num_tokens": 6247269.0,
      "step": 2825
    },
    {
      "entropy": 5.580532646179199,
      "epoch": 2.853252647503782,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045481227963440235,
      "loss": 5.5552,
      "mean_token_accuracy": 0.14393058121204377,
      "num_tokens": 6259094.0,
      "step": 2830
    },
    {
      "entropy": 5.662309265136718,
      "epoch": 2.858295511850731,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045457356026436295,
      "loss": 5.483,
      "mean_token_accuracy": 0.1481974482536316,
      "num_tokens": 6270012.0,
      "step": 2835
    },
    {
      "entropy": 5.528801679611206,
      "epoch": 2.8633383761976803,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045433428277555577,
      "loss": 5.486,
      "mean_token_accuracy": 0.15124562233686448,
      "num_tokens": 6280936.0,
      "step": 2840
    },
    {
      "entropy": 5.617532968521118,
      "epoch": 2.8683812405446294,
      "grad_norm": 1.0,
      "learning_rate": 0.0004540944479116609,
      "loss": 5.4852,
      "mean_token_accuracy": 0.15167278051376343,
      "num_tokens": 6291471.0,
      "step": 2845
    },
    {
      "entropy": 5.591040420532226,
      "epoch": 2.8734241048915785,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004538540564180906,
      "loss": 5.5498,
      "mean_token_accuracy": 0.1447784885764122,
      "num_tokens": 6303717.0,
      "step": 2850
    },
    {
      "entropy": 5.659421777725219,
      "epoch": 2.8784669692385276,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004536131090419871,
      "loss": 5.5347,
      "mean_token_accuracy": 0.14546699672937394,
      "num_tokens": 6312960.0,
      "step": 2855
    },
    {
      "entropy": 5.690258121490478,
      "epoch": 2.8835098335854763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004533716065322205,
      "loss": 5.5191,
      "mean_token_accuracy": 0.14920947849750518,
      "num_tokens": 6322558.0,
      "step": 2860
    },
    {
      "entropy": 5.709000444412231,
      "epoch": 2.8885526979324254,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004531295496393861,
      "loss": 5.6016,
      "mean_token_accuracy": 0.14819440245628357,
      "num_tokens": 6334763.0,
      "step": 2865
    },
    {
      "entropy": 5.636456823348999,
      "epoch": 2.8935955622793745,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004528869391158023,
      "loss": 5.5368,
      "mean_token_accuracy": 0.14484109878540039,
      "num_tokens": 6346678.0,
      "step": 2870
    },
    {
      "entropy": 5.617731189727783,
      "epoch": 2.8986384266263236,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045264377571550834,
      "loss": 5.5179,
      "mean_token_accuracy": 0.15168764740228652,
      "num_tokens": 6357495.0,
      "step": 2875
    },
    {
      "entropy": 5.540056657791138,
      "epoch": 2.9036812909732728,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004524000601942615,
      "loss": 5.4463,
      "mean_token_accuracy": 0.15301640406250955,
      "num_tokens": 6368557.0,
      "step": 2880
    },
    {
      "entropy": 5.5962952136993405,
      "epoch": 2.908724155320222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004521557933095354,
      "loss": 5.4694,
      "mean_token_accuracy": 0.15043903216719628,
      "num_tokens": 6378471.0,
      "step": 2885
    },
    {
      "entropy": 5.568690586090088,
      "epoch": 2.913767019667171,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000451910975820517,
      "loss": 5.5524,
      "mean_token_accuracy": 0.14270556718111038,
      "num_tokens": 6390826.0,
      "step": 2890
    },
    {
      "entropy": 5.598607730865479,
      "epoch": 2.91880988401412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045166560848810476,
      "loss": 5.3947,
      "mean_token_accuracy": 0.15188634693622588,
      "num_tokens": 6400438.0,
      "step": 2895
    },
    {
      "entropy": 5.525550794601441,
      "epoch": 2.9238527483610692,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045141969207490606,
      "loss": 5.5038,
      "mean_token_accuracy": 0.15191902816295624,
      "num_tokens": 6411533.0,
      "step": 2900
    },
    {
      "entropy": 5.56376576423645,
      "epoch": 2.9288956127080183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004511732273452347,
      "loss": 5.4578,
      "mean_token_accuracy": 0.15035423785448074,
      "num_tokens": 6421409.0,
      "step": 2905
    },
    {
      "entropy": 5.606322145462036,
      "epoch": 2.9339384770549675,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004509262150651089,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1473022684454918,
      "num_tokens": 6432738.0,
      "step": 2910
    },
    {
      "entropy": 5.606263065338135,
      "epoch": 2.938981341401916,
      "grad_norm": 1.125,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.4981,
      "mean_token_accuracy": 0.15135152339935304,
      "num_tokens": 6443665.0,
      "step": 2915
    },
    {
      "entropy": 5.579765367507934,
      "epoch": 2.9440242057488653,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004504305509260724,
      "loss": 5.4426,
      "mean_token_accuracy": 0.15822614431381227,
      "num_tokens": 6454849.0,
      "step": 2920
    },
    {
      "entropy": 5.567010116577149,
      "epoch": 2.9490670700958144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004501819006076974,
      "loss": 5.5083,
      "mean_token_accuracy": 0.1529279351234436,
      "num_tokens": 6465791.0,
      "step": 2925
    },
    {
      "entropy": 5.583162641525268,
      "epoch": 2.9541099344427635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044993270581993413,
      "loss": 5.4388,
      "mean_token_accuracy": 0.15790292471647263,
      "num_tokens": 6475740.0,
      "step": 2930
    },
    {
      "entropy": 5.565184497833252,
      "epoch": 2.9591527987897126,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00044968296733728587,
      "loss": 5.48,
      "mean_token_accuracy": 0.14736393392086028,
      "num_tokens": 6487594.0,
      "step": 2935
    },
    {
      "entropy": 5.522630405426026,
      "epoch": 2.9641956631366617,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004494326859359456,
      "loss": 5.4349,
      "mean_token_accuracy": 0.14542689472436904,
      "num_tokens": 6498683.0,
      "step": 2940
    },
    {
      "entropy": 5.564045667648315,
      "epoch": 2.969238527483611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044918186239379363,
      "loss": 5.4473,
      "mean_token_accuracy": 0.14891347140073777,
      "num_tokens": 6509069.0,
      "step": 2945
    },
    {
      "entropy": 5.619791841506958,
      "epoch": 2.9742813918305595,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004489304974903953,
      "loss": 5.4799,
      "mean_token_accuracy": 0.14885881543159485,
      "num_tokens": 6519480.0,
      "step": 2950
    },
    {
      "entropy": 5.550818204879761,
      "epoch": 2.9793242561775086,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004486785920069985,
      "loss": 5.5093,
      "mean_token_accuracy": 0.15276284217834474,
      "num_tokens": 6531010.0,
      "step": 2955
    },
    {
      "entropy": 5.6219322681427,
      "epoch": 2.9843671205244577,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004484261467265313,
      "loss": 5.4605,
      "mean_token_accuracy": 0.14902912825345993,
      "num_tokens": 6540292.0,
      "step": 2960
    },
    {
      "entropy": 5.664312839508057,
      "epoch": 2.989409984871407,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044817316243359956,
      "loss": 5.6143,
      "mean_token_accuracy": 0.13999737054109573,
      "num_tokens": 6553199.0,
      "step": 2965
    },
    {
      "entropy": 5.522039890289307,
      "epoch": 2.994452849218356,
      "grad_norm": 0.90625,
      "learning_rate": 0.00044791963991448427,
      "loss": 5.4789,
      "mean_token_accuracy": 0.15009307116270065,
      "num_tokens": 6564142.0,
      "step": 2970
    },
    {
      "entropy": 5.607178783416748,
      "epoch": 2.999495713565305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044766557995713915,
      "loss": 5.4602,
      "mean_token_accuracy": 0.15400594770908355,
      "num_tokens": 6574519.0,
      "step": 2975
    },
    {
      "entropy": 5.563734372456868,
      "epoch": 3.0040342914775593,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004474109833511885,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1663739780584971,
      "num_tokens": 6583348.0,
      "step": 2980
    },
    {
      "entropy": 5.530019807815552,
      "epoch": 3.0090771558245084,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004471558508879244,
      "loss": 5.2335,
      "mean_token_accuracy": 0.16050369441509246,
      "num_tokens": 6595704.0,
      "step": 2985
    },
    {
      "entropy": 5.53251576423645,
      "epoch": 3.0141200201714575,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004469001833603045,
      "loss": 5.1451,
      "mean_token_accuracy": 0.16149887889623643,
      "num_tokens": 6607177.0,
      "step": 2990
    },
    {
      "entropy": 5.47862663269043,
      "epoch": 3.0191628845184066,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004466439815629493,
      "loss": 5.1657,
      "mean_token_accuracy": 0.1669058695435524,
      "num_tokens": 6620798.0,
      "step": 2995
    },
    {
      "entropy": 5.5322123050689695,
      "epoch": 3.0242057488653553,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044638724629214,
      "loss": 5.2387,
      "mean_token_accuracy": 0.16908339858055116,
      "num_tokens": 6632972.0,
      "step": 3000
    },
    {
      "epoch": 3.0242057488653553,
      "eval_entropy": 5.518803171832446,
      "eval_loss": 5.829952239990234,
      "eval_mean_token_accuracy": 0.14320968427048383,
      "eval_num_tokens": 6632972.0,
      "eval_runtime": 2.8356,
      "eval_samples_per_second": 1106.637,
      "eval_steps_per_second": 138.594,
      "step": 3000
    },
    {
      "entropy": 5.498778438568115,
      "epoch": 3.0292486132123044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044612997834581557,
      "loss": 5.0509,
      "mean_token_accuracy": 0.1695736199617386,
      "num_tokens": 6643259.0,
      "step": 3005
    },
    {
      "entropy": 5.455741310119629,
      "epoch": 3.0342914775592535,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044587217852357086,
      "loss": 5.2386,
      "mean_token_accuracy": 0.16795216798782348,
      "num_tokens": 6655875.0,
      "step": 3010
    },
    {
      "entropy": 5.57458815574646,
      "epoch": 3.0393343419062027,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044561384762665363,
      "loss": 5.1601,
      "mean_token_accuracy": 0.16992995291948318,
      "num_tokens": 6667377.0,
      "step": 3015
    },
    {
      "entropy": 5.408328199386597,
      "epoch": 3.044377206253152,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004453549864579623,
      "loss": 5.1415,
      "mean_token_accuracy": 0.1619119167327881,
      "num_tokens": 6678293.0,
      "step": 3020
    },
    {
      "entropy": 5.49166464805603,
      "epoch": 3.049420070600101,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004450955958220433,
      "loss": 5.1973,
      "mean_token_accuracy": 0.16586864441633226,
      "num_tokens": 6690664.0,
      "step": 3025
    },
    {
      "entropy": 5.666604042053223,
      "epoch": 3.05446293494705,
      "grad_norm": 1.0,
      "learning_rate": 0.00044483567652508864,
      "loss": 5.2032,
      "mean_token_accuracy": 0.16186309605836868,
      "num_tokens": 6701465.0,
      "step": 3030
    },
    {
      "entropy": 5.437854290008545,
      "epoch": 3.059505799293999,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004445752293749337,
      "loss": 5.1995,
      "mean_token_accuracy": 0.1668174758553505,
      "num_tokens": 6713290.0,
      "step": 3035
    },
    {
      "entropy": 5.506089878082276,
      "epoch": 3.0645486636409482,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004443142551810541,
      "loss": 5.2168,
      "mean_token_accuracy": 0.1657101944088936,
      "num_tokens": 6723496.0,
      "step": 3040
    },
    {
      "entropy": 5.552100563049317,
      "epoch": 3.069591527987897,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044405275475456365,
      "loss": 5.1983,
      "mean_token_accuracy": 0.163461734354496,
      "num_tokens": 6734030.0,
      "step": 3045
    },
    {
      "entropy": 5.5378697395324705,
      "epoch": 3.074634392334846,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004437907289082116,
      "loss": 5.2287,
      "mean_token_accuracy": 0.16727308332920074,
      "num_tokens": 6745966.0,
      "step": 3050
    },
    {
      "entropy": 5.547027254104615,
      "epoch": 3.079677256681795,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044352817845638053,
      "loss": 5.1435,
      "mean_token_accuracy": 0.16268666386604308,
      "num_tokens": 6756982.0,
      "step": 3055
    },
    {
      "entropy": 5.420411109924316,
      "epoch": 3.0847201210287443,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004432651042150833,
      "loss": 5.1571,
      "mean_token_accuracy": 0.16135234385728836,
      "num_tokens": 6768950.0,
      "step": 3060
    },
    {
      "entropy": 5.402483034133911,
      "epoch": 3.0897629853756934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044300150700196054,
      "loss": 5.1307,
      "mean_token_accuracy": 0.15984653681516647,
      "num_tokens": 6779748.0,
      "step": 3065
    },
    {
      "entropy": 5.5331212997436525,
      "epoch": 3.0948058497226425,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004427373876362786,
      "loss": 5.2637,
      "mean_token_accuracy": 0.16265043020248413,
      "num_tokens": 6791456.0,
      "step": 3070
    },
    {
      "entropy": 5.467827606201172,
      "epoch": 3.0998487140695916,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004424727469389265,
      "loss": 5.1834,
      "mean_token_accuracy": 0.16567315012216569,
      "num_tokens": 6802813.0,
      "step": 3075
    },
    {
      "entropy": 5.574123048782349,
      "epoch": 3.1048915784165407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004422075857324137,
      "loss": 5.2427,
      "mean_token_accuracy": 0.16068820655345917,
      "num_tokens": 6813310.0,
      "step": 3080
    },
    {
      "entropy": 5.521329307556153,
      "epoch": 3.10993444276349,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004419419048408673,
      "loss": 5.2667,
      "mean_token_accuracy": 0.16046479642391204,
      "num_tokens": 6825593.0,
      "step": 3085
    },
    {
      "entropy": 5.441070795059204,
      "epoch": 3.1149773071104385,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004416757050900297,
      "loss": 5.1642,
      "mean_token_accuracy": 0.16903398483991622,
      "num_tokens": 6836279.0,
      "step": 3090
    },
    {
      "entropy": 5.441403913497925,
      "epoch": 3.1200201714573876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004414089873072557,
      "loss": 5.1424,
      "mean_token_accuracy": 0.16970179229974747,
      "num_tokens": 6846087.0,
      "step": 3095
    },
    {
      "entropy": 5.491785383224487,
      "epoch": 3.1250630358043368,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004411417523215105,
      "loss": 5.1299,
      "mean_token_accuracy": 0.1669689252972603,
      "num_tokens": 6857290.0,
      "step": 3100
    },
    {
      "entropy": 5.386426877975464,
      "epoch": 3.130105900151286,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004408740009633665,
      "loss": 5.2361,
      "mean_token_accuracy": 0.16384147852659225,
      "num_tokens": 6870060.0,
      "step": 3105
    },
    {
      "entropy": 5.56688084602356,
      "epoch": 3.135148764498235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044060573406500117,
      "loss": 5.1397,
      "mean_token_accuracy": 0.15771555602550508,
      "num_tokens": 6880813.0,
      "step": 3110
    },
    {
      "entropy": 5.425288105010987,
      "epoch": 3.140191628845184,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004403369524601941,
      "loss": 5.1509,
      "mean_token_accuracy": 0.17193858474493026,
      "num_tokens": 6891955.0,
      "step": 3115
    },
    {
      "entropy": 5.349575519561768,
      "epoch": 3.145234493192133,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004400676569843249,
      "loss": 5.1303,
      "mean_token_accuracy": 0.16631903350353242,
      "num_tokens": 6902873.0,
      "step": 3120
    },
    {
      "entropy": 5.5210484027862545,
      "epoch": 3.1502773575390823,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00043979784847436994,
      "loss": 5.1483,
      "mean_token_accuracy": 0.16382198929786682,
      "num_tokens": 6916104.0,
      "step": 3125
    },
    {
      "entropy": 5.48253321647644,
      "epoch": 3.1553202218860315,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395275277689004,
      "loss": 5.1654,
      "mean_token_accuracy": 0.16915008276700974,
      "num_tokens": 6926117.0,
      "step": 3130
    },
    {
      "entropy": 5.411616468429566,
      "epoch": 3.16036308623298,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004392566957080793,
      "loss": 5.1602,
      "mean_token_accuracy": 0.16095373928546905,
      "num_tokens": 6936257.0,
      "step": 3135
    },
    {
      "entropy": 5.496952152252197,
      "epoch": 3.1654059505799292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043898535313365884,
      "loss": 5.2959,
      "mean_token_accuracy": 0.15923235267400743,
      "num_tokens": 6948161.0,
      "step": 3140
    },
    {
      "entropy": 5.4647314071655275,
      "epoch": 3.1704488149268784,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000438713500888978,
      "loss": 5.1755,
      "mean_token_accuracy": 0.16748863309621811,
      "num_tokens": 6959014.0,
      "step": 3145
    },
    {
      "entropy": 5.465006303787232,
      "epoch": 3.1754916792738275,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004384411398189599,
      "loss": 5.1443,
      "mean_token_accuracy": 0.16520180255174638,
      "num_tokens": 6969180.0,
      "step": 3150
    },
    {
      "entropy": 5.440118408203125,
      "epoch": 3.1805345436207766,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000438168270770109,
      "loss": 5.2817,
      "mean_token_accuracy": 0.16402920633554458,
      "num_tokens": 6980702.0,
      "step": 3155
    },
    {
      "entropy": 5.530990362167358,
      "epoch": 3.1855774079677257,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004378948945905085,
      "loss": 5.2769,
      "mean_token_accuracy": 0.15514416843652726,
      "num_tokens": 6992904.0,
      "step": 3160
    },
    {
      "entropy": 5.461986017227173,
      "epoch": 3.190620272314675,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043762101212981805,
      "loss": 5.166,
      "mean_token_accuracy": 0.16778829246759414,
      "num_tokens": 7004355.0,
      "step": 3165
    },
    {
      "entropy": 5.458444547653198,
      "epoch": 3.195663136661624,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00043734662423927054,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1590945839881897,
      "num_tokens": 7014941.0,
      "step": 3170
    },
    {
      "entropy": 5.393497896194458,
      "epoch": 3.200706001008573,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004370717317716699,
      "loss": 5.2185,
      "mean_token_accuracy": 0.16269678771495819,
      "num_tokens": 7026208.0,
      "step": 3175
    },
    {
      "entropy": 5.51782603263855,
      "epoch": 3.205748865355522,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004367963355813882,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1580560639500618,
      "num_tokens": 7038065.0,
      "step": 3180
    },
    {
      "entropy": 5.545170831680298,
      "epoch": 3.210791729702471,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043652043652436323,
      "loss": 5.2436,
      "mean_token_accuracy": 0.16035007685422897,
      "num_tokens": 7049887.0,
      "step": 3185
    },
    {
      "entropy": 5.369115400314331,
      "epoch": 3.21583459404942,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004362440354580956,
      "loss": 5.1617,
      "mean_token_accuracy": 0.1672530859708786,
      "num_tokens": 7060056.0,
      "step": 3190
    },
    {
      "entropy": 5.413434743881226,
      "epoch": 3.220877458396369,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004359671332416462,
      "loss": 5.1881,
      "mean_token_accuracy": 0.16316429674625396,
      "num_tokens": 7071979.0,
      "step": 3195
    },
    {
      "entropy": 5.442484617233276,
      "epoch": 3.225920322743318,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004356897307356335,
      "loss": 5.2204,
      "mean_token_accuracy": 0.16305021792650223,
      "num_tokens": 7082534.0,
      "step": 3200
    },
    {
      "entropy": 5.467128324508667,
      "epoch": 3.2309631870902673,
      "grad_norm": 1.0,
      "learning_rate": 0.0004354118288022309,
      "loss": 5.2185,
      "mean_token_accuracy": 0.16607303470373153,
      "num_tokens": 7094737.0,
      "step": 3205
    },
    {
      "entropy": 5.457351493835449,
      "epoch": 3.2360060514372164,
      "grad_norm": 1.125,
      "learning_rate": 0.00043513342830516405,
      "loss": 5.2306,
      "mean_token_accuracy": 0.16553484052419662,
      "num_tokens": 7105483.0,
      "step": 3210
    },
    {
      "entropy": 5.41800708770752,
      "epoch": 3.2410489157841655,
      "grad_norm": 1.09375,
      "learning_rate": 0.000434854530109708,
      "loss": 5.1204,
      "mean_token_accuracy": 0.16772897243499757,
      "num_tokens": 7115110.0,
      "step": 3215
    },
    {
      "entropy": 5.467194128036499,
      "epoch": 3.2460917801311147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043457513508268487,
      "loss": 5.2007,
      "mean_token_accuracy": 0.16288117617368697,
      "num_tokens": 7125216.0,
      "step": 3220
    },
    {
      "entropy": 5.4844239234924315,
      "epoch": 3.2511346444780633,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00043429524409246083,
      "loss": 5.2418,
      "mean_token_accuracy": 0.16317311972379683,
      "num_tokens": 7136457.0,
      "step": 3225
    },
    {
      "entropy": 5.3839198589324955,
      "epoch": 3.2561775088250124,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004340148580089436,
      "loss": 5.218,
      "mean_token_accuracy": 0.16534755378961563,
      "num_tokens": 7147315.0,
      "step": 3230
    },
    {
      "entropy": 5.501992416381836,
      "epoch": 3.2612203731719616,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004337339777035794,
      "loss": 5.1912,
      "mean_token_accuracy": 0.1601543202996254,
      "num_tokens": 7158436.0,
      "step": 3235
    },
    {
      "entropy": 5.515243721008301,
      "epoch": 3.2662632375189107,
      "grad_norm": 1.078125,
      "learning_rate": 0.000433452604049351,
      "loss": 5.2561,
      "mean_token_accuracy": 0.15643265545368196,
      "num_tokens": 7168705.0,
      "step": 3240
    },
    {
      "entropy": 5.427344894409179,
      "epoch": 3.27130610186586,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043317073792077394,
      "loss": 5.1589,
      "mean_token_accuracy": 0.16633531004190444,
      "num_tokens": 7179200.0,
      "step": 3245
    },
    {
      "entropy": 5.479598760604858,
      "epoch": 3.276348966212809,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004328883801938947,
      "loss": 5.2449,
      "mean_token_accuracy": 0.15928739458322524,
      "num_tokens": 7189378.0,
      "step": 3250
    },
    {
      "entropy": 5.471361541748047,
      "epoch": 3.281391830559758,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043260553174628776,
      "loss": 5.1298,
      "mean_token_accuracy": 0.17073576748371125,
      "num_tokens": 7198726.0,
      "step": 3255
    },
    {
      "entropy": 5.450024509429932,
      "epoch": 3.286434694906707,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004323221934570524,
      "loss": 5.2056,
      "mean_token_accuracy": 0.16591388285160064,
      "num_tokens": 7210547.0,
      "step": 3260
    },
    {
      "entropy": 5.345797157287597,
      "epoch": 3.2914775592536563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043203836620681074,
      "loss": 5.1179,
      "mean_token_accuracy": 0.17085089385509492,
      "num_tokens": 7221871.0,
      "step": 3265
    },
    {
      "entropy": 5.473985004425049,
      "epoch": 3.2965204236006054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043175405087770435,
      "loss": 5.204,
      "mean_token_accuracy": 0.16670676320791245,
      "num_tokens": 7233928.0,
      "step": 3270
    },
    {
      "entropy": 5.407733964920044,
      "epoch": 3.301563287947554,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043146924835339186,
      "loss": 5.2319,
      "mean_token_accuracy": 0.1593266710639,
      "num_tokens": 7245263.0,
      "step": 3275
    },
    {
      "entropy": 5.585802793502808,
      "epoch": 3.306606152294503,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043118395951904605,
      "loss": 5.291,
      "mean_token_accuracy": 0.16548178642988204,
      "num_tokens": 7257941.0,
      "step": 3280
    },
    {
      "entropy": 5.458014726638794,
      "epoch": 3.3116490166414523,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004308981852613512,
      "loss": 5.1987,
      "mean_token_accuracy": 0.16094456762075424,
      "num_tokens": 7267854.0,
      "step": 3285
    },
    {
      "entropy": 5.418560647964478,
      "epoch": 3.3166918809884014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004306119264685003,
      "loss": 5.2602,
      "mean_token_accuracy": 0.16347412019968033,
      "num_tokens": 7279898.0,
      "step": 3290
    },
    {
      "entropy": 5.524970197677613,
      "epoch": 3.3217347453353505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303251840301923,
      "loss": 5.2706,
      "mean_token_accuracy": 0.1569972738623619,
      "num_tokens": 7291459.0,
      "step": 3295
    },
    {
      "entropy": 5.534268474578857,
      "epoch": 3.3267776096822996,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043003795883762933,
      "loss": 5.2914,
      "mean_token_accuracy": 0.16196816489100457,
      "num_tokens": 7302709.0,
      "step": 3300
    },
    {
      "entropy": 5.408970832824707,
      "epoch": 3.3318204740292487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042975025178351386,
      "loss": 5.1981,
      "mean_token_accuracy": 0.1602844253182411,
      "num_tokens": 7312541.0,
      "step": 3305
    },
    {
      "entropy": 5.445866823196411,
      "epoch": 3.336863338376198,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000429462063762046,
      "loss": 5.2243,
      "mean_token_accuracy": 0.1612634390592575,
      "num_tokens": 7323444.0,
      "step": 3310
    },
    {
      "entropy": 5.453121614456177,
      "epoch": 3.3419062027231465,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042917339566892095,
      "loss": 5.1671,
      "mean_token_accuracy": 0.1690024495124817,
      "num_tokens": 7334751.0,
      "step": 3315
    },
    {
      "entropy": 5.4250325679779055,
      "epoch": 3.3469490670700957,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004288842484013257,
      "loss": 5.259,
      "mean_token_accuracy": 0.15966995507478715,
      "num_tokens": 7347434.0,
      "step": 3320
    },
    {
      "entropy": 5.454699468612671,
      "epoch": 3.3519919314170448,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004285946228579365,
      "loss": 5.2701,
      "mean_token_accuracy": 0.1651109591126442,
      "num_tokens": 7358240.0,
      "step": 3325
    },
    {
      "entropy": 5.488889122009278,
      "epoch": 3.357034795763994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004283045199389164,
      "loss": 5.237,
      "mean_token_accuracy": 0.1610963061451912,
      "num_tokens": 7369542.0,
      "step": 3330
    },
    {
      "entropy": 5.450783538818359,
      "epoch": 3.362077660110943,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00042801394054591197,
      "loss": 5.2791,
      "mean_token_accuracy": 0.1544570133090019,
      "num_tokens": 7381885.0,
      "step": 3335
    },
    {
      "entropy": 5.425880765914917,
      "epoch": 3.367120524457892,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042772288558205057,
      "loss": 5.1737,
      "mean_token_accuracy": 0.16952318847179412,
      "num_tokens": 7392042.0,
      "step": 3340
    },
    {
      "entropy": 5.509827041625977,
      "epoch": 3.3721633888048412,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004274313559519377,
      "loss": 5.2513,
      "mean_token_accuracy": 0.16106278598308563,
      "num_tokens": 7403388.0,
      "step": 3345
    },
    {
      "entropy": 5.423347806930542,
      "epoch": 3.3772062531517903,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004271393525616544,
      "loss": 5.1632,
      "mean_token_accuracy": 0.16514763683080674,
      "num_tokens": 7413501.0,
      "step": 3350
    },
    {
      "entropy": 5.396876478195191,
      "epoch": 3.3822491174987395,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00042684687631875377,
      "loss": 5.0678,
      "mean_token_accuracy": 0.17176373898983002,
      "num_tokens": 7424529.0,
      "step": 3355
    },
    {
      "entropy": 5.384559059143067,
      "epoch": 3.3872919818456886,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042655392813225864,
      "loss": 5.242,
      "mean_token_accuracy": 0.16137355268001558,
      "num_tokens": 7434551.0,
      "step": 3360
    },
    {
      "entropy": 5.456484985351563,
      "epoch": 3.3923348461926373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042626050891265884,
      "loss": 5.2433,
      "mean_token_accuracy": 0.16101209819316864,
      "num_tokens": 7443898.0,
      "step": 3365
    },
    {
      "entropy": 5.458910989761352,
      "epoch": 3.3973777105395864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004259666195719081,
      "loss": 5.2172,
      "mean_token_accuracy": 0.1675860896706581,
      "num_tokens": 7454399.0,
      "step": 3370
    },
    {
      "entropy": 5.471589994430542,
      "epoch": 3.4024205748865355,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042567226102342134,
      "loss": 5.2639,
      "mean_token_accuracy": 0.16877782642841338,
      "num_tokens": 7465564.0,
      "step": 3375
    },
    {
      "entropy": 5.4233544826507565,
      "epoch": 3.4074634392334846,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004253774341820715,
      "loss": 5.2163,
      "mean_token_accuracy": 0.1603696510195732,
      "num_tokens": 7476619.0,
      "step": 3380
    },
    {
      "entropy": 5.406573724746704,
      "epoch": 3.4125063035804337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004250821399641874,
      "loss": 5.1855,
      "mean_token_accuracy": 0.16541039049625397,
      "num_tokens": 7487175.0,
      "step": 3385
    },
    {
      "entropy": 5.4396261215209964,
      "epoch": 3.417549167927383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004247863792875502,
      "loss": 5.1712,
      "mean_token_accuracy": 0.16160834431648255,
      "num_tokens": 7498152.0,
      "step": 3390
    },
    {
      "entropy": 5.5115687370300295,
      "epoch": 3.422592032274332,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042449015307139076,
      "loss": 5.1836,
      "mean_token_accuracy": 0.17419951260089875,
      "num_tokens": 7507914.0,
      "step": 3395
    },
    {
      "entropy": 5.383336353302002,
      "epoch": 3.427634896621281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042419346223638725,
      "loss": 5.2856,
      "mean_token_accuracy": 0.1578053817152977,
      "num_tokens": 7518941.0,
      "step": 3400
    },
    {
      "entropy": 5.3913439273834225,
      "epoch": 3.4326777609682297,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004238963077046616,
      "loss": 5.1124,
      "mean_token_accuracy": 0.16630617678165435,
      "num_tokens": 7529315.0,
      "step": 3405
    },
    {
      "entropy": 5.428023338317871,
      "epoch": 3.437720625315179,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004235986903997767,
      "loss": 5.2075,
      "mean_token_accuracy": 0.1669050171971321,
      "num_tokens": 7539888.0,
      "step": 3410
    },
    {
      "entropy": 5.478060293197632,
      "epoch": 3.442763489662128,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042330061124673437,
      "loss": 5.1955,
      "mean_token_accuracy": 0.16277887374162675,
      "num_tokens": 7550429.0,
      "step": 3415
    },
    {
      "entropy": 5.441017913818359,
      "epoch": 3.447806354009077,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042300207117197116,
      "loss": 5.1788,
      "mean_token_accuracy": 0.16205011457204818,
      "num_tokens": 7561051.0,
      "step": 3420
    },
    {
      "entropy": 5.431795597076416,
      "epoch": 3.452849218356026,
      "grad_norm": 1.125,
      "learning_rate": 0.00042270307110335657,
      "loss": 5.189,
      "mean_token_accuracy": 0.16427184343338014,
      "num_tokens": 7570847.0,
      "step": 3425
    },
    {
      "entropy": 5.370966148376465,
      "epoch": 3.4578920827029753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004224036119701899,
      "loss": 5.1816,
      "mean_token_accuracy": 0.16554466038942336,
      "num_tokens": 7581515.0,
      "step": 3430
    },
    {
      "entropy": 5.356352710723877,
      "epoch": 3.4629349470499244,
      "grad_norm": 0.953125,
      "learning_rate": 0.00042210369470319694,
      "loss": 5.1995,
      "mean_token_accuracy": 0.16370873153209686,
      "num_tokens": 7592306.0,
      "step": 3435
    },
    {
      "entropy": 5.504908418655395,
      "epoch": 3.4679778113968736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004218033202345274,
      "loss": 5.301,
      "mean_token_accuracy": 0.15796938687562942,
      "num_tokens": 7603264.0,
      "step": 3440
    },
    {
      "entropy": 5.465123796463013,
      "epoch": 3.4730206757438227,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004215024894977522,
      "loss": 5.2634,
      "mean_token_accuracy": 0.16270803660154343,
      "num_tokens": 7615603.0,
      "step": 3445
    },
    {
      "entropy": 5.411271572113037,
      "epoch": 3.478063540090772,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004212012034278601,
      "loss": 5.1508,
      "mean_token_accuracy": 0.17226357161998748,
      "num_tokens": 7626072.0,
      "step": 3450
    },
    {
      "entropy": 5.46341233253479,
      "epoch": 3.4831064044377205,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042089946296125513,
      "loss": 5.2184,
      "mean_token_accuracy": 0.16225174367427825,
      "num_tokens": 7636321.0,
      "step": 3455
    },
    {
      "entropy": 5.427269840240479,
      "epoch": 3.4881492687846696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042059726903575373,
      "loss": 5.1926,
      "mean_token_accuracy": 0.16928152441978456,
      "num_tokens": 7647709.0,
      "step": 3460
    },
    {
      "entropy": 5.445364999771118,
      "epoch": 3.4931921331316187,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004202946225905815,
      "loss": 5.1903,
      "mean_token_accuracy": 0.16657970994710922,
      "num_tokens": 7660195.0,
      "step": 3465
    },
    {
      "entropy": 5.401914834976196,
      "epoch": 3.498234997478568,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004199915245663706,
      "loss": 5.146,
      "mean_token_accuracy": 0.17144086360931396,
      "num_tokens": 7670274.0,
      "step": 3470
    },
    {
      "entropy": 5.493826055526734,
      "epoch": 3.503277861825517,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041968797590515673,
      "loss": 5.2115,
      "mean_token_accuracy": 0.161667200922966,
      "num_tokens": 7680327.0,
      "step": 3475
    },
    {
      "entropy": 5.383252143859863,
      "epoch": 3.508320726172466,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004193839775503761,
      "loss": 5.2211,
      "mean_token_accuracy": 0.16607730090618134,
      "num_tokens": 7691250.0,
      "step": 3480
    },
    {
      "entropy": 5.408378458023071,
      "epoch": 3.513363590519415,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00041907953044686253,
      "loss": 5.2517,
      "mean_token_accuracy": 0.16372339129447938,
      "num_tokens": 7703726.0,
      "step": 3485
    },
    {
      "entropy": 5.4567506313323975,
      "epoch": 3.518406454866364,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004187746355408448,
      "loss": 5.2795,
      "mean_token_accuracy": 0.16365584284067153,
      "num_tokens": 7715179.0,
      "step": 3490
    },
    {
      "entropy": 5.508245992660522,
      "epoch": 3.523449319213313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004184692937799431,
      "loss": 5.2743,
      "mean_token_accuracy": 0.16309838443994523,
      "num_tokens": 7726586.0,
      "step": 3495
    },
    {
      "entropy": 5.409369802474975,
      "epoch": 3.528492183560262,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041816350611316693,
      "loss": 5.2116,
      "mean_token_accuracy": 0.1633657142519951,
      "num_tokens": 7737350.0,
      "step": 3500
    },
    {
      "epoch": 3.528492183560262,
      "eval_entropy": 5.288684311107218,
      "eval_loss": 5.805794715881348,
      "eval_mean_token_accuracy": 0.14353060472102566,
      "eval_num_tokens": 7737350.0,
      "eval_runtime": 2.639,
      "eval_samples_per_second": 1189.077,
      "eval_steps_per_second": 148.919,
      "step": 3500
    },
    {
      "entropy": 5.396514368057251,
      "epoch": 3.533535047907211,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041785727349091117,
      "loss": 5.2034,
      "mean_token_accuracy": 0.16831117123365402,
      "num_tokens": 7749586.0,
      "step": 3505
    },
    {
      "entropy": 5.383462715148926,
      "epoch": 3.5385779122541603,
      "grad_norm": 1.03125,
      "learning_rate": 0.000417550596864954,
      "loss": 5.2043,
      "mean_token_accuracy": 0.1685790702700615,
      "num_tokens": 7761889.0,
      "step": 3510
    },
    {
      "entropy": 5.432257461547851,
      "epoch": 3.5436207766011094,
      "grad_norm": 1.125,
      "learning_rate": 0.00041724347718845344,
      "loss": 5.2018,
      "mean_token_accuracy": 0.16528122276067733,
      "num_tokens": 7772088.0,
      "step": 3515
    },
    {
      "entropy": 5.431580591201782,
      "epoch": 3.5486636409480585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041693591541594455,
      "loss": 5.2121,
      "mean_token_accuracy": 0.16515803933143616,
      "num_tokens": 7783087.0,
      "step": 3520
    },
    {
      "entropy": 5.400111246109009,
      "epoch": 3.5537065052950076,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004166279125033363,
      "loss": 5.1533,
      "mean_token_accuracy": 0.16039139032363892,
      "num_tokens": 7792919.0,
      "step": 3525
    },
    {
      "entropy": 5.350706243515015,
      "epoch": 3.5587493696419568,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00041631946940790893,
      "loss": 5.1893,
      "mean_token_accuracy": 0.16541917324066163,
      "num_tokens": 7803905.0,
      "step": 3530
    },
    {
      "entropy": 5.476314353942871,
      "epoch": 3.563792233988906,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041601058708831077,
      "loss": 5.2159,
      "mean_token_accuracy": 0.16536158919334412,
      "num_tokens": 7815651.0,
      "step": 3535
    },
    {
      "entropy": 5.398756647109986,
      "epoch": 3.568835098335855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000415701266504555,
      "loss": 5.1996,
      "mean_token_accuracy": 0.16630251854658126,
      "num_tokens": 7827247.0,
      "step": 3540
    },
    {
      "entropy": 5.432639455795288,
      "epoch": 3.5738779626828037,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004153915086180173,
      "loss": 5.2273,
      "mean_token_accuracy": 0.16272253841161727,
      "num_tokens": 7838461.0,
      "step": 3545
    },
    {
      "entropy": 5.353623676300049,
      "epoch": 3.578920827029753,
      "grad_norm": 1.0,
      "learning_rate": 0.0004150813143914323,
      "loss": 5.1592,
      "mean_token_accuracy": 0.16748233139514923,
      "num_tokens": 7849824.0,
      "step": 3550
    },
    {
      "entropy": 5.3674060821533205,
      "epoch": 3.583963691376702,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004147706847888908,
      "loss": 5.249,
      "mean_token_accuracy": 0.16401712149381636,
      "num_tokens": 7862167.0,
      "step": 3555
    },
    {
      "entropy": 5.460422658920288,
      "epoch": 3.589006555723651,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004144596207758367,
      "loss": 5.2182,
      "mean_token_accuracy": 0.164128115773201,
      "num_tokens": 7872320.0,
      "step": 3560
    },
    {
      "entropy": 5.418314218521118,
      "epoch": 3.5940494200706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004141481233190642,
      "loss": 5.1856,
      "mean_token_accuracy": 0.17166522443294524,
      "num_tokens": 7883150.0,
      "step": 3565
    },
    {
      "entropy": 5.3734376430511475,
      "epoch": 3.5990922844175492,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004138361933867145,
      "loss": 5.2621,
      "mean_token_accuracy": 0.15922135561704637,
      "num_tokens": 7895371.0,
      "step": 3570
    },
    {
      "entropy": 5.366262292861938,
      "epoch": 3.6041351487644984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004135238319482731,
      "loss": 5.1199,
      "mean_token_accuracy": 0.17200247198343277,
      "num_tokens": 7905669.0,
      "step": 3575
    },
    {
      "entropy": 5.384713935852051,
      "epoch": 3.609178013111447,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004132110399745666,
      "loss": 5.14,
      "mean_token_accuracy": 0.16676151305437087,
      "num_tokens": 7915714.0,
      "step": 3580
    },
    {
      "entropy": 5.372761964797974,
      "epoch": 3.614220877458396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004128978184377596,
      "loss": 5.2178,
      "mean_token_accuracy": 0.16233056485652925,
      "num_tokens": 7927110.0,
      "step": 3585
    },
    {
      "entropy": 5.463270425796509,
      "epoch": 3.6192637418053453,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004125841683113518,
      "loss": 5.2482,
      "mean_token_accuracy": 0.1620067447423935,
      "num_tokens": 7939170.0,
      "step": 3590
    },
    {
      "entropy": 5.435369348526001,
      "epoch": 3.6243066061522944,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00041227009057017523,
      "loss": 5.2776,
      "mean_token_accuracy": 0.15749395340681077,
      "num_tokens": 7952615.0,
      "step": 3595
    },
    {
      "entropy": 5.4100950241088865,
      "epoch": 3.6293494704992435,
      "grad_norm": 1.0,
      "learning_rate": 0.0004119555861903906,
      "loss": 5.2362,
      "mean_token_accuracy": 0.1592637062072754,
      "num_tokens": 7963946.0,
      "step": 3600
    },
    {
      "entropy": 5.429463863372803,
      "epoch": 3.6343923348461926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004116406561494847,
      "loss": 5.1996,
      "mean_token_accuracy": 0.1630503699183464,
      "num_tokens": 7975270.0,
      "step": 3605
    },
    {
      "entropy": 5.3208833694458,
      "epoch": 3.6394351991931417,
      "grad_norm": 1.015625,
      "learning_rate": 0.00041132530142626756,
      "loss": 5.1475,
      "mean_token_accuracy": 0.16887623071670532,
      "num_tokens": 7986008.0,
      "step": 3610
    },
    {
      "entropy": 5.435577297210694,
      "epoch": 3.644478063540091,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00041100952300086884,
      "loss": 5.3155,
      "mean_token_accuracy": 0.16070480048656463,
      "num_tokens": 7997416.0,
      "step": 3615
    },
    {
      "entropy": 5.420179557800293,
      "epoch": 3.64952092788704,
      "grad_norm": 0.953125,
      "learning_rate": 0.00041069332185473514,
      "loss": 5.1047,
      "mean_token_accuracy": 0.17083795964717866,
      "num_tokens": 8009120.0,
      "step": 3620
    },
    {
      "entropy": 5.393123483657837,
      "epoch": 3.654563792233989,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000410376698970627,
      "loss": 5.225,
      "mean_token_accuracy": 0.16107127219438552,
      "num_tokens": 8020325.0,
      "step": 3625
    },
    {
      "entropy": 5.374727487564087,
      "epoch": 3.659606656580938,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041005965533261554,
      "loss": 5.1692,
      "mean_token_accuracy": 0.16224971562623977,
      "num_tokens": 8031456.0,
      "step": 3630
    },
    {
      "entropy": 5.425358057022095,
      "epoch": 3.6646495209278873,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040974219192607973,
      "loss": 5.2134,
      "mean_token_accuracy": 0.16621486246585845,
      "num_tokens": 8042823.0,
      "step": 3635
    },
    {
      "entropy": 5.4966999053955075,
      "epoch": 3.669692385274836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040942430973770323,
      "loss": 5.3017,
      "mean_token_accuracy": 0.16085462123155594,
      "num_tokens": 8054516.0,
      "step": 3640
    },
    {
      "entropy": 5.421261405944824,
      "epoch": 3.674735249621785,
      "grad_norm": 1.0,
      "learning_rate": 0.0004091060097554712,
      "loss": 5.1837,
      "mean_token_accuracy": 0.1668559268116951,
      "num_tokens": 8064783.0,
      "step": 3645
    },
    {
      "entropy": 5.371801424026489,
      "epoch": 3.679778113968734,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004087872929686673,
      "loss": 5.1648,
      "mean_token_accuracy": 0.16690610349178314,
      "num_tokens": 8076085.0,
      "step": 3650
    },
    {
      "entropy": 5.4291423797607425,
      "epoch": 3.6848209783156833,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040846816036787056,
      "loss": 5.2886,
      "mean_token_accuracy": 0.15664487779140474,
      "num_tokens": 8088218.0,
      "step": 3655
    },
    {
      "entropy": 5.381323003768921,
      "epoch": 3.6898638426626325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004081486129449527,
      "loss": 5.2096,
      "mean_token_accuracy": 0.16191145330667495,
      "num_tokens": 8099768.0,
      "step": 3660
    },
    {
      "entropy": 5.407671165466309,
      "epoch": 3.6949067070095816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040782865169307423,
      "loss": 5.1982,
      "mean_token_accuracy": 0.16725882291793823,
      "num_tokens": 8109858.0,
      "step": 3665
    },
    {
      "entropy": 5.516709995269776,
      "epoch": 3.6999495713565302,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004075082776066822,
      "loss": 5.315,
      "mean_token_accuracy": 0.16229137778282166,
      "num_tokens": 8120262.0,
      "step": 3670
    },
    {
      "entropy": 5.50908842086792,
      "epoch": 3.7049924357034794,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004071874916815068,
      "loss": 5.249,
      "mean_token_accuracy": 0.16730919629335403,
      "num_tokens": 8131483.0,
      "step": 3675
    },
    {
      "entropy": 5.462291193008423,
      "epoch": 3.7100353000504285,
      "grad_norm": 1.125,
      "learning_rate": 0.00040686629491455787,
      "loss": 5.3096,
      "mean_token_accuracy": 0.15786902457475663,
      "num_tokens": 8143431.0,
      "step": 3680
    },
    {
      "entropy": 5.487177610397339,
      "epoch": 3.7150781643973776,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040654468830412247,
      "loss": 5.3054,
      "mean_token_accuracy": 0.16073084771633148,
      "num_tokens": 8154596.0,
      "step": 3685
    },
    {
      "entropy": 5.393495893478393,
      "epoch": 3.7201210287443267,
      "grad_norm": 0.921875,
      "learning_rate": 0.00040622267284976134,
      "loss": 5.285,
      "mean_token_accuracy": 0.1602349177002907,
      "num_tokens": 8165509.0,
      "step": 3690
    },
    {
      "entropy": 5.4206235885620115,
      "epoch": 3.725163893091276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040590024955230596,
      "loss": 5.1794,
      "mean_token_accuracy": 0.16264848709106444,
      "num_tokens": 8176275.0,
      "step": 3695
    },
    {
      "entropy": 5.384602308273315,
      "epoch": 3.730206757438225,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004055774194138554,
      "loss": 5.2701,
      "mean_token_accuracy": 0.1620113343000412,
      "num_tokens": 8187820.0,
      "step": 3700
    },
    {
      "entropy": 5.361208868026734,
      "epoch": 3.735249621785174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004052541834377731,
      "loss": 5.1528,
      "mean_token_accuracy": 0.16657100170850753,
      "num_tokens": 8197750.0,
      "step": 3705
    },
    {
      "entropy": 5.421066045761108,
      "epoch": 3.740292486132123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040493054262868397,
      "loss": 5.2555,
      "mean_token_accuracy": 0.16027550846338273,
      "num_tokens": 8209275.0,
      "step": 3710
    },
    {
      "entropy": 5.420915985107422,
      "epoch": 3.7453353504790723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040460649799247116,
      "loss": 5.3062,
      "mean_token_accuracy": 0.15899969637393951,
      "num_tokens": 8221280.0,
      "step": 3715
    },
    {
      "entropy": 5.433041524887085,
      "epoch": 3.7503782148260214,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004042820505362728,
      "loss": 5.1886,
      "mean_token_accuracy": 0.16738417595624924,
      "num_tokens": 8231679.0,
      "step": 3720
    },
    {
      "entropy": 5.381321430206299,
      "epoch": 3.7554210791729705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004039572012684792,
      "loss": 5.1863,
      "mean_token_accuracy": 0.16833597272634507,
      "num_tokens": 8242014.0,
      "step": 3725
    },
    {
      "entropy": 5.396518754959106,
      "epoch": 3.760463943519919,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040363195119872916,
      "loss": 5.1676,
      "mean_token_accuracy": 0.16874904185533524,
      "num_tokens": 8253128.0,
      "step": 3730
    },
    {
      "entropy": 5.333758163452148,
      "epoch": 3.7655068078668683,
      "grad_norm": 1.015625,
      "learning_rate": 0.00040330630133790757,
      "loss": 5.1154,
      "mean_token_accuracy": 0.17616755962371827,
      "num_tokens": 8264643.0,
      "step": 3735
    },
    {
      "entropy": 5.437575769424439,
      "epoch": 3.7705496722138174,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040298025269814165,
      "loss": 5.2636,
      "mean_token_accuracy": 0.16203719973564149,
      "num_tokens": 8275009.0,
      "step": 3740
    },
    {
      "entropy": 5.373467350006104,
      "epoch": 3.7755925365607665,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004026538062927981,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1589460164308548,
      "num_tokens": 8286321.0,
      "step": 3745
    },
    {
      "entropy": 5.402272272109985,
      "epoch": 3.7806354009077157,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040232696313647985,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1603735625743866,
      "num_tokens": 8297216.0,
      "step": 3750
    },
    {
      "entropy": 5.46411166191101,
      "epoch": 3.7856782652546648,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004019997242450231,
      "loss": 5.2181,
      "mean_token_accuracy": 0.160633085668087,
      "num_tokens": 8309053.0,
      "step": 3755
    },
    {
      "entropy": 5.400403308868408,
      "epoch": 3.7907211296016134,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004016720906354937,
      "loss": 5.2146,
      "mean_token_accuracy": 0.16017768532037735,
      "num_tokens": 8318840.0,
      "step": 3760
    },
    {
      "entropy": 5.415168762207031,
      "epoch": 3.7957639939485626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004013440633261845,
      "loss": 5.2165,
      "mean_token_accuracy": 0.16871240139007568,
      "num_tokens": 8329127.0,
      "step": 3765
    },
    {
      "entropy": 5.409310340881348,
      "epoch": 3.8008068582955117,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00040101564333661213,
      "loss": 5.1986,
      "mean_token_accuracy": 0.167653651535511,
      "num_tokens": 8340651.0,
      "step": 3770
    },
    {
      "entropy": 5.35351710319519,
      "epoch": 3.805849722642461,
      "grad_norm": 1.125,
      "learning_rate": 0.00040068683168751316,
      "loss": 5.2088,
      "mean_token_accuracy": 0.16815593391656875,
      "num_tokens": 8351399.0,
      "step": 3775
    },
    {
      "entropy": 5.4036847114562985,
      "epoch": 3.81089258698941,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040035762940084205,
      "loss": 5.1883,
      "mean_token_accuracy": 0.16491729170084,
      "num_tokens": 8361756.0,
      "step": 3780
    },
    {
      "entropy": 5.446819829940796,
      "epoch": 3.815935451336359,
      "grad_norm": 1.0,
      "learning_rate": 0.000400028037499767,
      "loss": 5.2239,
      "mean_token_accuracy": 0.16567778438329697,
      "num_tokens": 8372408.0,
      "step": 3785
    },
    {
      "entropy": 5.3853254318237305,
      "epoch": 3.820978315683308,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003996980570086673,
      "loss": 5.1576,
      "mean_token_accuracy": 0.1717057928442955,
      "num_tokens": 8382788.0,
      "step": 3790
    },
    {
      "entropy": 5.3489830493927,
      "epoch": 3.8260211800302573,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039936768895312987,
      "loss": 5.2317,
      "mean_token_accuracy": 0.16352340281009675,
      "num_tokens": 8393693.0,
      "step": 3795
    },
    {
      "entropy": 5.392658329010009,
      "epoch": 3.8310640443772064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039903693435994626,
      "loss": 5.2109,
      "mean_token_accuracy": 0.17214041948318481,
      "num_tokens": 8405163.0,
      "step": 3800
    },
    {
      "entropy": 5.402809000015258,
      "epoch": 3.8361069087241555,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00039870579425710937,
      "loss": 5.1963,
      "mean_token_accuracy": 0.16173276156187058,
      "num_tokens": 8416987.0,
      "step": 3805
    },
    {
      "entropy": 5.363797426223755,
      "epoch": 3.8411497730711046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003983742696738103,
      "loss": 5.2177,
      "mean_token_accuracy": 0.1640700489282608,
      "num_tokens": 8428864.0,
      "step": 3810
    },
    {
      "entropy": 5.416312217712402,
      "epoch": 3.8461926374180537,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003980423616404351,
      "loss": 5.1506,
      "mean_token_accuracy": 0.16375678032636642,
      "num_tokens": 8438970.0,
      "step": 3815
    },
    {
      "entropy": 5.49581208229065,
      "epoch": 3.8512355017650024,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003977100711885616,
      "loss": 5.304,
      "mean_token_accuracy": 0.15686767250299455,
      "num_tokens": 8448935.0,
      "step": 3820
    },
    {
      "entropy": 5.418037557601929,
      "epoch": 3.8562783661119515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039737739935095614,
      "loss": 5.2484,
      "mean_token_accuracy": 0.16554856598377227,
      "num_tokens": 8458908.0,
      "step": 3825
    },
    {
      "entropy": 5.415468406677246,
      "epoch": 3.8613212304589006,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003970443471615705,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1734945371747017,
      "num_tokens": 8469161.0,
      "step": 3830
    },
    {
      "entropy": 5.352152872085571,
      "epoch": 3.8663640948058497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003967109156555386,
      "loss": 5.1665,
      "mean_token_accuracy": 0.16743157356977462,
      "num_tokens": 8479821.0,
      "step": 3835
    },
    {
      "entropy": 5.417977142333984,
      "epoch": 3.871406959152799,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039637710586917324,
      "loss": 5.2605,
      "mean_token_accuracy": 0.16112958639860153,
      "num_tokens": 8489837.0,
      "step": 3840
    },
    {
      "entropy": 5.440296983718872,
      "epoch": 3.876449823499748,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039604291883996287,
      "loss": 5.2416,
      "mean_token_accuracy": 0.16606980115175246,
      "num_tokens": 8501241.0,
      "step": 3845
    },
    {
      "entropy": 5.442011117935181,
      "epoch": 3.8814926878466967,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039570835560656854,
      "loss": 5.2667,
      "mean_token_accuracy": 0.16614895313978195,
      "num_tokens": 8512558.0,
      "step": 3850
    },
    {
      "entropy": 5.3762805461883545,
      "epoch": 3.8865355521936458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003953734172088205,
      "loss": 5.1348,
      "mean_token_accuracy": 0.16708417236804962,
      "num_tokens": 8522870.0,
      "step": 3855
    },
    {
      "entropy": 5.390462827682495,
      "epoch": 3.891578416540595,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039503810468771494,
      "loss": 5.2317,
      "mean_token_accuracy": 0.16003523766994476,
      "num_tokens": 8533749.0,
      "step": 3860
    },
    {
      "entropy": 5.454382276535034,
      "epoch": 3.896621280887544,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003947024190854111,
      "loss": 5.2399,
      "mean_token_accuracy": 0.1711591064929962,
      "num_tokens": 8544700.0,
      "step": 3865
    },
    {
      "entropy": 5.412538242340088,
      "epoch": 3.901664145234493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039436636144522737,
      "loss": 5.2522,
      "mean_token_accuracy": 0.15667204558849335,
      "num_tokens": 8555236.0,
      "step": 3870
    },
    {
      "entropy": 5.39568943977356,
      "epoch": 3.9067070095814422,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039402993281163875,
      "loss": 5.2712,
      "mean_token_accuracy": 0.1599903002381325,
      "num_tokens": 8566597.0,
      "step": 3875
    },
    {
      "entropy": 5.431813669204712,
      "epoch": 3.9117498739283914,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003936931342302732,
      "loss": 5.2221,
      "mean_token_accuracy": 0.15816849023103713,
      "num_tokens": 8577297.0,
      "step": 3880
    },
    {
      "entropy": 5.474480056762696,
      "epoch": 3.9167927382753405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003933559667479084,
      "loss": 5.2422,
      "mean_token_accuracy": 0.16582781374454497,
      "num_tokens": 8587145.0,
      "step": 3885
    },
    {
      "entropy": 5.359107446670532,
      "epoch": 3.9218356026222896,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039301843141246876,
      "loss": 5.1215,
      "mean_token_accuracy": 0.17119301855564117,
      "num_tokens": 8597410.0,
      "step": 3890
    },
    {
      "entropy": 5.380753707885742,
      "epoch": 3.9268784669692387,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039268052927302185,
      "loss": 5.1872,
      "mean_token_accuracy": 0.16905289590358735,
      "num_tokens": 8607839.0,
      "step": 3895
    },
    {
      "entropy": 5.393512725830078,
      "epoch": 3.931921331316188,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003923422613797753,
      "loss": 5.1694,
      "mean_token_accuracy": 0.16536030620336534,
      "num_tokens": 8618517.0,
      "step": 3900
    },
    {
      "entropy": 5.32657642364502,
      "epoch": 3.936964195663137,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003920036287840734,
      "loss": 5.1453,
      "mean_token_accuracy": 0.17355929166078568,
      "num_tokens": 8629120.0,
      "step": 3905
    },
    {
      "entropy": 5.388219213485717,
      "epoch": 3.9420070600100856,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039166463253839434,
      "loss": 5.2074,
      "mean_token_accuracy": 0.17029678523540498,
      "num_tokens": 8640503.0,
      "step": 3910
    },
    {
      "entropy": 5.37358341217041,
      "epoch": 3.9470499243570347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003913252736963462,
      "loss": 5.2029,
      "mean_token_accuracy": 0.17363450974225997,
      "num_tokens": 8652575.0,
      "step": 3915
    },
    {
      "entropy": 5.416703367233277,
      "epoch": 3.952092788703984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000390985553312664,
      "loss": 5.1198,
      "mean_token_accuracy": 0.16988059133291245,
      "num_tokens": 8663646.0,
      "step": 3920
    },
    {
      "entropy": 5.38117995262146,
      "epoch": 3.957135653050933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003906454724432065,
      "loss": 5.2058,
      "mean_token_accuracy": 0.16442517936229706,
      "num_tokens": 8673953.0,
      "step": 3925
    },
    {
      "entropy": 5.350530624389648,
      "epoch": 3.962178517397882,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039030503214495296,
      "loss": 5.214,
      "mean_token_accuracy": 0.16484454721212388,
      "num_tokens": 8685388.0,
      "step": 3930
    },
    {
      "entropy": 5.380281543731689,
      "epoch": 3.967221381744831,
      "grad_norm": 1.125,
      "learning_rate": 0.0003899642334759996,
      "loss": 5.1925,
      "mean_token_accuracy": 0.16580954045057297,
      "num_tokens": 8695205.0,
      "step": 3935
    },
    {
      "entropy": 5.3921104907989506,
      "epoch": 3.97226424609178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003896230774955567,
      "loss": 5.184,
      "mean_token_accuracy": 0.16484033167362214,
      "num_tokens": 8706118.0,
      "step": 3940
    },
    {
      "entropy": 5.371186542510986,
      "epoch": 3.977307110438729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038928156526394476,
      "loss": 5.17,
      "mean_token_accuracy": 0.16910630613565444,
      "num_tokens": 8716926.0,
      "step": 3945
    },
    {
      "entropy": 5.447496032714843,
      "epoch": 3.982349974785678,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003889396978425918,
      "loss": 5.2969,
      "mean_token_accuracy": 0.16213430911302568,
      "num_tokens": 8729409.0,
      "step": 3950
    },
    {
      "entropy": 5.4057355403900145,
      "epoch": 3.987392839132627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003885974762940295,
      "loss": 5.235,
      "mean_token_accuracy": 0.1684113085269928,
      "num_tokens": 8740140.0,
      "step": 3955
    },
    {
      "entropy": 5.4019121646881105,
      "epoch": 3.9924357034795763,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003882549016818905,
      "loss": 5.1196,
      "mean_token_accuracy": 0.16654850393533707,
      "num_tokens": 8749176.0,
      "step": 3960
    },
    {
      "entropy": 5.355155277252197,
      "epoch": 3.9974785678265254,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038791197507090456,
      "loss": 5.2317,
      "mean_token_accuracy": 0.17030496895313263,
      "num_tokens": 8761575.0,
      "step": 3965
    },
    {
      "entropy": 5.3184362517462835,
      "epoch": 4.00201714573878,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00038756869752689527,
      "loss": 5.0291,
      "mean_token_accuracy": 0.17253867619567448,
      "num_tokens": 8771147.0,
      "step": 3970
    },
    {
      "entropy": 5.395121049880982,
      "epoch": 4.007060010085729,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003872250701167775,
      "loss": 4.8855,
      "mean_token_accuracy": 0.18297589272260667,
      "num_tokens": 8781936.0,
      "step": 3975
    },
    {
      "entropy": 5.361803436279297,
      "epoch": 4.012102874432678,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038688109390855297,
      "loss": 4.8943,
      "mean_token_accuracy": 0.18275870084762574,
      "num_tokens": 8791492.0,
      "step": 3980
    },
    {
      "entropy": 5.365276145935058,
      "epoch": 4.017145738779627,
      "grad_norm": 1.125,
      "learning_rate": 0.0003865367699713077,
      "loss": 4.8808,
      "mean_token_accuracy": 0.18234068602323533,
      "num_tokens": 8802552.0,
      "step": 3985
    },
    {
      "entropy": 5.399391603469849,
      "epoch": 4.022188603126576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038619209937520836,
      "loss": 4.9509,
      "mean_token_accuracy": 0.18303546756505967,
      "num_tokens": 8813885.0,
      "step": 3990
    },
    {
      "entropy": 5.3691075325012205,
      "epoch": 4.027231467473525,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038584708319149924,
      "loss": 4.911,
      "mean_token_accuracy": 0.18313807249069214,
      "num_tokens": 8825503.0,
      "step": 3995
    },
    {
      "entropy": 5.408253812789917,
      "epoch": 4.032274331820474,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038550172249249856,
      "loss": 4.9498,
      "mean_token_accuracy": 0.17092643231153487,
      "num_tokens": 8835581.0,
      "step": 4000
    },
    {
      "epoch": 4.032274331820474,
      "eval_entropy": 5.275534075331748,
      "eval_loss": 5.767982482910156,
      "eval_mean_token_accuracy": 0.14656087610178625,
      "eval_num_tokens": 8835581.0,
      "eval_runtime": 2.632,
      "eval_samples_per_second": 1192.261,
      "eval_steps_per_second": 149.318,
      "step": 4000
    },
    {
      "entropy": 5.322012996673584,
      "epoch": 4.0373171961674235,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003851560183515953,
      "loss": 4.8136,
      "mean_token_accuracy": 0.19483197331428528,
      "num_tokens": 8847453.0,
      "step": 4005
    },
    {
      "entropy": 5.329536008834839,
      "epoch": 4.042360060514373,
      "grad_norm": 1.078125,
      "learning_rate": 0.000384809971843246,
      "loss": 4.9626,
      "mean_token_accuracy": 0.18174424320459365,
      "num_tokens": 8858098.0,
      "step": 4010
    },
    {
      "entropy": 5.378550338745117,
      "epoch": 4.047402924861321,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000384463584042971,
      "loss": 4.8909,
      "mean_token_accuracy": 0.18192771077156067,
      "num_tokens": 8869601.0,
      "step": 4015
    },
    {
      "entropy": 5.264380741119385,
      "epoch": 4.05244578920827,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00038411685602735175,
      "loss": 4.8539,
      "mean_token_accuracy": 0.18428861498832702,
      "num_tokens": 8881926.0,
      "step": 4020
    },
    {
      "entropy": 5.3551019668579105,
      "epoch": 4.057488653555219,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038376978887402687,
      "loss": 4.9887,
      "mean_token_accuracy": 0.1788025438785553,
      "num_tokens": 8892714.0,
      "step": 4025
    },
    {
      "entropy": 5.37628664970398,
      "epoch": 4.062531517902168,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003834223836616891,
      "loss": 4.8979,
      "mean_token_accuracy": 0.18789646178483962,
      "num_tokens": 8905407.0,
      "step": 4030
    },
    {
      "entropy": 5.380304574966431,
      "epoch": 4.067574382249117,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038307464147008176,
      "loss": 4.9734,
      "mean_token_accuracy": 0.177698814868927,
      "num_tokens": 8915474.0,
      "step": 4035
    },
    {
      "entropy": 5.347956895828247,
      "epoch": 4.072617246596066,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003827265633799957,
      "loss": 4.9054,
      "mean_token_accuracy": 0.1812581092119217,
      "num_tokens": 8926168.0,
      "step": 4040
    },
    {
      "entropy": 5.398634195327759,
      "epoch": 4.0776601109430155,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003823781504732657,
      "loss": 4.9972,
      "mean_token_accuracy": 0.1739969864487648,
      "num_tokens": 8936995.0,
      "step": 4045
    },
    {
      "entropy": 5.345803499221802,
      "epoch": 4.082702975289965,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003820294038327669,
      "loss": 5.0253,
      "mean_token_accuracy": 0.17090386003255845,
      "num_tokens": 8947675.0,
      "step": 4050
    },
    {
      "entropy": 5.342918539047242,
      "epoch": 4.087745839636914,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003816803245424121,
      "loss": 4.9558,
      "mean_token_accuracy": 0.18279871344566345,
      "num_tokens": 8958486.0,
      "step": 4055
    },
    {
      "entropy": 5.429638242721557,
      "epoch": 4.092788703983863,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003813309136871478,
      "loss": 4.9519,
      "mean_token_accuracy": 0.18057543486356736,
      "num_tokens": 8968208.0,
      "step": 4060
    },
    {
      "entropy": 5.304712104797363,
      "epoch": 4.097831568330812,
      "grad_norm": 1.0,
      "learning_rate": 0.00038098117235295083,
      "loss": 4.8856,
      "mean_token_accuracy": 0.1835058867931366,
      "num_tokens": 8978327.0,
      "step": 4065
    },
    {
      "entropy": 5.297881746292115,
      "epoch": 4.102874432677761,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003806311016268254,
      "loss": 4.993,
      "mean_token_accuracy": 0.18010969460010529,
      "num_tokens": 8990944.0,
      "step": 4070
    },
    {
      "entropy": 5.372862434387207,
      "epoch": 4.10791729702471,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003802807025967995,
      "loss": 4.9426,
      "mean_token_accuracy": 0.1837949797511101,
      "num_tokens": 9002505.0,
      "step": 4075
    },
    {
      "entropy": 5.331052207946778,
      "epoch": 4.112960161371659,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003799299763519213,
      "loss": 4.9228,
      "mean_token_accuracy": 0.17937170416116716,
      "num_tokens": 9013453.0,
      "step": 4080
    },
    {
      "entropy": 5.318354177474975,
      "epoch": 4.118003025718608,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003795789239822561,
      "loss": 4.907,
      "mean_token_accuracy": 0.1838529258966446,
      "num_tokens": 9025318.0,
      "step": 4085
    },
    {
      "entropy": 5.3691548824310305,
      "epoch": 4.1230458900655576,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003792275465788826,
      "loss": 4.8933,
      "mean_token_accuracy": 0.1837931454181671,
      "num_tokens": 9034334.0,
      "step": 4090
    },
    {
      "entropy": 5.284818315505982,
      "epoch": 4.128088754412507,
      "grad_norm": 1.0,
      "learning_rate": 0.00037887584523389013,
      "loss": 4.9489,
      "mean_token_accuracy": 0.18161599189043046,
      "num_tokens": 9046096.0,
      "step": 4095
    },
    {
      "entropy": 5.302836942672729,
      "epoch": 4.133131618759456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037852382104037454,
      "loss": 4.9257,
      "mean_token_accuracy": 0.173905411362648,
      "num_tokens": 9057540.0,
      "step": 4100
    },
    {
      "entropy": 5.359889125823974,
      "epoch": 4.138174483106404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037817147509243497,
      "loss": 4.9046,
      "mean_token_accuracy": 0.18167727887630464,
      "num_tokens": 9068770.0,
      "step": 4105
    },
    {
      "entropy": 5.37003755569458,
      "epoch": 4.143217347453353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003778188084851709,
      "loss": 5.0133,
      "mean_token_accuracy": 0.17433297336101533,
      "num_tokens": 9080026.0,
      "step": 4110
    },
    {
      "entropy": 5.321903944015503,
      "epoch": 4.148260211800302,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037746582231467816,
      "loss": 4.9239,
      "mean_token_accuracy": 0.18804950565099715,
      "num_tokens": 9092013.0,
      "step": 4115
    },
    {
      "entropy": 5.292985630035401,
      "epoch": 4.153303076147251,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003771125176780461,
      "loss": 4.9535,
      "mean_token_accuracy": 0.17493669986724852,
      "num_tokens": 9104739.0,
      "step": 4120
    },
    {
      "entropy": 5.297138357162476,
      "epoch": 4.1583459404942005,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037675889567335345,
      "loss": 4.8874,
      "mean_token_accuracy": 0.18235434740781784,
      "num_tokens": 9116321.0,
      "step": 4125
    },
    {
      "entropy": 5.321177434921265,
      "epoch": 4.16338880484115,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037640495739966573,
      "loss": 4.8746,
      "mean_token_accuracy": 0.1875880092382431,
      "num_tokens": 9126844.0,
      "step": 4130
    },
    {
      "entropy": 5.229148197174072,
      "epoch": 4.168431669188099,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037605070395703124,
      "loss": 4.9068,
      "mean_token_accuracy": 0.1855694830417633,
      "num_tokens": 9137887.0,
      "step": 4135
    },
    {
      "entropy": 5.260153341293335,
      "epoch": 4.173474533535048,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037569613644647783,
      "loss": 4.8833,
      "mean_token_accuracy": 0.18526361286640167,
      "num_tokens": 9148087.0,
      "step": 4140
    },
    {
      "entropy": 5.271249055862427,
      "epoch": 4.178517397881997,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003753412559700094,
      "loss": 4.8474,
      "mean_token_accuracy": 0.1848575919866562,
      "num_tokens": 9159628.0,
      "step": 4145
    },
    {
      "entropy": 5.240087366104126,
      "epoch": 4.183560262228946,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037498606363060285,
      "loss": 4.922,
      "mean_token_accuracy": 0.178703011572361,
      "num_tokens": 9171295.0,
      "step": 4150
    },
    {
      "entropy": 5.298362970352173,
      "epoch": 4.188603126575895,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003746305605322041,
      "loss": 4.8956,
      "mean_token_accuracy": 0.18227899819612503,
      "num_tokens": 9181619.0,
      "step": 4155
    },
    {
      "entropy": 5.225590753555298,
      "epoch": 4.193645990922844,
      "grad_norm": 1.125,
      "learning_rate": 0.00037427474777972477,
      "loss": 4.8782,
      "mean_token_accuracy": 0.18804521411657332,
      "num_tokens": 9192184.0,
      "step": 4160
    },
    {
      "entropy": 5.288710069656372,
      "epoch": 4.198688855269793,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037391862647903943,
      "loss": 4.9808,
      "mean_token_accuracy": 0.17130756825208665,
      "num_tokens": 9201452.0,
      "step": 4165
    },
    {
      "entropy": 5.3450792789459225,
      "epoch": 4.2037317196167425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037356219773698106,
      "loss": 4.9374,
      "mean_token_accuracy": 0.1764432117342949,
      "num_tokens": 9210945.0,
      "step": 4170
    },
    {
      "entropy": 5.231933069229126,
      "epoch": 4.208774583963692,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037320546266133854,
      "loss": 4.8031,
      "mean_token_accuracy": 0.18961086124181747,
      "num_tokens": 9221577.0,
      "step": 4175
    },
    {
      "entropy": 5.26270170211792,
      "epoch": 4.213817448310641,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003728484223608526,
      "loss": 4.9545,
      "mean_token_accuracy": 0.17473041117191315,
      "num_tokens": 9232353.0,
      "step": 4180
    },
    {
      "entropy": 5.352227687835693,
      "epoch": 4.21886031265759,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037249107794521286,
      "loss": 5.0012,
      "mean_token_accuracy": 0.18204287737607955,
      "num_tokens": 9243157.0,
      "step": 4185
    },
    {
      "entropy": 5.302592039108276,
      "epoch": 4.223903177004539,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00037213343052505394,
      "loss": 4.9896,
      "mean_token_accuracy": 0.17643263339996337,
      "num_tokens": 9256750.0,
      "step": 4190
    },
    {
      "entropy": 5.305902051925659,
      "epoch": 4.228946041351487,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003717754812119523,
      "loss": 4.9628,
      "mean_token_accuracy": 0.18209224492311477,
      "num_tokens": 9268436.0,
      "step": 4195
    },
    {
      "entropy": 5.348188829421997,
      "epoch": 4.233988905698436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003714172311184226,
      "loss": 4.958,
      "mean_token_accuracy": 0.17616404294967652,
      "num_tokens": 9278326.0,
      "step": 4200
    },
    {
      "entropy": 5.349838638305664,
      "epoch": 4.2390317700453854,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003710586813579144,
      "loss": 4.9225,
      "mean_token_accuracy": 0.18225255310535432,
      "num_tokens": 9289788.0,
      "step": 4205
    },
    {
      "entropy": 5.338082122802734,
      "epoch": 4.244074634392335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037069983304480875,
      "loss": 4.9114,
      "mean_token_accuracy": 0.17900969237089157,
      "num_tokens": 9300932.0,
      "step": 4210
    },
    {
      "entropy": 5.307391595840454,
      "epoch": 4.249117498739284,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003703406872944144,
      "loss": 4.9921,
      "mean_token_accuracy": 0.18204918503761292,
      "num_tokens": 9313043.0,
      "step": 4215
    },
    {
      "entropy": 5.330126762390137,
      "epoch": 4.254160363086233,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003699812452229645,
      "loss": 4.9955,
      "mean_token_accuracy": 0.178223817050457,
      "num_tokens": 9324122.0,
      "step": 4220
    },
    {
      "entropy": 5.216471004486084,
      "epoch": 4.259203227433182,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036962150794761354,
      "loss": 4.8688,
      "mean_token_accuracy": 0.1861891895532608,
      "num_tokens": 9335745.0,
      "step": 4225
    },
    {
      "entropy": 5.327639532089234,
      "epoch": 4.264246091780131,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003692614765864331,
      "loss": 4.9319,
      "mean_token_accuracy": 0.18076154142618178,
      "num_tokens": 9346843.0,
      "step": 4230
    },
    {
      "entropy": 5.295945644378662,
      "epoch": 4.26928895612708,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036890115225840904,
      "loss": 4.9436,
      "mean_token_accuracy": 0.1751152455806732,
      "num_tokens": 9357782.0,
      "step": 4235
    },
    {
      "entropy": 5.322455215454101,
      "epoch": 4.274331820474029,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003685405360834376,
      "loss": 5.0595,
      "mean_token_accuracy": 0.17047233283519744,
      "num_tokens": 9370704.0,
      "step": 4240
    },
    {
      "entropy": 5.362507629394531,
      "epoch": 4.279374684820978,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036817962918232234,
      "loss": 4.9908,
      "mean_token_accuracy": 0.17333315759897233,
      "num_tokens": 9382174.0,
      "step": 4245
    },
    {
      "entropy": 5.279483604431152,
      "epoch": 4.2844175491679275,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036781843267677,
      "loss": 4.8877,
      "mean_token_accuracy": 0.18702801764011384,
      "num_tokens": 9393380.0,
      "step": 4250
    },
    {
      "entropy": 5.364893865585327,
      "epoch": 4.289460413514877,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003674569476893878,
      "loss": 4.9921,
      "mean_token_accuracy": 0.1790022909641266,
      "num_tokens": 9404194.0,
      "step": 4255
    },
    {
      "entropy": 5.292422771453857,
      "epoch": 4.294503277861826,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003670951753436794,
      "loss": 4.9009,
      "mean_token_accuracy": 0.18903275579214096,
      "num_tokens": 9415488.0,
      "step": 4260
    },
    {
      "entropy": 5.290180206298828,
      "epoch": 4.299546142208775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036673311676404163,
      "loss": 5.001,
      "mean_token_accuracy": 0.17764427214860917,
      "num_tokens": 9426351.0,
      "step": 4265
    },
    {
      "entropy": 5.363708353042602,
      "epoch": 4.304589006555724,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003663707730757608,
      "loss": 4.9882,
      "mean_token_accuracy": 0.1734233245253563,
      "num_tokens": 9438302.0,
      "step": 4270
    },
    {
      "entropy": 5.288618898391723,
      "epoch": 4.309631870902673,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036600814540500956,
      "loss": 4.8812,
      "mean_token_accuracy": 0.18032212257385255,
      "num_tokens": 9449004.0,
      "step": 4275
    },
    {
      "entropy": 5.242237520217896,
      "epoch": 4.314674735249622,
      "grad_norm": 1.03125,
      "learning_rate": 0.000365645234878843,
      "loss": 4.9087,
      "mean_token_accuracy": 0.1851394698023796,
      "num_tokens": 9459624.0,
      "step": 4280
    },
    {
      "entropy": 5.255999660491943,
      "epoch": 4.31971759959657,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036528204262519545,
      "loss": 4.9056,
      "mean_token_accuracy": 0.186630679666996,
      "num_tokens": 9470380.0,
      "step": 4285
    },
    {
      "entropy": 5.286920261383057,
      "epoch": 4.3247604639435195,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003649185697728768,
      "loss": 4.8918,
      "mean_token_accuracy": 0.18538452535867692,
      "num_tokens": 9480802.0,
      "step": 4290
    },
    {
      "entropy": 5.305649852752685,
      "epoch": 4.329803328290469,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036455481745156906,
      "loss": 4.9106,
      "mean_token_accuracy": 0.17623955756425858,
      "num_tokens": 9491842.0,
      "step": 4295
    },
    {
      "entropy": 5.2686515808105465,
      "epoch": 4.334846192637418,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003641907867918228,
      "loss": 4.9081,
      "mean_token_accuracy": 0.1867458403110504,
      "num_tokens": 9503398.0,
      "step": 4300
    },
    {
      "entropy": 5.333186626434326,
      "epoch": 4.339889056984367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003638264789250538,
      "loss": 5.0643,
      "mean_token_accuracy": 0.17986511588096618,
      "num_tokens": 9513915.0,
      "step": 4305
    },
    {
      "entropy": 5.313656663894653,
      "epoch": 4.344931921331316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036346189498353914,
      "loss": 4.9173,
      "mean_token_accuracy": 0.18442089557647706,
      "num_tokens": 9524630.0,
      "step": 4310
    },
    {
      "entropy": 5.275774383544922,
      "epoch": 4.349974785678265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003630970361004143,
      "loss": 4.9138,
      "mean_token_accuracy": 0.18355057686567305,
      "num_tokens": 9535135.0,
      "step": 4315
    },
    {
      "entropy": 5.365457153320312,
      "epoch": 4.355017650025214,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003627319034096691,
      "loss": 4.9636,
      "mean_token_accuracy": 0.1760061115026474,
      "num_tokens": 9544845.0,
      "step": 4320
    },
    {
      "entropy": 5.3485884189605715,
      "epoch": 4.360060514372163,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003623664980461443,
      "loss": 4.9921,
      "mean_token_accuracy": 0.1759295403957367,
      "num_tokens": 9555498.0,
      "step": 4325
    },
    {
      "entropy": 5.26085205078125,
      "epoch": 4.3651033787191125,
      "grad_norm": 0.96875,
      "learning_rate": 0.00036200082114552806,
      "loss": 4.9385,
      "mean_token_accuracy": 0.1820580005645752,
      "num_tokens": 9567995.0,
      "step": 4330
    },
    {
      "entropy": 5.318760681152344,
      "epoch": 4.370146243066062,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036163487384435294,
      "loss": 4.9893,
      "mean_token_accuracy": 0.1768825188279152,
      "num_tokens": 9578199.0,
      "step": 4335
    },
    {
      "entropy": 5.394182205200195,
      "epoch": 4.375189107413011,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003612686572799914,
      "loss": 5.0603,
      "mean_token_accuracy": 0.17752676010131835,
      "num_tokens": 9588633.0,
      "step": 4340
    },
    {
      "entropy": 5.336157178878784,
      "epoch": 4.38023197175996,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003609021725906529,
      "loss": 4.9906,
      "mean_token_accuracy": 0.17957373708486557,
      "num_tokens": 9600269.0,
      "step": 4345
    },
    {
      "entropy": 5.300542211532592,
      "epoch": 4.385274836106909,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036053542091538045,
      "loss": 4.9671,
      "mean_token_accuracy": 0.17539314478635787,
      "num_tokens": 9610699.0,
      "step": 4350
    },
    {
      "entropy": 5.326151561737061,
      "epoch": 4.390317700453858,
      "grad_norm": 1.25,
      "learning_rate": 0.00036016840339404664,
      "loss": 4.9406,
      "mean_token_accuracy": 0.18118180930614472,
      "num_tokens": 9620541.0,
      "step": 4355
    },
    {
      "entropy": 5.265413045883179,
      "epoch": 4.395360564800807,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003598011211673505,
      "loss": 4.8853,
      "mean_token_accuracy": 0.18843996226787568,
      "num_tokens": 9632048.0,
      "step": 4360
    },
    {
      "entropy": 5.293694972991943,
      "epoch": 4.400403429147756,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003594335753768135,
      "loss": 4.9446,
      "mean_token_accuracy": 0.18054304271936417,
      "num_tokens": 9643187.0,
      "step": 4365
    },
    {
      "entropy": 5.303615140914917,
      "epoch": 4.405446293494705,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003590657671647766,
      "loss": 4.9815,
      "mean_token_accuracy": 0.17570459544658662,
      "num_tokens": 9654375.0,
      "step": 4370
    },
    {
      "entropy": 5.328140211105347,
      "epoch": 4.4104891578416545,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035869769767439606,
      "loss": 5.0382,
      "mean_token_accuracy": 0.17248296737670898,
      "num_tokens": 9666702.0,
      "step": 4375
    },
    {
      "entropy": 5.426132678985596,
      "epoch": 4.415532022188603,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003583293680496404,
      "loss": 5.0917,
      "mean_token_accuracy": 0.17236345261335373,
      "num_tokens": 9677989.0,
      "step": 4380
    },
    {
      "entropy": 5.284139394760132,
      "epoch": 4.420574886535552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003579607794352867,
      "loss": 4.9108,
      "mean_token_accuracy": 0.18376108705997468,
      "num_tokens": 9688763.0,
      "step": 4385
    },
    {
      "entropy": 5.285767698287964,
      "epoch": 4.425617750882501,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035759193297691674,
      "loss": 4.9317,
      "mean_token_accuracy": 0.17890386879444123,
      "num_tokens": 9698457.0,
      "step": 4390
    },
    {
      "entropy": 5.268318605422974,
      "epoch": 4.43066061522945,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035722282982091373,
      "loss": 4.9254,
      "mean_token_accuracy": 0.1813647285103798,
      "num_tokens": 9709564.0,
      "step": 4395
    },
    {
      "entropy": 5.273524427413941,
      "epoch": 4.435703479576399,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003568534711144591,
      "loss": 4.9152,
      "mean_token_accuracy": 0.18003549575805664,
      "num_tokens": 9719231.0,
      "step": 4400
    },
    {
      "entropy": 5.32005934715271,
      "epoch": 4.440746343923348,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000356483858005528,
      "loss": 4.9524,
      "mean_token_accuracy": 0.17776781618595122,
      "num_tokens": 9729785.0,
      "step": 4405
    },
    {
      "entropy": 5.235485506057739,
      "epoch": 4.445789208270297,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035611399164288654,
      "loss": 4.8937,
      "mean_token_accuracy": 0.18520702570676803,
      "num_tokens": 9740292.0,
      "step": 4410
    },
    {
      "entropy": 5.31834044456482,
      "epoch": 4.450832072617247,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003557438731760879,
      "loss": 4.9759,
      "mean_token_accuracy": 0.1827839970588684,
      "num_tokens": 9750062.0,
      "step": 4415
    },
    {
      "entropy": 5.287930059432983,
      "epoch": 4.455874936964196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003553735037554689,
      "loss": 4.9682,
      "mean_token_accuracy": 0.17847125977277756,
      "num_tokens": 9760427.0,
      "step": 4420
    },
    {
      "entropy": 5.291294527053833,
      "epoch": 4.460917801311145,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003550028845321461,
      "loss": 4.9969,
      "mean_token_accuracy": 0.1830156534910202,
      "num_tokens": 9771842.0,
      "step": 4425
    },
    {
      "entropy": 5.305868196487427,
      "epoch": 4.465960665658094,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003546320166580127,
      "loss": 4.9186,
      "mean_token_accuracy": 0.1773524835705757,
      "num_tokens": 9782860.0,
      "step": 4430
    },
    {
      "entropy": 5.327865219116211,
      "epoch": 4.471003530005043,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035426090128573464,
      "loss": 4.9592,
      "mean_token_accuracy": 0.17592183649539947,
      "num_tokens": 9793361.0,
      "step": 4435
    },
    {
      "entropy": 5.265003108978272,
      "epoch": 4.476046394351992,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035388953956874686,
      "loss": 4.9105,
      "mean_token_accuracy": 0.17352977097034455,
      "num_tokens": 9805017.0,
      "step": 4440
    },
    {
      "entropy": 5.32849440574646,
      "epoch": 4.481089258698941,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003535179326612504,
      "loss": 5.0301,
      "mean_token_accuracy": 0.17886604517698287,
      "num_tokens": 9816333.0,
      "step": 4445
    },
    {
      "entropy": 5.405245208740235,
      "epoch": 4.48613212304589,
      "grad_norm": 1.015625,
      "learning_rate": 0.00035314608171820787,
      "loss": 5.1009,
      "mean_token_accuracy": 0.1734747365117073,
      "num_tokens": 9828835.0,
      "step": 4450
    },
    {
      "entropy": 5.3323007106781,
      "epoch": 4.4911749873928395,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035277398789534075,
      "loss": 5.027,
      "mean_token_accuracy": 0.18250549584627151,
      "num_tokens": 9839321.0,
      "step": 4455
    },
    {
      "entropy": 5.322406864166259,
      "epoch": 4.496217851739789,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003524016523491252,
      "loss": 4.9469,
      "mean_token_accuracy": 0.18716796189546586,
      "num_tokens": 9849913.0,
      "step": 4460
    },
    {
      "entropy": 5.303274822235108,
      "epoch": 4.501260716086737,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035202907623678865,
      "loss": 4.9651,
      "mean_token_accuracy": 0.17895520031452178,
      "num_tokens": 9861981.0,
      "step": 4465
    },
    {
      "entropy": 5.361339950561524,
      "epoch": 4.506303580433686,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003516562607163062,
      "loss": 5.0074,
      "mean_token_accuracy": 0.17951440215110778,
      "num_tokens": 9873189.0,
      "step": 4470
    },
    {
      "entropy": 5.331658411026001,
      "epoch": 4.511346444780635,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003512832069463971,
      "loss": 5.0003,
      "mean_token_accuracy": 0.17338586896657943,
      "num_tokens": 9884263.0,
      "step": 4475
    },
    {
      "entropy": 5.343599700927735,
      "epoch": 4.516389309127584,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035090991608652125,
      "loss": 5.0695,
      "mean_token_accuracy": 0.1765784353017807,
      "num_tokens": 9894958.0,
      "step": 4480
    },
    {
      "entropy": 5.306997060775757,
      "epoch": 4.521432173474533,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035053638929687504,
      "loss": 4.9235,
      "mean_token_accuracy": 0.18340528011322021,
      "num_tokens": 9906846.0,
      "step": 4485
    },
    {
      "entropy": 5.319222116470337,
      "epoch": 4.526475037821482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003501626277383884,
      "loss": 5.0205,
      "mean_token_accuracy": 0.1750070944428444,
      "num_tokens": 9917696.0,
      "step": 4490
    },
    {
      "entropy": 5.327042055130005,
      "epoch": 4.5315179021684315,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000349788632572721,
      "loss": 5.0285,
      "mean_token_accuracy": 0.17902330458164215,
      "num_tokens": 9928907.0,
      "step": 4495
    },
    {
      "entropy": 5.28035535812378,
      "epoch": 4.536560766515381,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003494144049622582,
      "loss": 4.9361,
      "mean_token_accuracy": 0.18162627071142196,
      "num_tokens": 9939239.0,
      "step": 4500
    },
    {
      "epoch": 4.536560766515381,
      "eval_entropy": 5.213590699601113,
      "eval_loss": 5.7545695304870605,
      "eval_mean_token_accuracy": 0.14845950744155104,
      "eval_num_tokens": 9939239.0,
      "eval_runtime": 2.6313,
      "eval_samples_per_second": 1192.577,
      "eval_steps_per_second": 149.357,
      "step": 4500
    },
    {
      "entropy": 5.324084281921387,
      "epoch": 4.54160363086233,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003490399460701083,
      "loss": 4.9896,
      "mean_token_accuracy": 0.17286743521690368,
      "num_tokens": 9951758.0,
      "step": 4505
    },
    {
      "entropy": 5.379778099060059,
      "epoch": 4.546646495209279,
      "grad_norm": 1.0625,
      "learning_rate": 0.000348665257060098,
      "loss": 4.9843,
      "mean_token_accuracy": 0.18531604409217833,
      "num_tokens": 9962356.0,
      "step": 4510
    },
    {
      "entropy": 5.27449049949646,
      "epoch": 4.551689359556228,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003482903390967695,
      "loss": 5.006,
      "mean_token_accuracy": 0.17380123883485793,
      "num_tokens": 9974601.0,
      "step": 4515
    },
    {
      "entropy": 5.255610466003418,
      "epoch": 4.556732223903177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00034791519334537635,
      "loss": 4.9902,
      "mean_token_accuracy": 0.176007217168808,
      "num_tokens": 9985549.0,
      "step": 4520
    },
    {
      "entropy": 5.240353918075561,
      "epoch": 4.561775088250126,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003475398209718804,
      "loss": 4.9038,
      "mean_token_accuracy": 0.17561270594596862,
      "num_tokens": 9996566.0,
      "step": 4525
    },
    {
      "entropy": 5.319602298736572,
      "epoch": 4.566817952597075,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003471642231429475,
      "loss": 4.994,
      "mean_token_accuracy": 0.17609466016292571,
      "num_tokens": 10007017.0,
      "step": 4530
    },
    {
      "entropy": 5.387630796432495,
      "epoch": 4.5718608169440245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003467884010259443,
      "loss": 5.0069,
      "mean_token_accuracy": 0.1834713935852051,
      "num_tokens": 10018393.0,
      "step": 4535
    },
    {
      "entropy": 5.25979790687561,
      "epoch": 4.576903681290974,
      "grad_norm": 1.046875,
      "learning_rate": 0.00034641235578893476,
      "loss": 4.9089,
      "mean_token_accuracy": 0.17968080341815948,
      "num_tokens": 10028998.0,
      "step": 4540
    },
    {
      "entropy": 5.283661794662476,
      "epoch": 4.581946545637923,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003460360886006761,
      "loss": 4.9728,
      "mean_token_accuracy": 0.17355560064315795,
      "num_tokens": 10039243.0,
      "step": 4545
    },
    {
      "entropy": 5.297887754440308,
      "epoch": 4.586989409984872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034565960063061534,
      "loss": 4.9163,
      "mean_token_accuracy": 0.18408143520355225,
      "num_tokens": 10049658.0,
      "step": 4550
    },
    {
      "entropy": 5.317724418640137,
      "epoch": 4.592032274331821,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003452828930488859,
      "loss": 4.9979,
      "mean_token_accuracy": 0.17836031317710876,
      "num_tokens": 10060778.0,
      "step": 4555
    },
    {
      "entropy": 5.286425495147705,
      "epoch": 4.597075138678769,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00034490596702630355,
      "loss": 4.9043,
      "mean_token_accuracy": 0.18284219205379487,
      "num_tokens": 10071123.0,
      "step": 4560
    },
    {
      "entropy": 5.244607543945312,
      "epoch": 4.602118003025718,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003445288237343632,
      "loss": 4.9127,
      "mean_token_accuracy": 0.1817759394645691,
      "num_tokens": 10082114.0,
      "step": 4565
    },
    {
      "entropy": 5.298414134979248,
      "epoch": 4.607160867372667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00034415146434523463,
      "loss": 4.9657,
      "mean_token_accuracy": 0.17481246292591096,
      "num_tokens": 10092419.0,
      "step": 4570
    },
    {
      "entropy": 5.2881810665130615,
      "epoch": 4.6122037317196165,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034377389003175984,
      "loss": 4.9043,
      "mean_token_accuracy": 0.1826196789741516,
      "num_tokens": 10103975.0,
      "step": 4575
    },
    {
      "entropy": 5.240405130386352,
      "epoch": 4.617246596066566,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003433961019674483,
      "loss": 4.9139,
      "mean_token_accuracy": 0.18590575456619263,
      "num_tokens": 10114963.0,
      "step": 4580
    },
    {
      "entropy": 5.312945413589477,
      "epoch": 4.622289460413515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003430181013264741,
      "loss": 5.0079,
      "mean_token_accuracy": 0.17221707701683045,
      "num_tokens": 10126477.0,
      "step": 4585
    },
    {
      "entropy": 5.317572689056396,
      "epoch": 4.627332324760464,
      "grad_norm": 1.125,
      "learning_rate": 0.0003426398892836719,
      "loss": 4.9959,
      "mean_token_accuracy": 0.18182409554719925,
      "num_tokens": 10137176.0,
      "step": 4590
    },
    {
      "entropy": 5.270642185211182,
      "epoch": 4.632375189107413,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003422614670145336,
      "loss": 4.9716,
      "mean_token_accuracy": 0.1788034200668335,
      "num_tokens": 10147736.0,
      "step": 4595
    },
    {
      "entropy": 5.309645748138427,
      "epoch": 4.637418053454362,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003418828356952041,
      "loss": 4.9273,
      "mean_token_accuracy": 0.17924056202173233,
      "num_tokens": 10159438.0,
      "step": 4600
    },
    {
      "entropy": 5.253974962234497,
      "epoch": 4.642460917801311,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034150399650247846,
      "loss": 4.9404,
      "mean_token_accuracy": 0.18331195563077926,
      "num_tokens": 10170209.0,
      "step": 4605
    },
    {
      "entropy": 5.279243230819702,
      "epoch": 4.64750378214826,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00034112495061379764,
      "loss": 4.9934,
      "mean_token_accuracy": 0.17971529066562653,
      "num_tokens": 10182258.0,
      "step": 4610
    },
    {
      "entropy": 5.245618152618408,
      "epoch": 4.652546646495209,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034074569920724483,
      "loss": 4.8528,
      "mean_token_accuracy": 0.1836523175239563,
      "num_tokens": 10192952.0,
      "step": 4615
    },
    {
      "entropy": 5.302034521102906,
      "epoch": 4.6575895108421586,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003403662434615423,
      "loss": 4.9694,
      "mean_token_accuracy": 0.1753387525677681,
      "num_tokens": 10204762.0,
      "step": 4620
    },
    {
      "entropy": 5.252137136459351,
      "epoch": 4.662632375189108,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033998658455604736,
      "loss": 4.9111,
      "mean_token_accuracy": 0.18440319895744323,
      "num_tokens": 10214436.0,
      "step": 4625
    },
    {
      "entropy": 5.278934335708618,
      "epoch": 4.667675239536057,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033960672367074827,
      "loss": 5.0214,
      "mean_token_accuracy": 0.17598165273666383,
      "num_tokens": 10226593.0,
      "step": 4630
    },
    {
      "entropy": 5.304606962203979,
      "epoch": 4.672718103883006,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003392266619862619,
      "loss": 4.9678,
      "mean_token_accuracy": 0.18117802888154982,
      "num_tokens": 10237727.0,
      "step": 4635
    },
    {
      "entropy": 5.291701173782348,
      "epoch": 4.677760968229954,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003388464006838284,
      "loss": 5.0292,
      "mean_token_accuracy": 0.1763583943247795,
      "num_tokens": 10250106.0,
      "step": 4640
    },
    {
      "entropy": 5.293747663497925,
      "epoch": 4.682803832576903,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033846594094530874,
      "loss": 4.9325,
      "mean_token_accuracy": 0.1815586656332016,
      "num_tokens": 10260805.0,
      "step": 4645
    },
    {
      "entropy": 5.301459932327271,
      "epoch": 4.687846696923852,
      "grad_norm": 1.125,
      "learning_rate": 0.0003380852839531806,
      "loss": 5.0108,
      "mean_token_accuracy": 0.1728121042251587,
      "num_tokens": 10271773.0,
      "step": 4650
    },
    {
      "entropy": 5.323439598083496,
      "epoch": 4.6928895612708015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033770443089053476,
      "loss": 4.9589,
      "mean_token_accuracy": 0.1839527815580368,
      "num_tokens": 10281826.0,
      "step": 4655
    },
    {
      "entropy": 5.312728309631348,
      "epoch": 4.697932425617751,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033732338294107116,
      "loss": 5.0312,
      "mean_token_accuracy": 0.17409167140722276,
      "num_tokens": 10292528.0,
      "step": 4660
    },
    {
      "entropy": 5.276774835586548,
      "epoch": 4.7029752899647,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033694214128909567,
      "loss": 4.895,
      "mean_token_accuracy": 0.18432387709617615,
      "num_tokens": 10304487.0,
      "step": 4665
    },
    {
      "entropy": 5.289554691314697,
      "epoch": 4.708018154311649,
      "grad_norm": 1.0,
      "learning_rate": 0.00033656070711951633,
      "loss": 4.9822,
      "mean_token_accuracy": 0.17810264080762864,
      "num_tokens": 10314977.0,
      "step": 4670
    },
    {
      "entropy": 5.359469032287597,
      "epoch": 4.713061018658598,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003361790816178391,
      "loss": 5.0155,
      "mean_token_accuracy": 0.17588816583156586,
      "num_tokens": 10327725.0,
      "step": 4675
    },
    {
      "entropy": 5.265970516204834,
      "epoch": 4.718103883005547,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003357972659701649,
      "loss": 4.9412,
      "mean_token_accuracy": 0.17558968663215638,
      "num_tokens": 10339227.0,
      "step": 4680
    },
    {
      "entropy": 5.298288440704345,
      "epoch": 4.723146747352496,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033541526136318564,
      "loss": 5.0009,
      "mean_token_accuracy": 0.1717066705226898,
      "num_tokens": 10350353.0,
      "step": 4685
    },
    {
      "entropy": 5.298790788650512,
      "epoch": 4.728189611699445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003350330689841804,
      "loss": 5.022,
      "mean_token_accuracy": 0.1791086569428444,
      "num_tokens": 10360874.0,
      "step": 4690
    },
    {
      "entropy": 5.286825227737427,
      "epoch": 4.733232476046394,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003346506900210118,
      "loss": 4.9535,
      "mean_token_accuracy": 0.18101237267255782,
      "num_tokens": 10372398.0,
      "step": 4695
    },
    {
      "entropy": 5.254783964157104,
      "epoch": 4.7382753403933435,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003342681256621227,
      "loss": 4.9572,
      "mean_token_accuracy": 0.183388552069664,
      "num_tokens": 10382491.0,
      "step": 4700
    },
    {
      "entropy": 5.305398654937744,
      "epoch": 4.743318204740293,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003338853770965319,
      "loss": 4.9718,
      "mean_token_accuracy": 0.17851891070604325,
      "num_tokens": 10393378.0,
      "step": 4705
    },
    {
      "entropy": 5.26035213470459,
      "epoch": 4.748361069087242,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033350244551383064,
      "loss": 4.9374,
      "mean_token_accuracy": 0.18415171056985855,
      "num_tokens": 10403638.0,
      "step": 4710
    },
    {
      "entropy": 5.288418579101562,
      "epoch": 4.753403933434191,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003331193321041793,
      "loss": 4.9903,
      "mean_token_accuracy": 0.17170993834733964,
      "num_tokens": 10414532.0,
      "step": 4715
    },
    {
      "entropy": 5.2860170841217045,
      "epoch": 4.75844679778114,
      "grad_norm": 1.125,
      "learning_rate": 0.00033273603805830296,
      "loss": 4.9351,
      "mean_token_accuracy": 0.1850849434733391,
      "num_tokens": 10425813.0,
      "step": 4720
    },
    {
      "entropy": 5.319452095031738,
      "epoch": 4.763489662128089,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033235256456748847,
      "loss": 4.9892,
      "mean_token_accuracy": 0.17580467015504836,
      "num_tokens": 10436640.0,
      "step": 4725
    },
    {
      "entropy": 5.258197927474976,
      "epoch": 4.768532526475038,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003319689128235804,
      "loss": 4.952,
      "mean_token_accuracy": 0.1830135777592659,
      "num_tokens": 10447254.0,
      "step": 4730
    },
    {
      "entropy": 5.2829296588897705,
      "epoch": 4.773575390821987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033158508401897707,
      "loss": 4.9619,
      "mean_token_accuracy": 0.1781514674425125,
      "num_tokens": 10458500.0,
      "step": 4735
    },
    {
      "entropy": 5.274430990219116,
      "epoch": 4.778618255168936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003312010793466273,
      "loss": 4.8744,
      "mean_token_accuracy": 0.18644857704639434,
      "num_tokens": 10468683.0,
      "step": 4740
    },
    {
      "entropy": 5.380297756195068,
      "epoch": 4.783661119515885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003308169000000265,
      "loss": 5.0569,
      "mean_token_accuracy": 0.17666614204645156,
      "num_tokens": 10480613.0,
      "step": 4745
    },
    {
      "entropy": 5.269877767562866,
      "epoch": 4.788703983862834,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003304325471732129,
      "loss": 4.9006,
      "mean_token_accuracy": 0.18918281495571138,
      "num_tokens": 10491680.0,
      "step": 4750
    },
    {
      "entropy": 5.275027322769165,
      "epoch": 4.793746848209783,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033004802206076396,
      "loss": 5.0084,
      "mean_token_accuracy": 0.1768597334623337,
      "num_tokens": 10502867.0,
      "step": 4755
    },
    {
      "entropy": 5.362795639038086,
      "epoch": 4.798789712556732,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032966332585779263,
      "loss": 5.0602,
      "mean_token_accuracy": 0.1718745246529579,
      "num_tokens": 10514892.0,
      "step": 4760
    },
    {
      "entropy": 5.311322546005249,
      "epoch": 4.803832576903681,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00032927845975994356,
      "loss": 4.9826,
      "mean_token_accuracy": 0.1818615108728409,
      "num_tokens": 10525615.0,
      "step": 4765
    },
    {
      "entropy": 5.275985479354858,
      "epoch": 4.80887544125063,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032889342496338953,
      "loss": 4.9196,
      "mean_token_accuracy": 0.18454082906246186,
      "num_tokens": 10536226.0,
      "step": 4770
    },
    {
      "entropy": 5.351872873306275,
      "epoch": 4.813918305597579,
      "grad_norm": 1.125,
      "learning_rate": 0.0003285082226648273,
      "loss": 4.9776,
      "mean_token_accuracy": 0.17930400967597962,
      "num_tokens": 10547159.0,
      "step": 4775
    },
    {
      "entropy": 5.248814821243286,
      "epoch": 4.8189611699445285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003281228540614749,
      "loss": 4.9621,
      "mean_token_accuracy": 0.17887850105762482,
      "num_tokens": 10557368.0,
      "step": 4780
    },
    {
      "entropy": 5.258347034454346,
      "epoch": 4.824004034291478,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032773732035106657,
      "loss": 4.9,
      "mean_token_accuracy": 0.18604593276977538,
      "num_tokens": 10568439.0,
      "step": 4785
    },
    {
      "entropy": 5.300298500061035,
      "epoch": 4.829046898638427,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032735162273185006,
      "loss": 5.0469,
      "mean_token_accuracy": 0.17380245476961137,
      "num_tokens": 10580977.0,
      "step": 4790
    },
    {
      "entropy": 5.3309986114501955,
      "epoch": 4.834089762985376,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003269657624025824,
      "loss": 4.9842,
      "mean_token_accuracy": 0.17700576931238174,
      "num_tokens": 10590783.0,
      "step": 4795
    },
    {
      "entropy": 5.2612633228302,
      "epoch": 4.839132627332325,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003265797405625265,
      "loss": 4.9093,
      "mean_token_accuracy": 0.1802224785089493,
      "num_tokens": 10601658.0,
      "step": 4800
    },
    {
      "entropy": 5.261012125015259,
      "epoch": 4.844175491679274,
      "grad_norm": 1.0,
      "learning_rate": 0.0003261935584114471,
      "loss": 5.0525,
      "mean_token_accuracy": 0.17058018445968628,
      "num_tokens": 10613828.0,
      "step": 4805
    },
    {
      "entropy": 5.271330070495606,
      "epoch": 4.849218356026223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003258072171496072,
      "loss": 4.8707,
      "mean_token_accuracy": 0.17985524535179137,
      "num_tokens": 10624417.0,
      "step": 4810
    },
    {
      "entropy": 5.294085311889648,
      "epoch": 4.854261220373172,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003254207179777644,
      "loss": 4.9815,
      "mean_token_accuracy": 0.18036548048257828,
      "num_tokens": 10635731.0,
      "step": 4815
    },
    {
      "entropy": 5.278293180465698,
      "epoch": 4.8593040847201205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000325034062097167,
      "loss": 4.9605,
      "mean_token_accuracy": 0.17985475659370423,
      "num_tokens": 10646087.0,
      "step": 4820
    },
    {
      "entropy": 5.302151203155518,
      "epoch": 4.86434694906707,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032464725070955046,
      "loss": 5.0143,
      "mean_token_accuracy": 0.17714819610118865,
      "num_tokens": 10657238.0,
      "step": 4825
    },
    {
      "entropy": 5.284213972091675,
      "epoch": 4.869389813414019,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032426028501713347,
      "loss": 4.986,
      "mean_token_accuracy": 0.18563233315944672,
      "num_tokens": 10667892.0,
      "step": 4830
    },
    {
      "entropy": 5.270573997497559,
      "epoch": 4.874432677760968,
      "grad_norm": 0.984375,
      "learning_rate": 0.00032387316622261425,
      "loss": 5.0115,
      "mean_token_accuracy": 0.18733932971954345,
      "num_tokens": 10679797.0,
      "step": 4835
    },
    {
      "entropy": 5.360124397277832,
      "epoch": 4.879475542107917,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032348589552916703,
      "loss": 5.0008,
      "mean_token_accuracy": 0.18354785889387132,
      "num_tokens": 10690622.0,
      "step": 4840
    },
    {
      "entropy": 5.354613971710205,
      "epoch": 4.884518406454866,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003230984741404381,
      "loss": 5.0295,
      "mean_token_accuracy": 0.1750412955880165,
      "num_tokens": 10702496.0,
      "step": 4845
    },
    {
      "entropy": 5.298963928222657,
      "epoch": 4.889561270801815,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003227109032605419,
      "loss": 5.0175,
      "mean_token_accuracy": 0.18698592931032182,
      "num_tokens": 10714438.0,
      "step": 4850
    },
    {
      "entropy": 5.406459331512451,
      "epoch": 4.894604135148764,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003223231840940578,
      "loss": 5.1017,
      "mean_token_accuracy": 0.1781959980726242,
      "num_tokens": 10725242.0,
      "step": 4855
    },
    {
      "entropy": 5.270995187759399,
      "epoch": 4.8996469994957135,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032193531784602595,
      "loss": 4.943,
      "mean_token_accuracy": 0.1840689316391945,
      "num_tokens": 10737546.0,
      "step": 4860
    },
    {
      "entropy": 5.280051898956299,
      "epoch": 4.904689863842663,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032154730572194346,
      "loss": 5.0365,
      "mean_token_accuracy": 0.17502037882804872,
      "num_tokens": 10748776.0,
      "step": 4865
    },
    {
      "entropy": 5.258845472335816,
      "epoch": 4.909732728189612,
      "grad_norm": 1.046875,
      "learning_rate": 0.000321159148927761,
      "loss": 4.9415,
      "mean_token_accuracy": 0.18133331835269928,
      "num_tokens": 10759593.0,
      "step": 4870
    },
    {
      "entropy": 5.289977407455444,
      "epoch": 4.914775592536561,
      "grad_norm": 1.203125,
      "learning_rate": 0.000320770848669879,
      "loss": 4.985,
      "mean_token_accuracy": 0.1744542673230171,
      "num_tokens": 10770303.0,
      "step": 4875
    },
    {
      "entropy": 5.327417039871216,
      "epoch": 4.91981845688351,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032038240615514345,
      "loss": 5.0046,
      "mean_token_accuracy": 0.17604739665985109,
      "num_tokens": 10781263.0,
      "step": 4880
    },
    {
      "entropy": 5.331995916366577,
      "epoch": 4.924861321230459,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003199938225908426,
      "loss": 4.9556,
      "mean_token_accuracy": 0.1845855623483658,
      "num_tokens": 10792759.0,
      "step": 4885
    },
    {
      "entropy": 5.240837478637696,
      "epoch": 4.929904185577408,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003196050991847033,
      "loss": 4.9548,
      "mean_token_accuracy": 0.18243360072374343,
      "num_tokens": 10804313.0,
      "step": 4890
    },
    {
      "entropy": 5.2612464427948,
      "epoch": 4.934947049924357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003192162371448868,
      "loss": 4.893,
      "mean_token_accuracy": 0.18721656501293182,
      "num_tokens": 10815866.0,
      "step": 4895
    },
    {
      "entropy": 5.359346342086792,
      "epoch": 4.939989914271306,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031882723767998517,
      "loss": 5.0433,
      "mean_token_accuracy": 0.17281465530395507,
      "num_tokens": 10827197.0,
      "step": 4900
    },
    {
      "entropy": 5.297011995315552,
      "epoch": 4.9450327786182555,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031843810199901795,
      "loss": 4.9778,
      "mean_token_accuracy": 0.18017227500677108,
      "num_tokens": 10838274.0,
      "step": 4905
    },
    {
      "entropy": 5.270789432525635,
      "epoch": 4.950075642965205,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003180488313114277,
      "loss": 4.9245,
      "mean_token_accuracy": 0.17935557961463927,
      "num_tokens": 10848523.0,
      "step": 4910
    },
    {
      "entropy": 5.2362542152404785,
      "epoch": 4.955118507312154,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003176594268270766,
      "loss": 4.9907,
      "mean_token_accuracy": 0.1809832438826561,
      "num_tokens": 10859867.0,
      "step": 4915
    },
    {
      "entropy": 5.271730947494507,
      "epoch": 4.960161371659103,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031726988975624293,
      "loss": 4.956,
      "mean_token_accuracy": 0.18322471231222154,
      "num_tokens": 10870032.0,
      "step": 4920
    },
    {
      "entropy": 5.245261096954346,
      "epoch": 4.965204236006051,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00031688022130961675,
      "loss": 4.8665,
      "mean_token_accuracy": 0.18327659815549852,
      "num_tokens": 10881629.0,
      "step": 4925
    },
    {
      "entropy": 5.251491498947144,
      "epoch": 4.970247100353,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003164904226982967,
      "loss": 4.9333,
      "mean_token_accuracy": 0.18830396682024003,
      "num_tokens": 10892513.0,
      "step": 4930
    },
    {
      "entropy": 5.2613194465637205,
      "epoch": 4.975289964699949,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031610049513378574,
      "loss": 4.893,
      "mean_token_accuracy": 0.1907330945134163,
      "num_tokens": 10903680.0,
      "step": 4935
    },
    {
      "entropy": 5.286007642745972,
      "epoch": 4.9803328290468984,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031571043982798786,
      "loss": 4.8948,
      "mean_token_accuracy": 0.181423918902874,
      "num_tokens": 10914117.0,
      "step": 4940
    },
    {
      "entropy": 5.261344003677368,
      "epoch": 4.985375693393848,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003153202579932039,
      "loss": 5.0307,
      "mean_token_accuracy": 0.1765132173895836,
      "num_tokens": 10925063.0,
      "step": 4945
    },
    {
      "entropy": 5.264973735809326,
      "epoch": 4.990418557740797,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00031492995084212794,
      "loss": 4.9233,
      "mean_token_accuracy": 0.1810626894235611,
      "num_tokens": 10937548.0,
      "step": 4950
    },
    {
      "entropy": 5.254584169387817,
      "epoch": 4.995461422087746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003145395195878437,
      "loss": 4.9475,
      "mean_token_accuracy": 0.18406419157981874,
      "num_tokens": 10948927.0,
      "step": 4955
    },
    {
      "entropy": 5.2572401364644366,
      "epoch": 5.0,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00031414896544382047,
      "loss": 4.8752,
      "mean_token_accuracy": 0.17638060947259268,
      "num_tokens": 10958200.0,
      "step": 4960
    },
    {
      "entropy": 5.347752618789673,
      "epoch": 5.005042864346949,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003137582896239095,
      "loss": 4.728,
      "mean_token_accuracy": 0.19896187633275986,
      "num_tokens": 10968904.0,
      "step": 4965
    },
    {
      "entropy": 5.237904071807861,
      "epoch": 5.010085728693898,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003133674933423404,
      "loss": 4.6917,
      "mean_token_accuracy": 0.19597846120595933,
      "num_tokens": 10979993.0,
      "step": 4970
    },
    {
      "entropy": 5.243738079071045,
      "epoch": 5.015128593040847,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031297657781371687,
      "loss": 4.6584,
      "mean_token_accuracy": 0.20072104930877685,
      "num_tokens": 10990203.0,
      "step": 4975
    },
    {
      "entropy": 5.3005434513092045,
      "epoch": 5.0201714573877965,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003125855442530135,
      "loss": 4.6853,
      "mean_token_accuracy": 0.20662225186824798,
      "num_tokens": 11001002.0,
      "step": 4980
    },
    {
      "entropy": 5.2116399765014645,
      "epoch": 5.025214321734746,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031219439387557155,
      "loss": 4.7641,
      "mean_token_accuracy": 0.1929519608616829,
      "num_tokens": 11012087.0,
      "step": 4985
    },
    {
      "entropy": 5.203783750534058,
      "epoch": 5.030257186081695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031180312789709535,
      "loss": 4.6454,
      "mean_token_accuracy": 0.20014435052871704,
      "num_tokens": 11022596.0,
      "step": 4990
    },
    {
      "entropy": 5.265881061553955,
      "epoch": 5.035300050428644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003114117475336487,
      "loss": 4.7467,
      "mean_token_accuracy": 0.1891510233283043,
      "num_tokens": 11034186.0,
      "step": 4995
    },
    {
      "entropy": 5.158915328979492,
      "epoch": 5.040342914775593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003110202540016509,
      "loss": 4.662,
      "mean_token_accuracy": 0.20034248530864715,
      "num_tokens": 11045536.0,
      "step": 5000
    },
    {
      "epoch": 5.040342914775593,
      "eval_entropy": 5.141179946239364,
      "eval_loss": 5.72727632522583,
      "eval_mean_token_accuracy": 0.15003919645044336,
      "eval_num_tokens": 11045536.0,
      "eval_runtime": 2.6321,
      "eval_samples_per_second": 1192.219,
      "eval_steps_per_second": 149.312,
      "step": 5000
    },
    {
      "entropy": 5.281191205978393,
      "epoch": 5.045385779122542,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00031062864851787267,
      "loss": 4.71,
      "mean_token_accuracy": 0.20173763036727904,
      "num_tokens": 11057763.0,
      "step": 5005
    },
    {
      "entropy": 5.286484956741333,
      "epoch": 5.05042864346949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031023693229943294,
      "loss": 4.7372,
      "mean_token_accuracy": 0.19276192933320999,
      "num_tokens": 11068121.0,
      "step": 5010
    },
    {
      "entropy": 5.221051025390625,
      "epoch": 5.055471507816439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003098451065637948,
      "loss": 4.7434,
      "mean_token_accuracy": 0.19354756325483322,
      "num_tokens": 11078970.0,
      "step": 5015
    },
    {
      "entropy": 5.303717517852784,
      "epoch": 5.0605143721633885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030945317252876163,
      "loss": 4.7534,
      "mean_token_accuracy": 0.18817539811134337,
      "num_tokens": 11090352.0,
      "step": 5020
    },
    {
      "entropy": 5.293031406402588,
      "epoch": 5.065557236510338,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030906113141247353,
      "loss": 4.763,
      "mean_token_accuracy": 0.197238053381443,
      "num_tokens": 11101299.0,
      "step": 5025
    },
    {
      "entropy": 5.239246082305908,
      "epoch": 5.070600100857287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030866898443340324,
      "loss": 4.6991,
      "mean_token_accuracy": 0.197691909968853,
      "num_tokens": 11110833.0,
      "step": 5030
    },
    {
      "entropy": 5.254098510742187,
      "epoch": 5.075642965204236,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003082767328103527,
      "loss": 4.7846,
      "mean_token_accuracy": 0.1888483464717865,
      "num_tokens": 11120814.0,
      "step": 5035
    },
    {
      "entropy": 5.32061243057251,
      "epoch": 5.080685829551185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030788437776244895,
      "loss": 4.7525,
      "mean_token_accuracy": 0.19328293800354004,
      "num_tokens": 11131604.0,
      "step": 5040
    },
    {
      "entropy": 5.286145162582398,
      "epoch": 5.085728693898134,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003074919205091405,
      "loss": 4.7678,
      "mean_token_accuracy": 0.1922394350171089,
      "num_tokens": 11142126.0,
      "step": 5045
    },
    {
      "entropy": 5.249022150039673,
      "epoch": 5.090771558245083,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003070993622701936,
      "loss": 4.7069,
      "mean_token_accuracy": 0.2020687699317932,
      "num_tokens": 11153260.0,
      "step": 5050
    },
    {
      "entropy": 5.2604598045349125,
      "epoch": 5.095814422592032,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030670670426568837,
      "loss": 4.6859,
      "mean_token_accuracy": 0.1989409953355789,
      "num_tokens": 11163303.0,
      "step": 5055
    },
    {
      "entropy": 5.1776457786560055,
      "epoch": 5.100857286938981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003063139477160147,
      "loss": 4.7265,
      "mean_token_accuracy": 0.18996165543794633,
      "num_tokens": 11173949.0,
      "step": 5060
    },
    {
      "entropy": 5.272495460510254,
      "epoch": 5.1059001512859306,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030592109384186935,
      "loss": 4.7377,
      "mean_token_accuracy": 0.19524364471435546,
      "num_tokens": 11184400.0,
      "step": 5065
    },
    {
      "entropy": 5.315284729003906,
      "epoch": 5.11094301563288,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000305528143864251,
      "loss": 4.8077,
      "mean_token_accuracy": 0.1947113960981369,
      "num_tokens": 11196037.0,
      "step": 5070
    },
    {
      "entropy": 5.292333316802979,
      "epoch": 5.115985879979829,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030513509900445725,
      "loss": 4.8267,
      "mean_token_accuracy": 0.19195764511823654,
      "num_tokens": 11207238.0,
      "step": 5075
    },
    {
      "entropy": 5.222452116012573,
      "epoch": 5.121028744326778,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030474196048408065,
      "loss": 4.6926,
      "mean_token_accuracy": 0.19556245356798171,
      "num_tokens": 11218394.0,
      "step": 5080
    },
    {
      "entropy": 5.230966091156006,
      "epoch": 5.126071608673727,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003043487295250049,
      "loss": 4.6883,
      "mean_token_accuracy": 0.2016897439956665,
      "num_tokens": 11228665.0,
      "step": 5085
    },
    {
      "entropy": 5.158874273300171,
      "epoch": 5.131114473020676,
      "grad_norm": 1.125,
      "learning_rate": 0.0003039554073494006,
      "loss": 4.6506,
      "mean_token_accuracy": 0.20119154900312425,
      "num_tokens": 11239014.0,
      "step": 5090
    },
    {
      "entropy": 5.236811542510987,
      "epoch": 5.136157337367624,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003035619951797224,
      "loss": 4.7431,
      "mean_token_accuracy": 0.19681891947984695,
      "num_tokens": 11250423.0,
      "step": 5095
    },
    {
      "entropy": 5.24750862121582,
      "epoch": 5.1412002017145735,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003031684942387043,
      "loss": 4.6599,
      "mean_token_accuracy": 0.2024094507098198,
      "num_tokens": 11260984.0,
      "step": 5100
    },
    {
      "entropy": 5.23399863243103,
      "epoch": 5.146243066061523,
      "grad_norm": 1.125,
      "learning_rate": 0.00030277490574935633,
      "loss": 4.7927,
      "mean_token_accuracy": 0.19361722320318223,
      "num_tokens": 11272103.0,
      "step": 5105
    },
    {
      "entropy": 5.222354173660278,
      "epoch": 5.151285930408472,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003023812309349605,
      "loss": 4.768,
      "mean_token_accuracy": 0.19484228938817977,
      "num_tokens": 11283948.0,
      "step": 5110
    },
    {
      "entropy": 5.248267316818238,
      "epoch": 5.156328794755421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003019874710190674,
      "loss": 4.705,
      "mean_token_accuracy": 0.2065915808081627,
      "num_tokens": 11295518.0,
      "step": 5115
    },
    {
      "entropy": 5.14952130317688,
      "epoch": 5.16137165910237,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003015936272254918,
      "loss": 4.6481,
      "mean_token_accuracy": 0.20521476566791536,
      "num_tokens": 11306244.0,
      "step": 5120
    },
    {
      "entropy": 5.179940462112427,
      "epoch": 5.166414523449319,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003011997007783094,
      "loss": 4.7035,
      "mean_token_accuracy": 0.1930046409368515,
      "num_tokens": 11317857.0,
      "step": 5125
    },
    {
      "entropy": 5.233994245529175,
      "epoch": 5.171457387796268,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003008056929018525,
      "loss": 4.7528,
      "mean_token_accuracy": 0.19374128580093383,
      "num_tokens": 11328620.0,
      "step": 5130
    },
    {
      "entropy": 5.1903636932373045,
      "epoch": 5.176500252143217,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030041160482070686,
      "loss": 4.7282,
      "mean_token_accuracy": 0.2032180979847908,
      "num_tokens": 11340428.0,
      "step": 5135
    },
    {
      "entropy": 5.228970384597778,
      "epoch": 5.181543116490166,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030001743775970714,
      "loss": 4.7249,
      "mean_token_accuracy": 0.19578366577625275,
      "num_tokens": 11350957.0,
      "step": 5140
    },
    {
      "entropy": 5.26102089881897,
      "epoch": 5.1865859808371155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029962319294393383,
      "loss": 4.7078,
      "mean_token_accuracy": 0.19729420393705369,
      "num_tokens": 11360780.0,
      "step": 5145
    },
    {
      "entropy": 5.181542253494262,
      "epoch": 5.191628845184065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002992288715987088,
      "loss": 4.6696,
      "mean_token_accuracy": 0.20200882256031036,
      "num_tokens": 11370696.0,
      "step": 5150
    },
    {
      "entropy": 5.166230916976929,
      "epoch": 5.196671709531014,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002988344749495919,
      "loss": 4.7126,
      "mean_token_accuracy": 0.20468319058418274,
      "num_tokens": 11381070.0,
      "step": 5155
    },
    {
      "entropy": 5.154423761367798,
      "epoch": 5.201714573877963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029844000422237714,
      "loss": 4.7001,
      "mean_token_accuracy": 0.1984965667128563,
      "num_tokens": 11391386.0,
      "step": 5160
    },
    {
      "entropy": 5.2157337188720705,
      "epoch": 5.206757438224912,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029804546064308854,
      "loss": 4.716,
      "mean_token_accuracy": 0.19292420893907547,
      "num_tokens": 11401428.0,
      "step": 5165
    },
    {
      "entropy": 5.221881437301636,
      "epoch": 5.211800302571861,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029765084543797657,
      "loss": 4.6999,
      "mean_token_accuracy": 0.1968621164560318,
      "num_tokens": 11412503.0,
      "step": 5170
    },
    {
      "entropy": 5.235488510131836,
      "epoch": 5.21684316691881,
      "grad_norm": 1.25,
      "learning_rate": 0.0002972561598335145,
      "loss": 4.7216,
      "mean_token_accuracy": 0.19793958961963654,
      "num_tokens": 11423340.0,
      "step": 5175
    },
    {
      "entropy": 5.222541809082031,
      "epoch": 5.221886031265759,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002968614050563943,
      "loss": 4.7352,
      "mean_token_accuracy": 0.19378636330366134,
      "num_tokens": 11434028.0,
      "step": 5180
    },
    {
      "entropy": 5.252810049057007,
      "epoch": 5.2269288956127085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002964665823335229,
      "loss": 4.8152,
      "mean_token_accuracy": 0.19323618710041046,
      "num_tokens": 11445457.0,
      "step": 5185
    },
    {
      "entropy": 5.226347970962524,
      "epoch": 5.231971759959657,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029607169289201836,
      "loss": 4.7535,
      "mean_token_accuracy": 0.19465828388929368,
      "num_tokens": 11455967.0,
      "step": 5190
    },
    {
      "entropy": 5.199074935913086,
      "epoch": 5.237014624306606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029567673795920625,
      "loss": 4.7439,
      "mean_token_accuracy": 0.1985458254814148,
      "num_tokens": 11468215.0,
      "step": 5195
    },
    {
      "entropy": 5.154666996002197,
      "epoch": 5.242057488653555,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029528171876261555,
      "loss": 4.6605,
      "mean_token_accuracy": 0.2029028683900833,
      "num_tokens": 11480329.0,
      "step": 5200
    },
    {
      "entropy": 5.274513483047485,
      "epoch": 5.247100353000504,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002948866365299751,
      "loss": 4.7555,
      "mean_token_accuracy": 0.19548195153474807,
      "num_tokens": 11490612.0,
      "step": 5205
    },
    {
      "entropy": 5.281692695617676,
      "epoch": 5.252143217347453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002944914924892096,
      "loss": 4.7496,
      "mean_token_accuracy": 0.19199593365192413,
      "num_tokens": 11501236.0,
      "step": 5210
    },
    {
      "entropy": 5.175805282592774,
      "epoch": 5.257186081694402,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002940962878684358,
      "loss": 4.7095,
      "mean_token_accuracy": 0.195647232234478,
      "num_tokens": 11513204.0,
      "step": 5215
    },
    {
      "entropy": 5.182984256744385,
      "epoch": 5.262228946041351,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002937010238959588,
      "loss": 4.7511,
      "mean_token_accuracy": 0.1943848878145218,
      "num_tokens": 11524954.0,
      "step": 5220
    },
    {
      "entropy": 5.24591817855835,
      "epoch": 5.2672718103883005,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002933057018002681,
      "loss": 4.7393,
      "mean_token_accuracy": 0.19283102005720137,
      "num_tokens": 11537756.0,
      "step": 5225
    },
    {
      "entropy": 5.271328020095825,
      "epoch": 5.27231467473525,
      "grad_norm": 1.203125,
      "learning_rate": 0.000292910322810034,
      "loss": 4.7455,
      "mean_token_accuracy": 0.1985904261469841,
      "num_tokens": 11549920.0,
      "step": 5230
    },
    {
      "entropy": 5.199133586883545,
      "epoch": 5.277357539082199,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029251488815410344,
      "loss": 4.7492,
      "mean_token_accuracy": 0.1957603007555008,
      "num_tokens": 11560642.0,
      "step": 5235
    },
    {
      "entropy": 5.157462549209595,
      "epoch": 5.282400403429148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029211939906149644,
      "loss": 4.7282,
      "mean_token_accuracy": 0.195216104388237,
      "num_tokens": 11571773.0,
      "step": 5240
    },
    {
      "entropy": 5.254365873336792,
      "epoch": 5.287443267776097,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029172385676140225,
      "loss": 4.7874,
      "mean_token_accuracy": 0.19785969704389572,
      "num_tokens": 11583550.0,
      "step": 5245
    },
    {
      "entropy": 5.2908775806427,
      "epoch": 5.292486132123046,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029132826248317537,
      "loss": 4.769,
      "mean_token_accuracy": 0.18914690464735032,
      "num_tokens": 11593844.0,
      "step": 5250
    },
    {
      "entropy": 5.224659013748169,
      "epoch": 5.297528996469995,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002909326174563322,
      "loss": 4.7296,
      "mean_token_accuracy": 0.19573808312416077,
      "num_tokens": 11604325.0,
      "step": 5255
    },
    {
      "entropy": 5.172271251678467,
      "epoch": 5.302571860816944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029053692291054636,
      "loss": 4.6858,
      "mean_token_accuracy": 0.19416798502206803,
      "num_tokens": 11615172.0,
      "step": 5260
    },
    {
      "entropy": 5.235822582244873,
      "epoch": 5.307614725163893,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029014118007564564,
      "loss": 4.7648,
      "mean_token_accuracy": 0.19445224106311798,
      "num_tokens": 11626064.0,
      "step": 5265
    },
    {
      "entropy": 5.265423154830932,
      "epoch": 5.3126575895108425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002897453901816081,
      "loss": 4.7798,
      "mean_token_accuracy": 0.19302175790071488,
      "num_tokens": 11636852.0,
      "step": 5270
    },
    {
      "entropy": 5.2245806694030765,
      "epoch": 5.317700453857791,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028934955445855773,
      "loss": 4.7871,
      "mean_token_accuracy": 0.19751407504081725,
      "num_tokens": 11648720.0,
      "step": 5275
    },
    {
      "entropy": 5.230935764312744,
      "epoch": 5.32274331820474,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028895367413676095,
      "loss": 4.7582,
      "mean_token_accuracy": 0.19794594794511794,
      "num_tokens": 11661271.0,
      "step": 5280
    },
    {
      "entropy": 5.251638698577881,
      "epoch": 5.327786182551689,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002885577504466232,
      "loss": 4.7391,
      "mean_token_accuracy": 0.1919041559100151,
      "num_tokens": 11672232.0,
      "step": 5285
    },
    {
      "entropy": 5.205961894989014,
      "epoch": 5.332829046898638,
      "grad_norm": 1.125,
      "learning_rate": 0.00028816178461868425,
      "loss": 4.7062,
      "mean_token_accuracy": 0.19907160103321075,
      "num_tokens": 11683258.0,
      "step": 5290
    },
    {
      "entropy": 5.2376491069793705,
      "epoch": 5.337871911245587,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002877657778836151,
      "loss": 4.6898,
      "mean_token_accuracy": 0.19311168938875198,
      "num_tokens": 11693407.0,
      "step": 5295
    },
    {
      "entropy": 5.185609817504883,
      "epoch": 5.342914775592536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002873697314722138,
      "loss": 4.779,
      "mean_token_accuracy": 0.19829982668161392,
      "num_tokens": 11704227.0,
      "step": 5300
    },
    {
      "entropy": 5.258454895019531,
      "epoch": 5.3479576399394855,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002869736466154018,
      "loss": 4.7693,
      "mean_token_accuracy": 0.1916856974363327,
      "num_tokens": 11715738.0,
      "step": 5305
    },
    {
      "entropy": 5.273810195922851,
      "epoch": 5.353000504286435,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028657752454421974,
      "loss": 4.7573,
      "mean_token_accuracy": 0.19490979015827178,
      "num_tokens": 11726158.0,
      "step": 5310
    },
    {
      "entropy": 5.15630087852478,
      "epoch": 5.358043368633384,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002861813664898245,
      "loss": 4.7106,
      "mean_token_accuracy": 0.20360922068357468,
      "num_tokens": 11736998.0,
      "step": 5315
    },
    {
      "entropy": 5.3174700260162355,
      "epoch": 5.363086232980333,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028578517368348425,
      "loss": 4.8553,
      "mean_token_accuracy": 0.188724522292614,
      "num_tokens": 11748818.0,
      "step": 5320
    },
    {
      "entropy": 5.201540946960449,
      "epoch": 5.368129097327282,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002853889473565753,
      "loss": 4.6881,
      "mean_token_accuracy": 0.19262525141239167,
      "num_tokens": 11760116.0,
      "step": 5325
    },
    {
      "entropy": 5.2184741497039795,
      "epoch": 5.373171961674231,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002849926887405784,
      "loss": 4.7801,
      "mean_token_accuracy": 0.19618999660015107,
      "num_tokens": 11771113.0,
      "step": 5330
    },
    {
      "entropy": 5.230961608886719,
      "epoch": 5.37821482602118,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002845963990670744,
      "loss": 4.6754,
      "mean_token_accuracy": 0.19523147493600845,
      "num_tokens": 11781628.0,
      "step": 5335
    },
    {
      "entropy": 5.2365968227386475,
      "epoch": 5.383257690368129,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028420007956774077,
      "loss": 4.7869,
      "mean_token_accuracy": 0.1947922334074974,
      "num_tokens": 11792363.0,
      "step": 5340
    },
    {
      "entropy": 5.190263271331787,
      "epoch": 5.388300554715078,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028380373147434746,
      "loss": 4.7425,
      "mean_token_accuracy": 0.19842376559972763,
      "num_tokens": 11802794.0,
      "step": 5345
    },
    {
      "entropy": 5.198542594909668,
      "epoch": 5.3933434190620275,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028340735601875376,
      "loss": 4.7351,
      "mean_token_accuracy": 0.19469986259937286,
      "num_tokens": 11813575.0,
      "step": 5350
    },
    {
      "entropy": 5.278315496444702,
      "epoch": 5.398386283408977,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002830109544329036,
      "loss": 4.8369,
      "mean_token_accuracy": 0.18049127012491226,
      "num_tokens": 11824408.0,
      "step": 5355
    },
    {
      "entropy": 5.202161359786987,
      "epoch": 5.403429147755926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028261452794882223,
      "loss": 4.6792,
      "mean_token_accuracy": 0.20581565499305726,
      "num_tokens": 11835980.0,
      "step": 5360
    },
    {
      "entropy": 5.19088978767395,
      "epoch": 5.408472012102875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028221807779861243,
      "loss": 4.7236,
      "mean_token_accuracy": 0.1982305258512497,
      "num_tokens": 11846422.0,
      "step": 5365
    },
    {
      "entropy": 5.3386811256408695,
      "epoch": 5.413514876449823,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002818216052144502,
      "loss": 4.8639,
      "mean_token_accuracy": 0.19375443160533906,
      "num_tokens": 11858512.0,
      "step": 5370
    },
    {
      "entropy": 5.223774719238281,
      "epoch": 5.418557740796772,
      "grad_norm": 1.25,
      "learning_rate": 0.0002814251114285816,
      "loss": 4.703,
      "mean_token_accuracy": 0.20014239251613616,
      "num_tokens": 11869319.0,
      "step": 5375
    },
    {
      "entropy": 5.206186103820801,
      "epoch": 5.423600605143721,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002810285976733186,
      "loss": 4.781,
      "mean_token_accuracy": 0.19317810386419296,
      "num_tokens": 11881161.0,
      "step": 5380
    },
    {
      "entropy": 5.2189970970153805,
      "epoch": 5.4286434694906704,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002806320651810349,
      "loss": 4.7625,
      "mean_token_accuracy": 0.18827890753746032,
      "num_tokens": 11890762.0,
      "step": 5385
    },
    {
      "entropy": 5.221167182922363,
      "epoch": 5.43368633383762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002802355151841627,
      "loss": 4.7651,
      "mean_token_accuracy": 0.1935282826423645,
      "num_tokens": 11900851.0,
      "step": 5390
    },
    {
      "entropy": 5.198607492446899,
      "epoch": 5.438729198184569,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002798389489151886,
      "loss": 4.7184,
      "mean_token_accuracy": 0.1925155833363533,
      "num_tokens": 11910899.0,
      "step": 5395
    },
    {
      "entropy": 5.330115699768067,
      "epoch": 5.443772062531518,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027944236760664947,
      "loss": 4.8746,
      "mean_token_accuracy": 0.18778094947338103,
      "num_tokens": 11922823.0,
      "step": 5400
    },
    {
      "entropy": 5.254288005828857,
      "epoch": 5.448814926878467,
      "grad_norm": 1.125,
      "learning_rate": 0.00027904577249112945,
      "loss": 4.7875,
      "mean_token_accuracy": 0.19317295402288437,
      "num_tokens": 11933366.0,
      "step": 5405
    },
    {
      "entropy": 5.157413530349731,
      "epoch": 5.453857791225416,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002786491648012552,
      "loss": 4.7119,
      "mean_token_accuracy": 0.19343763142824172,
      "num_tokens": 11945484.0,
      "step": 5410
    },
    {
      "entropy": 5.263805627822876,
      "epoch": 5.458900655572365,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002782525457696926,
      "loss": 4.7998,
      "mean_token_accuracy": 0.19454885572195052,
      "num_tokens": 11955855.0,
      "step": 5415
    },
    {
      "entropy": 5.217371320724487,
      "epoch": 5.463943519919314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002778559166291427,
      "loss": 4.6594,
      "mean_token_accuracy": 0.20352775305509568,
      "num_tokens": 11966597.0,
      "step": 5420
    },
    {
      "entropy": 5.227929544448853,
      "epoch": 5.468986384266263,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002774592786123379,
      "loss": 4.8157,
      "mean_token_accuracy": 0.1897737830877304,
      "num_tokens": 11978388.0,
      "step": 5425
    },
    {
      "entropy": 5.245231103897095,
      "epoch": 5.4740292486132125,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002770626329520386,
      "loss": 4.7235,
      "mean_token_accuracy": 0.19348430186510085,
      "num_tokens": 11988685.0,
      "step": 5430
    },
    {
      "entropy": 5.258587741851807,
      "epoch": 5.479072112960162,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002766659808810285,
      "loss": 4.8233,
      "mean_token_accuracy": 0.18947571218013765,
      "num_tokens": 12001412.0,
      "step": 5435
    },
    {
      "entropy": 5.164519309997559,
      "epoch": 5.484114977307111,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027626932363211134,
      "loss": 4.6739,
      "mean_token_accuracy": 0.20838394165039062,
      "num_tokens": 12013100.0,
      "step": 5440
    },
    {
      "entropy": 5.258546590805054,
      "epoch": 5.48915784165406,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002758726624381071,
      "loss": 4.7805,
      "mean_token_accuracy": 0.19472405910491944,
      "num_tokens": 12024954.0,
      "step": 5445
    },
    {
      "entropy": 5.240252923965454,
      "epoch": 5.494200706001009,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002754759985318478,
      "loss": 4.7126,
      "mean_token_accuracy": 0.20070272833108901,
      "num_tokens": 12036201.0,
      "step": 5450
    },
    {
      "entropy": 5.159111738204956,
      "epoch": 5.499243570347957,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002750793331461741,
      "loss": 4.7388,
      "mean_token_accuracy": 0.19494573324918746,
      "num_tokens": 12047039.0,
      "step": 5455
    },
    {
      "entropy": 5.266638898849488,
      "epoch": 5.504286434694906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002746826675139313,
      "loss": 4.8141,
      "mean_token_accuracy": 0.19487137198448182,
      "num_tokens": 12057578.0,
      "step": 5460
    },
    {
      "entropy": 5.28081865310669,
      "epoch": 5.509329299041855,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027428600286796503,
      "loss": 4.8078,
      "mean_token_accuracy": 0.1915408715605736,
      "num_tokens": 12067947.0,
      "step": 5465
    },
    {
      "entropy": 5.2249456405639645,
      "epoch": 5.5143721633888045,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027388934044111836,
      "loss": 4.734,
      "mean_token_accuracy": 0.19665402621030809,
      "num_tokens": 12078932.0,
      "step": 5470
    },
    {
      "entropy": 5.251396512985229,
      "epoch": 5.519415027735754,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002734926814662271,
      "loss": 4.796,
      "mean_token_accuracy": 0.1950599282979965,
      "num_tokens": 12089743.0,
      "step": 5475
    },
    {
      "entropy": 5.282199096679688,
      "epoch": 5.524457892082703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027309602717611673,
      "loss": 4.7899,
      "mean_token_accuracy": 0.19252176880836486,
      "num_tokens": 12102038.0,
      "step": 5480
    },
    {
      "entropy": 5.2331770896911625,
      "epoch": 5.529500756429652,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027269937880359786,
      "loss": 4.7604,
      "mean_token_accuracy": 0.19788573533296586,
      "num_tokens": 12112324.0,
      "step": 5485
    },
    {
      "entropy": 5.224417448043823,
      "epoch": 5.534543620776601,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027230273758146263,
      "loss": 4.7489,
      "mean_token_accuracy": 0.19597203582525252,
      "num_tokens": 12124533.0,
      "step": 5490
    },
    {
      "entropy": 5.254444026947022,
      "epoch": 5.53958648512355,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002719061047424813,
      "loss": 4.7872,
      "mean_token_accuracy": 0.19194820672273635,
      "num_tokens": 12136054.0,
      "step": 5495
    },
    {
      "entropy": 5.214131259918213,
      "epoch": 5.544629349470499,
      "grad_norm": 1.125,
      "learning_rate": 0.0002715094815193978,
      "loss": 4.7785,
      "mean_token_accuracy": 0.1901981309056282,
      "num_tokens": 12147495.0,
      "step": 5500
    },
    {
      "epoch": 5.544629349470499,
      "eval_entropy": 5.107799089591921,
      "eval_loss": 5.713109016418457,
      "eval_mean_token_accuracy": 0.15286463650734977,
      "eval_num_tokens": 12147495.0,
      "eval_runtime": 2.6423,
      "eval_samples_per_second": 1187.595,
      "eval_steps_per_second": 148.733,
      "step": 5500
    },
    {
      "entropy": 5.181309795379638,
      "epoch": 5.549672213817448,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002711128691449262,
      "loss": 4.6927,
      "mean_token_accuracy": 0.2034272789955139,
      "num_tokens": 12157018.0,
      "step": 5505
    },
    {
      "entropy": 5.234472560882568,
      "epoch": 5.5547150781643975,
      "grad_norm": 1.140625,
      "learning_rate": 0.000270716268851747,
      "loss": 4.7581,
      "mean_token_accuracy": 0.19358992725610732,
      "num_tokens": 12168002.0,
      "step": 5510
    },
    {
      "entropy": 5.2318216323852536,
      "epoch": 5.559757942511347,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00027031968187250305,
      "loss": 4.8019,
      "mean_token_accuracy": 0.19219024777412413,
      "num_tokens": 12181508.0,
      "step": 5515
    },
    {
      "entropy": 5.276045560836792,
      "epoch": 5.564800806858296,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002699231094397957,
      "loss": 4.8169,
      "mean_token_accuracy": 0.18816379010677337,
      "num_tokens": 12191844.0,
      "step": 5520
    },
    {
      "entropy": 5.210687828063965,
      "epoch": 5.569843671205245,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002695265527861815,
      "loss": 4.7844,
      "mean_token_accuracy": 0.1913653701543808,
      "num_tokens": 12202243.0,
      "step": 5525
    },
    {
      "entropy": 5.319292402267456,
      "epoch": 5.574886535552194,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002691300131441674,
      "loss": 4.8384,
      "mean_token_accuracy": 0.1935153052210808,
      "num_tokens": 12215333.0,
      "step": 5530
    },
    {
      "entropy": 5.25596923828125,
      "epoch": 5.579929399899143,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002687334917462079,
      "loss": 4.7388,
      "mean_token_accuracy": 0.19931214600801467,
      "num_tokens": 12225000.0,
      "step": 5535
    },
    {
      "entropy": 5.241505527496338,
      "epoch": 5.584972264246092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002683369898247005,
      "loss": 4.8152,
      "mean_token_accuracy": 0.19502967596054077,
      "num_tokens": 12236851.0,
      "step": 5540
    },
    {
      "entropy": 5.198987579345703,
      "epoch": 5.590015128593041,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002679405086119825,
      "loss": 4.7693,
      "mean_token_accuracy": 0.19551405161619187,
      "num_tokens": 12247639.0,
      "step": 5545
    },
    {
      "entropy": 5.202404022216797,
      "epoch": 5.59505799293999,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00026754404934032657,
      "loss": 4.7078,
      "mean_token_accuracy": 0.1998487576842308,
      "num_tokens": 12259647.0,
      "step": 5550
    },
    {
      "entropy": 5.275854444503784,
      "epoch": 5.600100857286939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002671476132419374,
      "loss": 4.8277,
      "mean_token_accuracy": 0.18575831800699233,
      "num_tokens": 12271489.0,
      "step": 5555
    },
    {
      "entropy": 5.20933084487915,
      "epoch": 5.605143721633888,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002667512015489472,
      "loss": 4.7872,
      "mean_token_accuracy": 0.1983243554830551,
      "num_tokens": 12282217.0,
      "step": 5560
    },
    {
      "entropy": 5.270116758346558,
      "epoch": 5.610186585980837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000266354815493413,
      "loss": 4.8006,
      "mean_token_accuracy": 0.190126633644104,
      "num_tokens": 12292519.0,
      "step": 5565
    },
    {
      "entropy": 5.166496419906617,
      "epoch": 5.615229450327786,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026595845630731166,
      "loss": 4.6227,
      "mean_token_accuracy": 0.20484723299741744,
      "num_tokens": 12303071.0,
      "step": 5570
    },
    {
      "entropy": 5.223015069961548,
      "epoch": 5.620272314674735,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026556212522253666,
      "loss": 4.7451,
      "mean_token_accuracy": 0.1969309315085411,
      "num_tokens": 12314262.0,
      "step": 5575
    },
    {
      "entropy": 5.281633520126343,
      "epoch": 5.625315179021684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002651658234708941,
      "loss": 4.8537,
      "mean_token_accuracy": 0.19031688421964646,
      "num_tokens": 12325257.0,
      "step": 5580
    },
    {
      "entropy": 5.249491024017334,
      "epoch": 5.630358043368633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026476955228409924,
      "loss": 4.7961,
      "mean_token_accuracy": 0.19422131329774855,
      "num_tokens": 12337066.0,
      "step": 5585
    },
    {
      "entropy": 5.1858726978302006,
      "epoch": 5.635400907715582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026437331289377175,
      "loss": 4.7493,
      "mean_token_accuracy": 0.19711011201143264,
      "num_tokens": 12349338.0,
      "step": 5590
    },
    {
      "entropy": 5.222999334335327,
      "epoch": 5.640443772062532,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002639771065314329,
      "loss": 4.7482,
      "mean_token_accuracy": 0.1965946599841118,
      "num_tokens": 12359536.0,
      "step": 5595
    },
    {
      "entropy": 5.212452030181884,
      "epoch": 5.645486636409481,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026358093442850134,
      "loss": 4.7688,
      "mean_token_accuracy": 0.19090568125247956,
      "num_tokens": 12369688.0,
      "step": 5600
    },
    {
      "entropy": 5.260433721542358,
      "epoch": 5.65052950075643,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002631847978162889,
      "loss": 4.7554,
      "mean_token_accuracy": 0.19530148804187775,
      "num_tokens": 12380288.0,
      "step": 5605
    },
    {
      "entropy": 5.147199964523315,
      "epoch": 5.655572365103379,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026278869792599745,
      "loss": 4.7295,
      "mean_token_accuracy": 0.1997440204024315,
      "num_tokens": 12391383.0,
      "step": 5610
    },
    {
      "entropy": 5.237601280212402,
      "epoch": 5.660615229450328,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002623926359887145,
      "loss": 4.7562,
      "mean_token_accuracy": 0.19333217144012452,
      "num_tokens": 12402213.0,
      "step": 5615
    },
    {
      "entropy": 5.126544713973999,
      "epoch": 5.665658093797277,
      "grad_norm": 1.25,
      "learning_rate": 0.00026199661323540973,
      "loss": 4.7524,
      "mean_token_accuracy": 0.1906924843788147,
      "num_tokens": 12412882.0,
      "step": 5620
    },
    {
      "entropy": 5.256667757034302,
      "epoch": 5.670700958144226,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026160063089693096,
      "loss": 4.808,
      "mean_token_accuracy": 0.1871042549610138,
      "num_tokens": 12423548.0,
      "step": 5625
    },
    {
      "entropy": 5.30175895690918,
      "epoch": 5.675743822491175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002612046902040006,
      "loss": 4.8133,
      "mean_token_accuracy": 0.18467668890953065,
      "num_tokens": 12434693.0,
      "step": 5630
    },
    {
      "entropy": 5.251724338531494,
      "epoch": 5.680786686838124,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002608087923872111,
      "loss": 4.7887,
      "mean_token_accuracy": 0.1958966836333275,
      "num_tokens": 12444502.0,
      "step": 5635
    },
    {
      "entropy": 5.272950029373169,
      "epoch": 5.685829551185073,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026041293867702235,
      "loss": 4.8426,
      "mean_token_accuracy": 0.1890074595808983,
      "num_tokens": 12456991.0,
      "step": 5640
    },
    {
      "entropy": 5.221624565124512,
      "epoch": 5.690872415532022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026001713030375664,
      "loss": 4.7224,
      "mean_token_accuracy": 0.19554358571767808,
      "num_tokens": 12466408.0,
      "step": 5645
    },
    {
      "entropy": 5.203773927688599,
      "epoch": 5.695915279878971,
      "grad_norm": 1.046875,
      "learning_rate": 0.00025962136849759564,
      "loss": 4.7878,
      "mean_token_accuracy": 0.19384435415267945,
      "num_tokens": 12479174.0,
      "step": 5650
    },
    {
      "entropy": 5.192513179779053,
      "epoch": 5.70095814422592,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025922565448857605,
      "loss": 4.7283,
      "mean_token_accuracy": 0.20808654725551606,
      "num_tokens": 12489149.0,
      "step": 5655
    },
    {
      "entropy": 5.340596914291382,
      "epoch": 5.706001008572869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025882998950658623,
      "loss": 4.8294,
      "mean_token_accuracy": 0.19129183441400527,
      "num_tokens": 12500417.0,
      "step": 5660
    },
    {
      "entropy": 5.208664751052856,
      "epoch": 5.711043872919818,
      "grad_norm": 1.125,
      "learning_rate": 0.00025843437478136196,
      "loss": 4.7068,
      "mean_token_accuracy": 0.1988130584359169,
      "num_tokens": 12511172.0,
      "step": 5665
    },
    {
      "entropy": 5.1949462890625,
      "epoch": 5.716086737266767,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025803881154248305,
      "loss": 4.6718,
      "mean_token_accuracy": 0.19958071559667587,
      "num_tokens": 12523203.0,
      "step": 5670
    },
    {
      "entropy": 5.240204000473023,
      "epoch": 5.7211296016137165,
      "grad_norm": 1.0,
      "learning_rate": 0.0002576433010193691,
      "loss": 4.7577,
      "mean_token_accuracy": 0.19309308528900146,
      "num_tokens": 12534224.0,
      "step": 5675
    },
    {
      "entropy": 5.138685178756714,
      "epoch": 5.726172465960666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000257247844441276,
      "loss": 4.615,
      "mean_token_accuracy": 0.20383508056402205,
      "num_tokens": 12545008.0,
      "step": 5680
    },
    {
      "entropy": 5.19198169708252,
      "epoch": 5.731215330307615,
      "grad_norm": 1.25,
      "learning_rate": 0.00025685244303729187,
      "loss": 4.7667,
      "mean_token_accuracy": 0.1919328212738037,
      "num_tokens": 12557888.0,
      "step": 5685
    },
    {
      "entropy": 5.220616292953491,
      "epoch": 5.736258194654564,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002564570980363334,
      "loss": 4.8422,
      "mean_token_accuracy": 0.18901431560516357,
      "num_tokens": 12569847.0,
      "step": 5690
    },
    {
      "entropy": 5.186214351654053,
      "epoch": 5.741301059001513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025606181066714195,
      "loss": 4.6566,
      "mean_token_accuracy": 0.2075856864452362,
      "num_tokens": 12580044.0,
      "step": 5695
    },
    {
      "entropy": 5.300652885437012,
      "epoch": 5.746343923348462,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002556665821582798,
      "loss": 4.7234,
      "mean_token_accuracy": 0.20001891106367112,
      "num_tokens": 12591286.0,
      "step": 5700
    },
    {
      "entropy": 5.258606052398681,
      "epoch": 5.751386787695411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002552714137381264,
      "loss": 4.7576,
      "mean_token_accuracy": 0.19304202795028685,
      "num_tokens": 12601987.0,
      "step": 5705
    },
    {
      "entropy": 5.209339904785156,
      "epoch": 5.75642965204236,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002548763066348741,
      "loss": 4.7491,
      "mean_token_accuracy": 0.19406707137823104,
      "num_tokens": 12612488.0,
      "step": 5710
    },
    {
      "entropy": 5.207173871994018,
      "epoch": 5.7614725163893095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002544812620765251,
      "loss": 4.7054,
      "mean_token_accuracy": 0.19700933992862701,
      "num_tokens": 12623098.0,
      "step": 5715
    },
    {
      "entropy": 5.164046049118042,
      "epoch": 5.766515380736259,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002540862812908868,
      "loss": 4.7357,
      "mean_token_accuracy": 0.20163710713386535,
      "num_tokens": 12634092.0,
      "step": 5720
    },
    {
      "entropy": 5.218921852111817,
      "epoch": 5.771558245083208,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002536913655055687,
      "loss": 4.7452,
      "mean_token_accuracy": 0.18917413502931596,
      "num_tokens": 12643959.0,
      "step": 5725
    },
    {
      "entropy": 5.195008897781372,
      "epoch": 5.776601109430157,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002532965159479781,
      "loss": 4.6655,
      "mean_token_accuracy": 0.20110440999269485,
      "num_tokens": 12653517.0,
      "step": 5730
    },
    {
      "entropy": 5.250275993347168,
      "epoch": 5.781643973777105,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002529017338453167,
      "loss": 4.8056,
      "mean_token_accuracy": 0.1931455537676811,
      "num_tokens": 12665963.0,
      "step": 5735
    },
    {
      "entropy": 5.174979734420776,
      "epoch": 5.786686838124054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002525070204245761,
      "loss": 4.7207,
      "mean_token_accuracy": 0.20448999404907225,
      "num_tokens": 12677572.0,
      "step": 5740
    },
    {
      "entropy": 5.269372367858887,
      "epoch": 5.791729702471003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025211237691253507,
      "loss": 4.7692,
      "mean_token_accuracy": 0.19401110261678695,
      "num_tokens": 12688560.0,
      "step": 5745
    },
    {
      "entropy": 5.2143761157989506,
      "epoch": 5.796772566817952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025171780453575447,
      "loss": 4.7467,
      "mean_token_accuracy": 0.1988970845937729,
      "num_tokens": 12700390.0,
      "step": 5750
    },
    {
      "entropy": 5.188512897491455,
      "epoch": 5.8018154311649015,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002513233045205746,
      "loss": 4.7373,
      "mean_token_accuracy": 0.20142971277236937,
      "num_tokens": 12710728.0,
      "step": 5755
    },
    {
      "entropy": 5.131809329986572,
      "epoch": 5.806858295511851,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002509288780931102,
      "loss": 4.6942,
      "mean_token_accuracy": 0.2000270038843155,
      "num_tokens": 12721148.0,
      "step": 5760
    },
    {
      "entropy": 5.208822536468506,
      "epoch": 5.8119011598588,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002505345264792482,
      "loss": 4.847,
      "mean_token_accuracy": 0.18846339583396912,
      "num_tokens": 12732911.0,
      "step": 5765
    },
    {
      "entropy": 5.290615797042847,
      "epoch": 5.816944024205749,
      "grad_norm": 1.078125,
      "learning_rate": 0.00025014025090464205,
      "loss": 4.7387,
      "mean_token_accuracy": 0.1964803084731102,
      "num_tokens": 12744358.0,
      "step": 5770
    },
    {
      "entropy": 5.2039361000061035,
      "epoch": 5.821986888552698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024974605259470975,
      "loss": 4.7453,
      "mean_token_accuracy": 0.1972372740507126,
      "num_tokens": 12753925.0,
      "step": 5775
    },
    {
      "entropy": 5.200745725631714,
      "epoch": 5.827029752899647,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024935193277462835,
      "loss": 4.7575,
      "mean_token_accuracy": 0.20077791661024094,
      "num_tokens": 12765038.0,
      "step": 5780
    },
    {
      "entropy": 5.231180620193482,
      "epoch": 5.832072617246596,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002489578926693317,
      "loss": 4.7235,
      "mean_token_accuracy": 0.19682149440050126,
      "num_tokens": 12775751.0,
      "step": 5785
    },
    {
      "entropy": 5.204396820068359,
      "epoch": 5.837115481593545,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002485639335035053,
      "loss": 4.797,
      "mean_token_accuracy": 0.18370633572340012,
      "num_tokens": 12786538.0,
      "step": 5790
    },
    {
      "entropy": 5.17878212928772,
      "epoch": 5.842158345940494,
      "grad_norm": 1.109375,
      "learning_rate": 0.00024817005650158345,
      "loss": 4.7954,
      "mean_token_accuracy": 0.18706645667552949,
      "num_tokens": 12798612.0,
      "step": 5795
    },
    {
      "entropy": 5.243161821365357,
      "epoch": 5.8472012102874436,
      "grad_norm": 1.265625,
      "learning_rate": 0.00024777626288774473,
      "loss": 4.7017,
      "mean_token_accuracy": 0.19595236629247664,
      "num_tokens": 12810681.0,
      "step": 5800
    },
    {
      "entropy": 5.1484133243560795,
      "epoch": 5.852244074634393,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002473825538859091,
      "loss": 4.7117,
      "mean_token_accuracy": 0.19342807084321975,
      "num_tokens": 12822077.0,
      "step": 5805
    },
    {
      "entropy": 5.148173475265503,
      "epoch": 5.857286938981342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002469889307197331,
      "loss": 4.7247,
      "mean_token_accuracy": 0.19496287703514098,
      "num_tokens": 12833558.0,
      "step": 5810
    },
    {
      "entropy": 5.227203512191773,
      "epoch": 5.86232980332829,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002465953946126065,
      "loss": 4.8547,
      "mean_token_accuracy": 0.1929842323064804,
      "num_tokens": 12846759.0,
      "step": 5815
    },
    {
      "entropy": 5.230418062210083,
      "epoch": 5.867372667675239,
      "grad_norm": 1.171875,
      "learning_rate": 0.00024620194678764875,
      "loss": 4.7648,
      "mean_token_accuracy": 0.19328173398971557,
      "num_tokens": 12857626.0,
      "step": 5820
    },
    {
      "entropy": 5.310681533813477,
      "epoch": 5.872415532022188,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002458085884677046,
      "loss": 4.855,
      "mean_token_accuracy": 0.18769813776016236,
      "num_tokens": 12870576.0,
      "step": 5825
    },
    {
      "entropy": 5.207978773117065,
      "epoch": 5.877458396369137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024541532087534097,
      "loss": 4.6426,
      "mean_token_accuracy": 0.20561135560274124,
      "num_tokens": 12880446.0,
      "step": 5830
    },
    {
      "entropy": 5.188767528533935,
      "epoch": 5.8825012607160865,
      "grad_norm": 1.125,
      "learning_rate": 0.0002450221452328424,
      "loss": 4.7974,
      "mean_token_accuracy": 0.1909268617630005,
      "num_tokens": 12891945.0,
      "step": 5835
    },
    {
      "entropy": 5.258281993865967,
      "epoch": 5.887544125063036,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024462906276220814,
      "loss": 4.8479,
      "mean_token_accuracy": 0.19595733731985093,
      "num_tokens": 12902143.0,
      "step": 5840
    },
    {
      "entropy": 5.2476927757263185,
      "epoch": 5.892586989409985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002442360746851473,
      "loss": 4.7814,
      "mean_token_accuracy": 0.19582216143608094,
      "num_tokens": 12913636.0,
      "step": 5845
    },
    {
      "entropy": 5.2833775043487545,
      "epoch": 5.897629853756934,
      "grad_norm": 1.15625,
      "learning_rate": 0.00024384318222307605,
      "loss": 4.8624,
      "mean_token_accuracy": 0.18303285539150238,
      "num_tokens": 12924924.0,
      "step": 5850
    },
    {
      "entropy": 5.226479244232178,
      "epoch": 5.902672718103883,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002434503865971131,
      "loss": 4.7664,
      "mean_token_accuracy": 0.1963245913386345,
      "num_tokens": 12936844.0,
      "step": 5855
    },
    {
      "entropy": 5.197439193725586,
      "epoch": 5.907715582450832,
      "grad_norm": 1.125,
      "learning_rate": 0.0002430576890280765,
      "loss": 4.7137,
      "mean_token_accuracy": 0.1971079081296921,
      "num_tokens": 12947268.0,
      "step": 5860
    },
    {
      "entropy": 5.1597517967224125,
      "epoch": 5.912758446797781,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00024266509073647913,
      "loss": 4.7012,
      "mean_token_accuracy": 0.19995448589324952,
      "num_tokens": 12957987.0,
      "step": 5865
    },
    {
      "entropy": 5.130568599700927,
      "epoch": 5.91780131114473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00024227259294252578,
      "loss": 4.6643,
      "mean_token_accuracy": 0.19997601062059403,
      "num_tokens": 12969533.0,
      "step": 5870
    },
    {
      "entropy": 5.230861759185791,
      "epoch": 5.922844175491679,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002418801968661084,
      "loss": 4.7884,
      "mean_token_accuracy": 0.19343979358673097,
      "num_tokens": 12979175.0,
      "step": 5875
    },
    {
      "entropy": 5.214277172088623,
      "epoch": 5.9278870398386285,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024148790372680324,
      "loss": 4.7292,
      "mean_token_accuracy": 0.19006066471338273,
      "num_tokens": 12989781.0,
      "step": 5880
    },
    {
      "entropy": 5.151724815368652,
      "epoch": 5.932929904185578,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00024109571474386631,
      "loss": 4.745,
      "mean_token_accuracy": 0.19279673993587493,
      "num_tokens": 13000281.0,
      "step": 5885
    },
    {
      "entropy": 5.238187265396118,
      "epoch": 5.937972768532527,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00024070363113623018,
      "loss": 4.8123,
      "mean_token_accuracy": 0.20003381222486497,
      "num_tokens": 13013025.0,
      "step": 5890
    },
    {
      "entropy": 5.154749584197998,
      "epoch": 5.943015632879476,
      "grad_norm": 1.015625,
      "learning_rate": 0.00024031165412249938,
      "loss": 4.6947,
      "mean_token_accuracy": 0.19811999797821045,
      "num_tokens": 13024827.0,
      "step": 5895
    },
    {
      "entropy": 5.220423936843872,
      "epoch": 5.948058497226425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000239919784920948,
      "loss": 4.7409,
      "mean_token_accuracy": 0.19765564799308777,
      "num_tokens": 13035157.0,
      "step": 5900
    },
    {
      "entropy": 5.203606748580933,
      "epoch": 5.953101361573374,
      "grad_norm": 1.171875,
      "learning_rate": 0.00023952802474951418,
      "loss": 4.7651,
      "mean_token_accuracy": 0.19300366193056107,
      "num_tokens": 13045937.0,
      "step": 5905
    },
    {
      "entropy": 5.248757553100586,
      "epoch": 5.958144225920323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023913637482579774,
      "loss": 4.834,
      "mean_token_accuracy": 0.18689459413290024,
      "num_tokens": 13057514.0,
      "step": 5910
    },
    {
      "entropy": 5.176025104522705,
      "epoch": 5.9631870902672715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00023874483636705563,
      "loss": 4.7253,
      "mean_token_accuracy": 0.19215343445539473,
      "num_tokens": 13068452.0,
      "step": 5915
    },
    {
      "entropy": 5.238114643096924,
      "epoch": 5.968229954614221,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023835341059019838,
      "loss": 4.7229,
      "mean_token_accuracy": 0.2026422992348671,
      "num_tokens": 13079974.0,
      "step": 5920
    },
    {
      "entropy": 5.114340448379517,
      "epoch": 5.97327281896117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023796209871178642,
      "loss": 4.7239,
      "mean_token_accuracy": 0.20155558735132217,
      "num_tokens": 13093430.0,
      "step": 5925
    },
    {
      "entropy": 5.18604211807251,
      "epoch": 5.978315683308119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023757090194802605,
      "loss": 4.7877,
      "mean_token_accuracy": 0.19459179937839508,
      "num_tokens": 13103986.0,
      "step": 5930
    },
    {
      "entropy": 5.191823101043701,
      "epoch": 5.983358547655068,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023717982151476592,
      "loss": 4.7062,
      "mean_token_accuracy": 0.20038924217224122,
      "num_tokens": 13114220.0,
      "step": 5935
    },
    {
      "entropy": 5.2114959239959715,
      "epoch": 5.988401412002017,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023678885862749282,
      "loss": 4.727,
      "mean_token_accuracy": 0.18876763433218002,
      "num_tokens": 13125492.0,
      "step": 5940
    },
    {
      "entropy": 5.228824710845947,
      "epoch": 5.993444276348966,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00023639801450132863,
      "loss": 4.7764,
      "mean_token_accuracy": 0.19640225917100906,
      "num_tokens": 13136314.0,
      "step": 5945
    },
    {
      "entropy": 5.174434328079224,
      "epoch": 5.998487140695915,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023600729035102585,
      "loss": 4.6939,
      "mean_token_accuracy": 0.19694596976041795,
      "num_tokens": 13147097.0,
      "step": 5950
    },
    {
      "entropy": 5.1626733673943415,
      "epoch": 6.0030257186081695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002356166873909642,
      "loss": 4.6062,
      "mean_token_accuracy": 0.20793199208047655,
      "num_tokens": 13156211.0,
      "step": 5955
    },
    {
      "entropy": 5.138905000686646,
      "epoch": 6.008068582955119,
      "grad_norm": 1.03125,
      "learning_rate": 0.00023522620683514645,
      "loss": 4.5134,
      "mean_token_accuracy": 0.21669998466968537,
      "num_tokens": 13168542.0,
      "step": 5960
    },
    {
      "entropy": 5.2159322738647464,
      "epoch": 6.013111447302068,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023483584989719553,
      "loss": 4.5713,
      "mean_token_accuracy": 0.21309427171945572,
      "num_tokens": 13179481.0,
      "step": 5965
    },
    {
      "entropy": 5.155423021316528,
      "epoch": 6.018154311649017,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023444561779034957,
      "loss": 4.4583,
      "mean_token_accuracy": 0.22517074942588805,
      "num_tokens": 13189499.0,
      "step": 5970
    },
    {
      "entropy": 5.183023118972779,
      "epoch": 6.023197175995966,
      "grad_norm": 1.0625,
      "learning_rate": 0.00023405551172745904,
      "loss": 4.5786,
      "mean_token_accuracy": 0.20624102503061295,
      "num_tokens": 13200865.0,
      "step": 5975
    },
    {
      "entropy": 5.169784450531006,
      "epoch": 6.028240040342915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00023366553292098247,
      "loss": 4.5566,
      "mean_token_accuracy": 0.2049471467733383,
      "num_tokens": 13211479.0,
      "step": 5980
    },
    {
      "entropy": 5.185241746902466,
      "epoch": 6.033282904689864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002332756825829832,
      "loss": 4.5371,
      "mean_token_accuracy": 0.21350208520889283,
      "num_tokens": 13222673.0,
      "step": 5985
    },
    {
      "entropy": 5.250400352478027,
      "epoch": 6.038325769036813,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002328859619251249,
      "loss": 4.5813,
      "mean_token_accuracy": 0.21167807877063752,
      "num_tokens": 13234728.0,
      "step": 5990
    },
    {
      "entropy": 5.2426896572113035,
      "epoch": 6.043368633383762,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023249637215866847,
      "loss": 4.5522,
      "mean_token_accuracy": 0.2049570620059967,
      "num_tokens": 13245476.0,
      "step": 5995
    },
    {
      "entropy": 5.18251633644104,
      "epoch": 6.048411497730711,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002321069144944678,
      "loss": 4.533,
      "mean_token_accuracy": 0.2136307939887047,
      "num_tokens": 13256337.0,
      "step": 6000
    },
    {
      "epoch": 6.048411497730711,
      "eval_entropy": 5.098752895384345,
      "eval_loss": 5.7129316329956055,
      "eval_mean_token_accuracy": 0.15298832941601295,
      "eval_num_tokens": 13256337.0,
      "eval_runtime": 2.8404,
      "eval_samples_per_second": 1104.779,
      "eval_steps_per_second": 138.361,
      "step": 6000
    },
    {
      "entropy": 5.082426071166992,
      "epoch": 6.05345436207766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023171759014296639,
      "loss": 4.4971,
      "mean_token_accuracy": 0.20884496122598648,
      "num_tokens": 13267531.0,
      "step": 6005
    },
    {
      "entropy": 5.23685655593872,
      "epoch": 6.058497226424609,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023132840031419316,
      "loss": 4.5941,
      "mean_token_accuracy": 0.20492557287216187,
      "num_tokens": 13279570.0,
      "step": 6010
    },
    {
      "entropy": 5.2473255634307865,
      "epoch": 6.063540090771558,
      "grad_norm": 1.125,
      "learning_rate": 0.00023093934621775926,
      "loss": 4.5712,
      "mean_token_accuracy": 0.21231979578733445,
      "num_tokens": 13290766.0,
      "step": 6015
    },
    {
      "entropy": 5.133472681045532,
      "epoch": 6.068582955118507,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023055042906285368,
      "loss": 4.4944,
      "mean_token_accuracy": 0.21481449753046036,
      "num_tokens": 13300853.0,
      "step": 6020
    },
    {
      "entropy": 5.155392694473266,
      "epoch": 6.073625819465456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002301616500582398,
      "loss": 4.5793,
      "mean_token_accuracy": 0.2072348937392235,
      "num_tokens": 13311555.0,
      "step": 6025
    },
    {
      "entropy": 5.2166766166687015,
      "epoch": 6.078668683812405,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00022977301041225207,
      "loss": 4.6155,
      "mean_token_accuracy": 0.20886164903640747,
      "num_tokens": 13323844.0,
      "step": 6030
    },
    {
      "entropy": 5.173759126663208,
      "epoch": 6.083711548159354,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022938451133279114,
      "loss": 4.558,
      "mean_token_accuracy": 0.2109983891248703,
      "num_tokens": 13335386.0,
      "step": 6035
    },
    {
      "entropy": 5.179766225814819,
      "epoch": 6.088754412506304,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022899615402732133,
      "loss": 4.5368,
      "mean_token_accuracy": 0.20905002653598787,
      "num_tokens": 13346796.0,
      "step": 6040
    },
    {
      "entropy": 5.193929147720337,
      "epoch": 6.093797276853253,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022860793970286592,
      "loss": 4.4743,
      "mean_token_accuracy": 0.21688882857561112,
      "num_tokens": 13356976.0,
      "step": 6045
    },
    {
      "entropy": 5.177070045471192,
      "epoch": 6.098840141200202,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002282198695660042,
      "loss": 4.6594,
      "mean_token_accuracy": 0.2055477261543274,
      "num_tokens": 13368224.0,
      "step": 6050
    },
    {
      "entropy": 5.189565944671631,
      "epoch": 6.103883005547151,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022783194482286684,
      "loss": 4.5619,
      "mean_token_accuracy": 0.2097763881087303,
      "num_tokens": 13380196.0,
      "step": 6055
    },
    {
      "entropy": 5.177653074264526,
      "epoch": 6.1089258698941,
      "grad_norm": 1.125,
      "learning_rate": 0.0002274441666791332,
      "loss": 4.5482,
      "mean_token_accuracy": 0.21607325226068497,
      "num_tokens": 13390611.0,
      "step": 6060
    },
    {
      "entropy": 5.22979769706726,
      "epoch": 6.113968734241049,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022705653634002648,
      "loss": 4.6295,
      "mean_token_accuracy": 0.2055166870355606,
      "num_tokens": 13401763.0,
      "step": 6065
    },
    {
      "entropy": 5.184354305267334,
      "epoch": 6.119011598587998,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022666905501031077,
      "loss": 4.6367,
      "mean_token_accuracy": 0.20524520874023439,
      "num_tokens": 13412843.0,
      "step": 6070
    },
    {
      "entropy": 5.125191974639892,
      "epoch": 6.124054462934947,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002262817238942869,
      "loss": 4.5457,
      "mean_token_accuracy": 0.21493498980998993,
      "num_tokens": 13425932.0,
      "step": 6075
    },
    {
      "entropy": 5.215444135665893,
      "epoch": 6.1290973272818965,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002258945441957892,
      "loss": 4.5351,
      "mean_token_accuracy": 0.2144885778427124,
      "num_tokens": 13437888.0,
      "step": 6080
    },
    {
      "entropy": 5.234664249420166,
      "epoch": 6.134140191628846,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002255075171181807,
      "loss": 4.6301,
      "mean_token_accuracy": 0.21060454845428467,
      "num_tokens": 13449121.0,
      "step": 6085
    },
    {
      "entropy": 5.194630479812622,
      "epoch": 6.139183055975794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022512064386435082,
      "loss": 4.5288,
      "mean_token_accuracy": 0.21392538249492646,
      "num_tokens": 13459403.0,
      "step": 6090
    },
    {
      "entropy": 5.237185287475586,
      "epoch": 6.144225920322743,
      "grad_norm": 1.25,
      "learning_rate": 0.0002247339256367103,
      "loss": 4.5666,
      "mean_token_accuracy": 0.20767272263765335,
      "num_tokens": 13468842.0,
      "step": 6095
    },
    {
      "entropy": 5.17270073890686,
      "epoch": 6.149268784669692,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022434736363718854,
      "loss": 4.5716,
      "mean_token_accuracy": 0.20732024312019348,
      "num_tokens": 13480938.0,
      "step": 6100
    },
    {
      "entropy": 5.112171173095703,
      "epoch": 6.154311649016641,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022396095906722897,
      "loss": 4.485,
      "mean_token_accuracy": 0.21847066134214402,
      "num_tokens": 13490846.0,
      "step": 6105
    },
    {
      "entropy": 5.179958200454712,
      "epoch": 6.15935451336359,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002235747131277861,
      "loss": 4.583,
      "mean_token_accuracy": 0.2104949563741684,
      "num_tokens": 13502549.0,
      "step": 6110
    },
    {
      "entropy": 5.193667221069336,
      "epoch": 6.164397377710539,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022318862701932102,
      "loss": 4.5556,
      "mean_token_accuracy": 0.21251111775636672,
      "num_tokens": 13512493.0,
      "step": 6115
    },
    {
      "entropy": 5.201950550079346,
      "epoch": 6.1694402420574885,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022280270194179836,
      "loss": 4.6133,
      "mean_token_accuracy": 0.20933446437120437,
      "num_tokens": 13523429.0,
      "step": 6120
    },
    {
      "entropy": 5.254075574874878,
      "epoch": 6.174483106404438,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022241693909468214,
      "loss": 4.639,
      "mean_token_accuracy": 0.20262111574411393,
      "num_tokens": 13535119.0,
      "step": 6125
    },
    {
      "entropy": 5.143132257461548,
      "epoch": 6.179525970751387,
      "grad_norm": 1.15625,
      "learning_rate": 0.00022203133967693217,
      "loss": 4.5478,
      "mean_token_accuracy": 0.21019617319107056,
      "num_tokens": 13546395.0,
      "step": 6130
    },
    {
      "entropy": 5.183080005645752,
      "epoch": 6.184568835098336,
      "grad_norm": 1.140625,
      "learning_rate": 0.00022164590488700042,
      "loss": 4.5749,
      "mean_token_accuracy": 0.21021025776863098,
      "num_tokens": 13556704.0,
      "step": 6135
    },
    {
      "entropy": 5.247574996948242,
      "epoch": 6.189611699445285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00022126063592282696,
      "loss": 4.5756,
      "mean_token_accuracy": 0.20780618190765382,
      "num_tokens": 13568410.0,
      "step": 6140
    },
    {
      "entropy": 5.242210054397583,
      "epoch": 6.194654563792234,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00022087553398183684,
      "loss": 4.6281,
      "mean_token_accuracy": 0.205714413523674,
      "num_tokens": 13580146.0,
      "step": 6145
    },
    {
      "entropy": 5.186987781524659,
      "epoch": 6.199697428139183,
      "grad_norm": 1.203125,
      "learning_rate": 0.00022049060026093555,
      "loss": 4.5662,
      "mean_token_accuracy": 0.20294594019651413,
      "num_tokens": 13592010.0,
      "step": 6150
    },
    {
      "entropy": 5.195967054367065,
      "epoch": 6.204740292486132,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00022010583595650617,
      "loss": 4.5407,
      "mean_token_accuracy": 0.21505373120307922,
      "num_tokens": 13602240.0,
      "step": 6155
    },
    {
      "entropy": 5.194594621658325,
      "epoch": 6.2097831568330815,
      "grad_norm": 1.15625,
      "learning_rate": 0.00021972124226440483,
      "loss": 4.5941,
      "mean_token_accuracy": 0.21471176743507386,
      "num_tokens": 13612455.0,
      "step": 6160
    },
    {
      "entropy": 5.183549547195435,
      "epoch": 6.214826021180031,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021933682037995772,
      "loss": 4.5818,
      "mean_token_accuracy": 0.2092095211148262,
      "num_tokens": 13622232.0,
      "step": 6165
    },
    {
      "entropy": 5.2338450908660885,
      "epoch": 6.21986888552698,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021895257149795694,
      "loss": 4.6213,
      "mean_token_accuracy": 0.20560964792966843,
      "num_tokens": 13633916.0,
      "step": 6170
    },
    {
      "entropy": 5.254591035842895,
      "epoch": 6.224911749873929,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002185684968126569,
      "loss": 4.6512,
      "mean_token_accuracy": 0.19962091445922853,
      "num_tokens": 13644477.0,
      "step": 6175
    },
    {
      "entropy": 5.251878166198731,
      "epoch": 6.229954614220877,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002181845975177703,
      "loss": 4.6361,
      "mean_token_accuracy": 0.207310451567173,
      "num_tokens": 13656648.0,
      "step": 6180
    },
    {
      "entropy": 5.215914058685303,
      "epoch": 6.234997478567826,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021780087480646548,
      "loss": 4.6327,
      "mean_token_accuracy": 0.2002495691180229,
      "num_tokens": 13667077.0,
      "step": 6185
    },
    {
      "entropy": 5.213680458068848,
      "epoch": 6.240040342914775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002174173298713611,
      "loss": 4.588,
      "mean_token_accuracy": 0.20517800599336625,
      "num_tokens": 13678382.0,
      "step": 6190
    },
    {
      "entropy": 5.161611461639405,
      "epoch": 6.245083207261724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021703396390452385,
      "loss": 4.5855,
      "mean_token_accuracy": 0.20608027875423432,
      "num_tokens": 13688757.0,
      "step": 6195
    },
    {
      "entropy": 5.196043872833252,
      "epoch": 6.2501260716086735,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002166507780974639,
      "loss": 4.6387,
      "mean_token_accuracy": 0.20285409986972808,
      "num_tokens": 13700169.0,
      "step": 6200
    },
    {
      "entropy": 5.223274278640747,
      "epoch": 6.255168935955623,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002162677736411317,
      "loss": 4.6078,
      "mean_token_accuracy": 0.202001591026783,
      "num_tokens": 13710686.0,
      "step": 6205
    },
    {
      "entropy": 5.186983585357666,
      "epoch": 6.260211800302572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00021588495172591377,
      "loss": 4.5988,
      "mean_token_accuracy": 0.21144547462463378,
      "num_tokens": 13722263.0,
      "step": 6210
    },
    {
      "entropy": 5.165117502212524,
      "epoch": 6.265254664649521,
      "grad_norm": 1.09375,
      "learning_rate": 0.00021550231354162957,
      "loss": 4.6218,
      "mean_token_accuracy": 0.20867250561714173,
      "num_tokens": 13733978.0,
      "step": 6215
    },
    {
      "entropy": 5.19357852935791,
      "epoch": 6.27029752899647,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00021511986027752733,
      "loss": 4.5788,
      "mean_token_accuracy": 0.2118837982416153,
      "num_tokens": 13746902.0,
      "step": 6220
    },
    {
      "entropy": 5.228260612487793,
      "epoch": 6.275340393343419,
      "grad_norm": 1.25,
      "learning_rate": 0.0002147375931222807,
      "loss": 4.5749,
      "mean_token_accuracy": 0.20851545929908752,
      "num_tokens": 13758174.0,
      "step": 6225
    },
    {
      "entropy": 5.1534820079803465,
      "epoch": 6.280383257690368,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021435551326398478,
      "loss": 4.5361,
      "mean_token_accuracy": 0.21433932036161424,
      "num_tokens": 13768287.0,
      "step": 6230
    },
    {
      "entropy": 5.175978660583496,
      "epoch": 6.285426122037317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00021397362189015263,
      "loss": 4.5945,
      "mean_token_accuracy": 0.20848405510187148,
      "num_tokens": 13781730.0,
      "step": 6235
    },
    {
      "entropy": 5.13439826965332,
      "epoch": 6.290468986384266,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021359192018771127,
      "loss": 4.5172,
      "mean_token_accuracy": 0.21607965528964995,
      "num_tokens": 13792136.0,
      "step": 6240
    },
    {
      "entropy": 5.0893322944641115,
      "epoch": 6.2955118507312156,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021321040934299852,
      "loss": 4.5733,
      "mean_token_accuracy": 0.21500775665044786,
      "num_tokens": 13804256.0,
      "step": 6245
    },
    {
      "entropy": 5.198020076751709,
      "epoch": 6.300554715078165,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021282909054175903,
      "loss": 4.55,
      "mean_token_accuracy": 0.21607322990894318,
      "num_tokens": 13814217.0,
      "step": 6250
    },
    {
      "entropy": 5.197706985473633,
      "epoch": 6.305597579425114,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002124479649691402,
      "loss": 4.528,
      "mean_token_accuracy": 0.2135801985859871,
      "num_tokens": 13824934.0,
      "step": 6255
    },
    {
      "entropy": 5.1761970043182375,
      "epoch": 6.310640443772063,
      "grad_norm": 1.25,
      "learning_rate": 0.00021206703380968923,
      "loss": 4.5401,
      "mean_token_accuracy": 0.21754625737667083,
      "num_tokens": 13836053.0,
      "step": 6260
    },
    {
      "entropy": 5.139744520187378,
      "epoch": 6.315683308119011,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021168629824734887,
      "loss": 4.5388,
      "mean_token_accuracy": 0.2147700384259224,
      "num_tokens": 13847571.0,
      "step": 6265
    },
    {
      "entropy": 5.181746244430542,
      "epoch": 6.32072617246596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021130575946545416,
      "loss": 4.5166,
      "mean_token_accuracy": 0.21209434270858765,
      "num_tokens": 13858704.0,
      "step": 6270
    },
    {
      "entropy": 5.138604068756104,
      "epoch": 6.325769036812909,
      "grad_norm": 1.171875,
      "learning_rate": 0.00021092541864672825,
      "loss": 4.5701,
      "mean_token_accuracy": 0.21043870151042937,
      "num_tokens": 13870577.0,
      "step": 6275
    },
    {
      "entropy": 5.1456702709197994,
      "epoch": 6.3308119011598585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00021054527697327935,
      "loss": 4.5565,
      "mean_token_accuracy": 0.21179594546556474,
      "num_tokens": 13881884.0,
      "step": 6280
    },
    {
      "entropy": 5.180002117156983,
      "epoch": 6.335854765506808,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021016533562659634,
      "loss": 4.6071,
      "mean_token_accuracy": 0.20812753289937974,
      "num_tokens": 13893458.0,
      "step": 6285
    },
    {
      "entropy": 5.194777107238769,
      "epoch": 6.340897629853757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020978559578754586,
      "loss": 4.5985,
      "mean_token_accuracy": 0.2063929706811905,
      "num_tokens": 13904001.0,
      "step": 6290
    },
    {
      "entropy": 5.215136575698852,
      "epoch": 6.345940494200706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000209406058636368,
      "loss": 4.5776,
      "mean_token_accuracy": 0.21138615757226945,
      "num_tokens": 13915019.0,
      "step": 6295
    },
    {
      "entropy": 5.148127460479737,
      "epoch": 6.350983358547655,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020902672535267315,
      "loss": 4.5305,
      "mean_token_accuracy": 0.21563087850809098,
      "num_tokens": 13924217.0,
      "step": 6300
    },
    {
      "entropy": 5.1839264869689945,
      "epoch": 6.356026222894604,
      "grad_norm": 1.140625,
      "learning_rate": 0.00020864759711543762,
      "loss": 4.6123,
      "mean_token_accuracy": 0.20721697956323623,
      "num_tokens": 13935000.0,
      "step": 6305
    },
    {
      "entropy": 5.2034289836883545,
      "epoch": 6.361069087241553,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002082686751030009,
      "loss": 4.5968,
      "mean_token_accuracy": 0.20836371332406997,
      "num_tokens": 13944295.0,
      "step": 6310
    },
    {
      "entropy": 5.233879613876343,
      "epoch": 6.366111951588502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002078899604930613,
      "loss": 4.5803,
      "mean_token_accuracy": 0.20929054021835328,
      "num_tokens": 13955093.0,
      "step": 6315
    },
    {
      "entropy": 5.123318481445312,
      "epoch": 6.371154815935451,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020751145446267267,
      "loss": 4.4662,
      "mean_token_accuracy": 0.2183082401752472,
      "num_tokens": 13965803.0,
      "step": 6320
    },
    {
      "entropy": 5.1577098846435545,
      "epoch": 6.3761976802824005,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002071331581882403,
      "loss": 4.6236,
      "mean_token_accuracy": 0.2057637020945549,
      "num_tokens": 13975826.0,
      "step": 6325
    },
    {
      "entropy": 5.175602388381958,
      "epoch": 6.38124054462935,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020675507284551793,
      "loss": 4.6017,
      "mean_token_accuracy": 0.20969808995723724,
      "num_tokens": 13987456.0,
      "step": 6330
    },
    {
      "entropy": 5.261620950698853,
      "epoch": 6.386283408976299,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00020637719960960333,
      "loss": 4.6161,
      "mean_token_accuracy": 0.20457296967506408,
      "num_tokens": 13998425.0,
      "step": 6335
    },
    {
      "entropy": 5.249192237854004,
      "epoch": 6.391326273323248,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020599953965493535,
      "loss": 4.552,
      "mean_token_accuracy": 0.20694674998521806,
      "num_tokens": 14009205.0,
      "step": 6340
    },
    {
      "entropy": 5.151955318450928,
      "epoch": 6.396369137670197,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020562209415528982,
      "loss": 4.5602,
      "mean_token_accuracy": 0.20886103063821793,
      "num_tokens": 14019654.0,
      "step": 6345
    },
    {
      "entropy": 5.170197343826294,
      "epoch": 6.401412002017146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00020524486428377605,
      "loss": 4.5866,
      "mean_token_accuracy": 0.21326925456523896,
      "num_tokens": 14030540.0,
      "step": 6350
    },
    {
      "entropy": 5.142443513870239,
      "epoch": 6.406454866364095,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002048678512128332,
      "loss": 4.5383,
      "mean_token_accuracy": 0.20592473745346068,
      "num_tokens": 14039894.0,
      "step": 6355
    },
    {
      "entropy": 5.1516721725463865,
      "epoch": 6.411497730711044,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020449105611422658,
      "loss": 4.5454,
      "mean_token_accuracy": 0.21459872275590897,
      "num_tokens": 14051507.0,
      "step": 6360
    },
    {
      "entropy": 5.188652658462525,
      "epoch": 6.416540595057993,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00020411448015904415,
      "loss": 4.6308,
      "mean_token_accuracy": 0.20156168490648269,
      "num_tokens": 14062137.0,
      "step": 6365
    },
    {
      "entropy": 5.1740625381469725,
      "epoch": 6.421583459404942,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020373812451769248,
      "loss": 4.528,
      "mean_token_accuracy": 0.21415633857250213,
      "num_tokens": 14072787.0,
      "step": 6370
    },
    {
      "entropy": 5.255901098251343,
      "epoch": 6.426626323751891,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020336199035989372,
      "loss": 4.6138,
      "mean_token_accuracy": 0.20252186805009842,
      "num_tokens": 14081705.0,
      "step": 6375
    },
    {
      "entropy": 5.1842756271362305,
      "epoch": 6.43166918809884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002029860788546814,
      "loss": 4.6037,
      "mean_token_accuracy": 0.20670073181390763,
      "num_tokens": 14093619.0,
      "step": 6380
    },
    {
      "entropy": 5.161433219909668,
      "epoch": 6.436712052445789,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00020261039117039725,
      "loss": 4.5948,
      "mean_token_accuracy": 0.20324929505586625,
      "num_tokens": 14104616.0,
      "step": 6385
    },
    {
      "entropy": 5.183618021011353,
      "epoch": 6.441754916792738,
      "grad_norm": 1.15625,
      "learning_rate": 0.00020223492847468722,
      "loss": 4.5667,
      "mean_token_accuracy": 0.21981403976678848,
      "num_tokens": 14115200.0,
      "step": 6390
    },
    {
      "entropy": 5.159724521636963,
      "epoch": 6.446797781139687,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002018596919344981,
      "loss": 4.5816,
      "mean_token_accuracy": 0.2080659806728363,
      "num_tokens": 14126570.0,
      "step": 6395
    },
    {
      "entropy": 5.1927032470703125,
      "epoch": 6.451840645486636,
      "grad_norm": 1.171875,
      "learning_rate": 0.00020148468271607352,
      "loss": 4.6236,
      "mean_token_accuracy": 0.19845569878816605,
      "num_tokens": 14137558.0,
      "step": 6400
    },
    {
      "entropy": 5.156425333023071,
      "epoch": 6.4568835098335855,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020110990198495105,
      "loss": 4.515,
      "mean_token_accuracy": 0.21909228712320328,
      "num_tokens": 14148162.0,
      "step": 6405
    },
    {
      "entropy": 5.188686466217041,
      "epoch": 6.461926374180535,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002007353509059575,
      "loss": 4.5838,
      "mean_token_accuracy": 0.20717293471097947,
      "num_tokens": 14158593.0,
      "step": 6410
    },
    {
      "entropy": 5.19873366355896,
      "epoch": 6.466969238527484,
      "grad_norm": 1.125,
      "learning_rate": 0.00020036103064320664,
      "loss": 4.5859,
      "mean_token_accuracy": 0.20157547742128373,
      "num_tokens": 14170065.0,
      "step": 6415
    },
    {
      "entropy": 5.160470628738404,
      "epoch": 6.472012102874433,
      "grad_norm": 1.296875,
      "learning_rate": 0.00019998694236009423,
      "loss": 4.4961,
      "mean_token_accuracy": 0.21519092619419097,
      "num_tokens": 14180315.0,
      "step": 6420
    },
    {
      "entropy": 5.165270090103149,
      "epoch": 6.477054967221382,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001996130872192955,
      "loss": 4.5366,
      "mean_token_accuracy": 0.21647629737854004,
      "num_tokens": 14191634.0,
      "step": 6425
    },
    {
      "entropy": 5.1952508926391605,
      "epoch": 6.482097831568331,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019923946638276052,
      "loss": 4.6016,
      "mean_token_accuracy": 0.2130608916282654,
      "num_tokens": 14204022.0,
      "step": 6430
    },
    {
      "entropy": 5.1614264965057375,
      "epoch": 6.48714069591528,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001988660810117118,
      "loss": 4.5904,
      "mean_token_accuracy": 0.20497036576271058,
      "num_tokens": 14213808.0,
      "step": 6435
    },
    {
      "entropy": 5.153904247283935,
      "epoch": 6.492183560262229,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001984929322666395,
      "loss": 4.5483,
      "mean_token_accuracy": 0.2136194184422493,
      "num_tokens": 14224436.0,
      "step": 6440
    },
    {
      "entropy": 5.236234521865844,
      "epoch": 6.4972264246091775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019812002130729857,
      "loss": 4.591,
      "mean_token_accuracy": 0.21542836129665374,
      "num_tokens": 14234689.0,
      "step": 6445
    },
    {
      "entropy": 5.184971666336059,
      "epoch": 6.502269288956127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00019774734929270483,
      "loss": 4.5478,
      "mean_token_accuracy": 0.21050220876932144,
      "num_tokens": 14245868.0,
      "step": 6450
    },
    {
      "entropy": 5.186728811264038,
      "epoch": 6.507312153303076,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019737491738113167,
      "loss": 4.6302,
      "mean_token_accuracy": 0.20799246728420256,
      "num_tokens": 14257545.0,
      "step": 6455
    },
    {
      "entropy": 5.1994654655456545,
      "epoch": 6.512355017650025,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019700272673010588,
      "loss": 4.6269,
      "mean_token_accuracy": 0.20820811688899993,
      "num_tokens": 14268766.0,
      "step": 6460
    },
    {
      "entropy": 5.193571281433106,
      "epoch": 6.517397881996974,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001966307784964046,
      "loss": 4.6417,
      "mean_token_accuracy": 0.1999490439891815,
      "num_tokens": 14281036.0,
      "step": 6465
    },
    {
      "entropy": 5.182217645645141,
      "epoch": 6.522440746343923,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00019625907383605162,
      "loss": 4.6469,
      "mean_token_accuracy": 0.204874250292778,
      "num_tokens": 14293040.0,
      "step": 6470
    },
    {
      "entropy": 5.127573871612549,
      "epoch": 6.527483610690872,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019588761390431354,
      "loss": 4.5282,
      "mean_token_accuracy": 0.21490539759397506,
      "num_tokens": 14303783.0,
      "step": 6475
    },
    {
      "entropy": 5.191330862045288,
      "epoch": 6.532526475037821,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019551639985569642,
      "loss": 4.568,
      "mean_token_accuracy": 0.21617287993431092,
      "num_tokens": 14315070.0,
      "step": 6480
    },
    {
      "entropy": 5.163919162750244,
      "epoch": 6.5375693393847705,
      "grad_norm": 1.140625,
      "learning_rate": 0.00019514543284394203,
      "loss": 4.5437,
      "mean_token_accuracy": 0.21120428591966628,
      "num_tokens": 14325794.0,
      "step": 6485
    },
    {
      "entropy": 5.128492164611816,
      "epoch": 6.54261220373172,
      "grad_norm": 1.234375,
      "learning_rate": 0.00019477471402202452,
      "loss": 4.5701,
      "mean_token_accuracy": 0.21161746829748154,
      "num_tokens": 14336813.0,
      "step": 6490
    },
    {
      "entropy": 5.223765516281128,
      "epoch": 6.547655068078669,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019440424454214645,
      "loss": 4.5706,
      "mean_token_accuracy": 0.21472148597240448,
      "num_tokens": 14347107.0,
      "step": 6495
    },
    {
      "entropy": 5.205308675765991,
      "epoch": 6.552697932425618,
      "grad_norm": 1.25,
      "learning_rate": 0.00019403402555573547,
      "loss": 4.6325,
      "mean_token_accuracy": 0.2114514708518982,
      "num_tokens": 14359694.0,
      "step": 6500
    },
    {
      "epoch": 6.552697932425618,
      "eval_entropy": 5.097094129363392,
      "eval_loss": 5.6999921798706055,
      "eval_mean_token_accuracy": 0.15392748990662528,
      "eval_num_tokens": 14359694.0,
      "eval_runtime": 2.6317,
      "eval_samples_per_second": 1192.379,
      "eval_steps_per_second": 149.332,
      "step": 6500
    },
    {
      "entropy": 5.130491876602173,
      "epoch": 6.557740796772567,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00019366405821344074,
      "loss": 4.5879,
      "mean_token_accuracy": 0.2102610409259796,
      "num_tokens": 14370203.0,
      "step": 6505
    },
    {
      "entropy": 5.155288314819336,
      "epoch": 6.562783661119516,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001932943436651295,
      "loss": 4.6292,
      "mean_token_accuracy": 0.21059160828590393,
      "num_tokens": 14384484.0,
      "step": 6510
    },
    {
      "entropy": 5.149600982666016,
      "epoch": 6.567826525466465,
      "grad_norm": 1.265625,
      "learning_rate": 0.00019292488305988276,
      "loss": 4.5849,
      "mean_token_accuracy": 0.2041059121489525,
      "num_tokens": 14395637.0,
      "step": 6515
    },
    {
      "entropy": 5.295889139175415,
      "epoch": 6.572869389813414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001925556775459929,
      "loss": 4.6549,
      "mean_token_accuracy": 0.20161703377962112,
      "num_tokens": 14406870.0,
      "step": 6520
    },
    {
      "entropy": 5.216554355621338,
      "epoch": 6.577912254160363,
      "grad_norm": 1.234375,
      "learning_rate": 0.000192186728270959,
      "loss": 4.6132,
      "mean_token_accuracy": 0.2054870307445526,
      "num_tokens": 14417480.0,
      "step": 6525
    },
    {
      "entropy": 5.225105142593383,
      "epoch": 6.5829551185073125,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000191818036381484,
      "loss": 4.5504,
      "mean_token_accuracy": 0.21343085467815398,
      "num_tokens": 14427508.0,
      "step": 6530
    },
    {
      "entropy": 5.063633012771606,
      "epoch": 6.587997982854262,
      "grad_norm": 1.125,
      "learning_rate": 0.00019144960302347065,
      "loss": 4.4543,
      "mean_token_accuracy": 0.21671183705329894,
      "num_tokens": 14439113.0,
      "step": 6535
    },
    {
      "entropy": 5.184521198272705,
      "epoch": 6.593040847201211,
      "grad_norm": 1.203125,
      "learning_rate": 0.00019108142934201854,
      "loss": 4.6616,
      "mean_token_accuracy": 0.20681383907794954,
      "num_tokens": 14449832.0,
      "step": 6540
    },
    {
      "entropy": 5.21746187210083,
      "epoch": 6.598083711548159,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019071351648141977,
      "loss": 4.5765,
      "mean_token_accuracy": 0.21272105127573013,
      "num_tokens": 14459301.0,
      "step": 6545
    },
    {
      "entropy": 5.1809553623199465,
      "epoch": 6.603126575895108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00019034586558515625,
      "loss": 4.5693,
      "mean_token_accuracy": 0.2061496451497078,
      "num_tokens": 14470634.0,
      "step": 6550
    },
    {
      "entropy": 5.149834728240966,
      "epoch": 6.608169440242057,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018997847779589518,
      "loss": 4.5996,
      "mean_token_accuracy": 0.20664474964141846,
      "num_tokens": 14481226.0,
      "step": 6555
    },
    {
      "entropy": 5.214832401275634,
      "epoch": 6.613212304589006,
      "grad_norm": 1.3125,
      "learning_rate": 0.00018961135425548642,
      "loss": 4.58,
      "mean_token_accuracy": 0.214280803501606,
      "num_tokens": 14491810.0,
      "step": 6560
    },
    {
      "entropy": 5.126802253723144,
      "epoch": 6.6182551689359554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018924449610495836,
      "loss": 4.5101,
      "mean_token_accuracy": 0.21520401239395143,
      "num_tokens": 14503406.0,
      "step": 6565
    },
    {
      "entropy": 5.139012336730957,
      "epoch": 6.623298033282905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00018887790448451458,
      "loss": 4.5512,
      "mean_token_accuracy": 0.2125125616788864,
      "num_tokens": 14514706.0,
      "step": 6570
    },
    {
      "entropy": 5.162431764602661,
      "epoch": 6.628340897629854,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001885115805335304,
      "loss": 4.6277,
      "mean_token_accuracy": 0.2061990424990654,
      "num_tokens": 14526168.0,
      "step": 6575
    },
    {
      "entropy": 5.173348140716553,
      "epoch": 6.633383761976803,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018814552539054896,
      "loss": 4.5144,
      "mean_token_accuracy": 0.2138102248311043,
      "num_tokens": 14536762.0,
      "step": 6580
    },
    {
      "entropy": 5.186430263519287,
      "epoch": 6.638426626323752,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00018777974019327826,
      "loss": 4.6226,
      "mean_token_accuracy": 0.2106246158480644,
      "num_tokens": 14548101.0,
      "step": 6585
    },
    {
      "entropy": 5.223225069046021,
      "epoch": 6.643469490670701,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018741422607858693,
      "loss": 4.6149,
      "mean_token_accuracy": 0.2099144384264946,
      "num_tokens": 14559504.0,
      "step": 6590
    },
    {
      "entropy": 5.280319452285767,
      "epoch": 6.64851235501765,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018704898418250143,
      "loss": 4.6538,
      "mean_token_accuracy": 0.2066337212920189,
      "num_tokens": 14570558.0,
      "step": 6595
    },
    {
      "entropy": 5.207668399810791,
      "epoch": 6.653555219364599,
      "grad_norm": 1.15625,
      "learning_rate": 0.00018668401564020192,
      "loss": 4.6758,
      "mean_token_accuracy": 0.2020114004611969,
      "num_tokens": 14581328.0,
      "step": 6600
    },
    {
      "entropy": 5.196731662750244,
      "epoch": 6.658598083711548,
      "grad_norm": 1.125,
      "learning_rate": 0.00018631932158601905,
      "loss": 4.625,
      "mean_token_accuracy": 0.206864795088768,
      "num_tokens": 14593351.0,
      "step": 6605
    },
    {
      "entropy": 5.1712418556213375,
      "epoch": 6.6636409480584975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018595490315343038,
      "loss": 4.59,
      "mean_token_accuracy": 0.2127873942255974,
      "num_tokens": 14604837.0,
      "step": 6610
    },
    {
      "entropy": 5.219285392761231,
      "epoch": 6.668683812405447,
      "grad_norm": 1.25,
      "learning_rate": 0.00018559076147505688,
      "loss": 4.7086,
      "mean_token_accuracy": 0.1970106318593025,
      "num_tokens": 14614897.0,
      "step": 6615
    },
    {
      "entropy": 5.219979286193848,
      "epoch": 6.673726676752396,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018522689768265911,
      "loss": 4.6608,
      "mean_token_accuracy": 0.2062785267829895,
      "num_tokens": 14625974.0,
      "step": 6620
    },
    {
      "entropy": 5.180649328231811,
      "epoch": 6.678769541099344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00018486331290713427,
      "loss": 4.5514,
      "mean_token_accuracy": 0.2047770008444786,
      "num_tokens": 14636286.0,
      "step": 6625
    },
    {
      "entropy": 5.230290794372559,
      "epoch": 6.683812405446293,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018450000827851217,
      "loss": 4.6123,
      "mean_token_accuracy": 0.2125822424888611,
      "num_tokens": 14646389.0,
      "step": 6630
    },
    {
      "entropy": 5.231647968292236,
      "epoch": 6.688855269793242,
      "grad_norm": 1.265625,
      "learning_rate": 0.000184136984925952,
      "loss": 4.6199,
      "mean_token_accuracy": 0.21079246699810028,
      "num_tokens": 14656762.0,
      "step": 6635
    },
    {
      "entropy": 5.18216724395752,
      "epoch": 6.693898134140191,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001837742439777386,
      "loss": 4.5738,
      "mean_token_accuracy": 0.20315176099538804,
      "num_tokens": 14668333.0,
      "step": 6640
    },
    {
      "entropy": 5.211260795593262,
      "epoch": 6.69894099848714,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018341178656127948,
      "loss": 4.5965,
      "mean_token_accuracy": 0.20475521683692932,
      "num_tokens": 14678269.0,
      "step": 6645
    },
    {
      "entropy": 5.225475120544433,
      "epoch": 6.7039838628340895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00018304961380310038,
      "loss": 4.5815,
      "mean_token_accuracy": 0.20184740126132966,
      "num_tokens": 14688938.0,
      "step": 6650
    },
    {
      "entropy": 5.138168478012085,
      "epoch": 6.709026727181039,
      "grad_norm": 1.1875,
      "learning_rate": 0.00018268772682884273,
      "loss": 4.6377,
      "mean_token_accuracy": 0.20295664966106414,
      "num_tokens": 14700458.0,
      "step": 6655
    },
    {
      "entropy": 5.146854066848755,
      "epoch": 6.714069591527988,
      "grad_norm": 1.125,
      "learning_rate": 0.00018232612676325944,
      "loss": 4.6018,
      "mean_token_accuracy": 0.2048575684428215,
      "num_tokens": 14711456.0,
      "step": 6660
    },
    {
      "entropy": 5.212303829193115,
      "epoch": 6.719112455874937,
      "grad_norm": 1.03125,
      "learning_rate": 0.0001819648147302121,
      "loss": 4.6163,
      "mean_token_accuracy": 0.20662684440612794,
      "num_tokens": 14723112.0,
      "step": 6665
    },
    {
      "entropy": 5.249505996704102,
      "epoch": 6.724155320221886,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018160379185266662,
      "loss": 4.6611,
      "mean_token_accuracy": 0.1988249585032463,
      "num_tokens": 14735024.0,
      "step": 6670
    },
    {
      "entropy": 5.149381828308106,
      "epoch": 6.729198184568835,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00018124305925269058,
      "loss": 4.557,
      "mean_token_accuracy": 0.21637227684259414,
      "num_tokens": 14746481.0,
      "step": 6675
    },
    {
      "entropy": 5.137785720825195,
      "epoch": 6.734241048915784,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018088261805144918,
      "loss": 4.5093,
      "mean_token_accuracy": 0.21421371400356293,
      "num_tokens": 14758010.0,
      "step": 6680
    },
    {
      "entropy": 5.171672725677491,
      "epoch": 6.739283913262733,
      "grad_norm": 1.140625,
      "learning_rate": 0.00018052246936920192,
      "loss": 4.6272,
      "mean_token_accuracy": 0.20709343552589415,
      "num_tokens": 14769353.0,
      "step": 6685
    },
    {
      "entropy": 5.182807683944702,
      "epoch": 6.7443267776096825,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00018016261432529928,
      "loss": 4.5826,
      "mean_token_accuracy": 0.20868770629167557,
      "num_tokens": 14779690.0,
      "step": 6690
    },
    {
      "entropy": 5.211354398727417,
      "epoch": 6.749369641956632,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017980305403817893,
      "loss": 4.6164,
      "mean_token_accuracy": 0.20536208301782607,
      "num_tokens": 14789902.0,
      "step": 6695
    },
    {
      "entropy": 5.2027839660644535,
      "epoch": 6.754412506303581,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017944378962536279,
      "loss": 4.6163,
      "mean_token_accuracy": 0.21072118431329728,
      "num_tokens": 14799880.0,
      "step": 6700
    },
    {
      "entropy": 5.213867473602295,
      "epoch": 6.75945537065053,
      "grad_norm": 1.046875,
      "learning_rate": 0.00017908482220345268,
      "loss": 4.635,
      "mean_token_accuracy": 0.20993876904249192,
      "num_tokens": 14811516.0,
      "step": 6705
    },
    {
      "entropy": 5.190399169921875,
      "epoch": 6.764498234997479,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017872615288812787,
      "loss": 4.5672,
      "mean_token_accuracy": 0.21313044130802156,
      "num_tokens": 14823930.0,
      "step": 6710
    },
    {
      "entropy": 5.249651479721069,
      "epoch": 6.769541099344428,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017836778279414057,
      "loss": 4.6489,
      "mean_token_accuracy": 0.20421119779348373,
      "num_tokens": 14835867.0,
      "step": 6715
    },
    {
      "entropy": 5.189886093139648,
      "epoch": 6.774583963691377,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017800971303531354,
      "loss": 4.6115,
      "mean_token_accuracy": 0.20635173171758653,
      "num_tokens": 14846261.0,
      "step": 6720
    },
    {
      "entropy": 5.20282940864563,
      "epoch": 6.779626828038325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001776519447245356,
      "loss": 4.6266,
      "mean_token_accuracy": 0.20908863991498947,
      "num_tokens": 14856051.0,
      "step": 6725
    },
    {
      "entropy": 5.20092282295227,
      "epoch": 6.7846696923852745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00017729447897375927,
      "loss": 4.6396,
      "mean_token_accuracy": 0.20754891335964204,
      "num_tokens": 14867089.0,
      "step": 6730
    },
    {
      "entropy": 5.168969297409058,
      "epoch": 6.789712556732224,
      "grad_norm": 1.15625,
      "learning_rate": 0.00017693731689399605,
      "loss": 4.6028,
      "mean_token_accuracy": 0.2038504123687744,
      "num_tokens": 14877896.0,
      "step": 6735
    },
    {
      "entropy": 5.183093643188476,
      "epoch": 6.794755421079173,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017658045959531416,
      "loss": 4.527,
      "mean_token_accuracy": 0.21362382918596268,
      "num_tokens": 14890027.0,
      "step": 6740
    },
    {
      "entropy": 5.241463041305542,
      "epoch": 6.799798285426122,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017622390818683404,
      "loss": 4.5821,
      "mean_token_accuracy": 0.21604207903146744,
      "num_tokens": 14901411.0,
      "step": 6745
    },
    {
      "entropy": 5.209956359863281,
      "epoch": 6.804841149773071,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00017586766377672593,
      "loss": 4.5129,
      "mean_token_accuracy": 0.220297010242939,
      "num_tokens": 14910710.0,
      "step": 6750
    },
    {
      "entropy": 5.1855470657348635,
      "epoch": 6.80988401412002,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017551172747220552,
      "loss": 4.57,
      "mean_token_accuracy": 0.21107130497694016,
      "num_tokens": 14921809.0,
      "step": 6755
    },
    {
      "entropy": 5.173009538650513,
      "epoch": 6.814926878466969,
      "grad_norm": 1.140625,
      "learning_rate": 0.00017515610037953112,
      "loss": 4.639,
      "mean_token_accuracy": 0.19935745596885682,
      "num_tokens": 14933584.0,
      "step": 6760
    },
    {
      "entropy": 5.170605230331421,
      "epoch": 6.819969742813918,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001748007836039999,
      "loss": 4.5668,
      "mean_token_accuracy": 0.21055862605571746,
      "num_tokens": 14944237.0,
      "step": 6765
    },
    {
      "entropy": 5.158693742752075,
      "epoch": 6.825012607160867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017444577824994475,
      "loss": 4.6234,
      "mean_token_accuracy": 0.21646799594163896,
      "num_tokens": 14955085.0,
      "step": 6770
    },
    {
      "entropy": 5.178486251831055,
      "epoch": 6.8300554715078166,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001740910854207303,
      "loss": 4.6791,
      "mean_token_accuracy": 0.20297423154115676,
      "num_tokens": 14967190.0,
      "step": 6775
    },
    {
      "entropy": 5.205316209793091,
      "epoch": 6.835098335854766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00017373670621875016,
      "loss": 4.6282,
      "mean_token_accuracy": 0.21252540946006776,
      "num_tokens": 14978513.0,
      "step": 6780
    },
    {
      "entropy": 5.240381336212158,
      "epoch": 6.840141200201715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000173382641745423,
      "loss": 4.6254,
      "mean_token_accuracy": 0.20698291659355164,
      "num_tokens": 14990412.0,
      "step": 6785
    },
    {
      "entropy": 5.2067045211792,
      "epoch": 6.845184064548664,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001730288931011893,
      "loss": 4.6311,
      "mean_token_accuracy": 0.20755360424518585,
      "num_tokens": 15001144.0,
      "step": 6790
    },
    {
      "entropy": 5.209230756759643,
      "epoch": 6.850226928895613,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017267546138550823,
      "loss": 4.621,
      "mean_token_accuracy": 0.20631584227085115,
      "num_tokens": 15013398.0,
      "step": 6795
    },
    {
      "entropy": 5.186916923522949,
      "epoch": 6.855269793242562,
      "grad_norm": 1.21875,
      "learning_rate": 0.00017232234769685358,
      "loss": 4.5918,
      "mean_token_accuracy": 0.2124532088637352,
      "num_tokens": 15024795.0,
      "step": 6800
    },
    {
      "entropy": 5.163653230667114,
      "epoch": 6.86031265758951,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00017196955313271086,
      "loss": 4.57,
      "mean_token_accuracy": 0.21256305426359176,
      "num_tokens": 15036269.0,
      "step": 6805
    },
    {
      "entropy": 5.166349840164185,
      "epoch": 6.8653555219364595,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017161707878957368,
      "loss": 4.6186,
      "mean_token_accuracy": 0.20506519228219985,
      "num_tokens": 15047161.0,
      "step": 6810
    },
    {
      "entropy": 5.193677663803101,
      "epoch": 6.870398386283409,
      "grad_norm": 1.109375,
      "learning_rate": 0.00017126492576294053,
      "loss": 4.5857,
      "mean_token_accuracy": 0.21502889990806578,
      "num_tokens": 15058191.0,
      "step": 6815
    },
    {
      "entropy": 5.153065395355225,
      "epoch": 6.875441250630358,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001709130951473109,
      "loss": 4.5062,
      "mean_token_accuracy": 0.2173328310251236,
      "num_tokens": 15069376.0,
      "step": 6820
    },
    {
      "entropy": 5.173217678070069,
      "epoch": 6.880484114977307,
      "grad_norm": 1.234375,
      "learning_rate": 0.00017056158803618277,
      "loss": 4.6958,
      "mean_token_accuracy": 0.20229191184043885,
      "num_tokens": 15080538.0,
      "step": 6825
    },
    {
      "entropy": 5.188806962966919,
      "epoch": 6.885526979324256,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017021040552204814,
      "loss": 4.6491,
      "mean_token_accuracy": 0.20888241231441498,
      "num_tokens": 15092193.0,
      "step": 6830
    },
    {
      "entropy": 5.185833692550659,
      "epoch": 6.890569843671205,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016985954869639048,
      "loss": 4.5848,
      "mean_token_accuracy": 0.21802718937397003,
      "num_tokens": 15102741.0,
      "step": 6835
    },
    {
      "entropy": 5.18069314956665,
      "epoch": 6.895612708018154,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950901864968076,
      "loss": 4.4912,
      "mean_token_accuracy": 0.21149059981107712,
      "num_tokens": 15113602.0,
      "step": 6840
    },
    {
      "entropy": 5.213942813873291,
      "epoch": 6.900655572365103,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016915881647137465,
      "loss": 4.6199,
      "mean_token_accuracy": 0.21102805584669113,
      "num_tokens": 15125092.0,
      "step": 6845
    },
    {
      "entropy": 5.151121330261231,
      "epoch": 6.905698436712052,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001688089432499084,
      "loss": 4.5382,
      "mean_token_accuracy": 0.21257967054843901,
      "num_tokens": 15136205.0,
      "step": 6850
    },
    {
      "entropy": 5.124761867523193,
      "epoch": 6.9107413010590015,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016845940007269617,
      "loss": 4.5517,
      "mean_token_accuracy": 0.21635200828313828,
      "num_tokens": 15147198.0,
      "step": 6855
    },
    {
      "entropy": 5.147335243225098,
      "epoch": 6.915784165405951,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016811018802612615,
      "loss": 4.5963,
      "mean_token_accuracy": 0.20654091089963914,
      "num_tokens": 15157608.0,
      "step": 6860
    },
    {
      "entropy": 5.207252168655396,
      "epoch": 6.9208270297529,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016776130819555758,
      "loss": 4.5393,
      "mean_token_accuracy": 0.21678743809461593,
      "num_tokens": 15168132.0,
      "step": 6865
    },
    {
      "entropy": 5.163896656036377,
      "epoch": 6.925869894099849,
      "grad_norm": 1.25,
      "learning_rate": 0.00016741276166531684,
      "loss": 4.5512,
      "mean_token_accuracy": 0.21622716784477233,
      "num_tokens": 15178565.0,
      "step": 6870
    },
    {
      "entropy": 5.083998203277588,
      "epoch": 6.930912758446798,
      "grad_norm": 1.25,
      "learning_rate": 0.00016706454951869468,
      "loss": 4.5193,
      "mean_token_accuracy": 0.21345730125904083,
      "num_tokens": 15189577.0,
      "step": 6875
    },
    {
      "entropy": 5.155003881454467,
      "epoch": 6.935955622793747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00016671667283794245,
      "loss": 4.5731,
      "mean_token_accuracy": 0.2113265424966812,
      "num_tokens": 15201897.0,
      "step": 6880
    },
    {
      "entropy": 5.176406192779541,
      "epoch": 6.940998487140696,
      "grad_norm": 1.171875,
      "learning_rate": 0.00016636913270426888,
      "loss": 4.5523,
      "mean_token_accuracy": 0.2038802370429039,
      "num_tokens": 15212266.0,
      "step": 6885
    },
    {
      "entropy": 5.145376443862915,
      "epoch": 6.946041351487645,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016602193019783673,
      "loss": 4.5824,
      "mean_token_accuracy": 0.2078516185283661,
      "num_tokens": 15221911.0,
      "step": 6890
    },
    {
      "entropy": 5.235925102233887,
      "epoch": 6.9510842158345945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016567506639775947,
      "loss": 4.6555,
      "mean_token_accuracy": 0.20694834738969803,
      "num_tokens": 15232210.0,
      "step": 6895
    },
    {
      "entropy": 5.144708156585693,
      "epoch": 6.956127080181544,
      "grad_norm": 1.203125,
      "learning_rate": 0.00016532854238209783,
      "loss": 4.6281,
      "mean_token_accuracy": 0.20101002603769302,
      "num_tokens": 15242872.0,
      "step": 6900
    },
    {
      "entropy": 5.152430772781372,
      "epoch": 6.961169944528492,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001649823592278563,
      "loss": 4.6022,
      "mean_token_accuracy": 0.20615483671426774,
      "num_tokens": 15254901.0,
      "step": 6905
    },
    {
      "entropy": 5.1881012439727785,
      "epoch": 6.966212808875441,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016463651801098023,
      "loss": 4.5642,
      "mean_token_accuracy": 0.20887210518121718,
      "num_tokens": 15265370.0,
      "step": 6910
    },
    {
      "entropy": 5.17701268196106,
      "epoch": 6.97125567322239,
      "grad_norm": 1.328125,
      "learning_rate": 0.00016429101980635193,
      "loss": 4.5451,
      "mean_token_accuracy": 0.2166115403175354,
      "num_tokens": 15275062.0,
      "step": 6915
    },
    {
      "entropy": 5.137642240524292,
      "epoch": 6.976298537569339,
      "grad_norm": 1.25,
      "learning_rate": 0.00016394586568778795,
      "loss": 4.4798,
      "mean_token_accuracy": 0.2090232789516449,
      "num_tokens": 15285463.0,
      "step": 6920
    },
    {
      "entropy": 5.123339223861694,
      "epoch": 6.981341401916288,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016360105672803515,
      "loss": 4.541,
      "mean_token_accuracy": 0.21386461108922958,
      "num_tokens": 15297023.0,
      "step": 6925
    },
    {
      "entropy": 5.1613012790679935,
      "epoch": 6.986384266263237,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016325659399876792,
      "loss": 4.6198,
      "mean_token_accuracy": 0.20218446850776672,
      "num_tokens": 15309136.0,
      "step": 6930
    },
    {
      "entropy": 5.209257936477661,
      "epoch": 6.9914271306101865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016291247857058412,
      "loss": 4.719,
      "mean_token_accuracy": 0.19685543030500413,
      "num_tokens": 15323419.0,
      "step": 6935
    },
    {
      "entropy": 5.195684957504272,
      "epoch": 6.996469994957136,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016256871151300268,
      "loss": 4.6482,
      "mean_token_accuracy": 0.20272963494062424,
      "num_tokens": 15334242.0,
      "step": 6940
    },
    {
      "entropy": 5.191173871358235,
      "epoch": 7.00100857286939,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00016222529389445928,
      "loss": 4.59,
      "mean_token_accuracy": 0.2113943745692571,
      "num_tokens": 15344106.0,
      "step": 6945
    },
    {
      "entropy": 5.118384122848511,
      "epoch": 7.006051437216339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00016188222678230395,
      "loss": 4.3993,
      "mean_token_accuracy": 0.22623356580734252,
      "num_tokens": 15355556.0,
      "step": 6950
    },
    {
      "entropy": 5.126711082458496,
      "epoch": 7.011094301563288,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001615395112427972,
      "loss": 4.434,
      "mean_token_accuracy": 0.21991039514541627,
      "num_tokens": 15365636.0,
      "step": 6955
    },
    {
      "entropy": 5.14275426864624,
      "epoch": 7.016137165910237,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001611971483411069,
      "loss": 4.4498,
      "mean_token_accuracy": 0.21908948570489883,
      "num_tokens": 15376177.0,
      "step": 6960
    },
    {
      "entropy": 5.1958695411682125,
      "epoch": 7.021180030257186,
      "grad_norm": 1.15625,
      "learning_rate": 0.00016085513914130463,
      "loss": 4.4321,
      "mean_token_accuracy": 0.21812065839767455,
      "num_tokens": 15387096.0,
      "step": 6965
    },
    {
      "entropy": 5.188603067398072,
      "epoch": 7.026222894604135,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016051348470636306,
      "loss": 4.4482,
      "mean_token_accuracy": 0.21887545436620712,
      "num_tokens": 15397450.0,
      "step": 6970
    },
    {
      "entropy": 5.198639726638794,
      "epoch": 7.0312657589510845,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016017218609815195,
      "loss": 4.4369,
      "mean_token_accuracy": 0.2192506641149521,
      "num_tokens": 15406958.0,
      "step": 6975
    },
    {
      "entropy": 5.1605198860168455,
      "epoch": 7.036308623298034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001598312443774353,
      "loss": 4.5016,
      "mean_token_accuracy": 0.2176968514919281,
      "num_tokens": 15417644.0,
      "step": 6980
    },
    {
      "entropy": 5.157814073562622,
      "epoch": 7.041351487644983,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015949066060386784,
      "loss": 4.3838,
      "mean_token_accuracy": 0.22596095651388168,
      "num_tokens": 15428561.0,
      "step": 6985
    },
    {
      "entropy": 5.129917573928833,
      "epoch": 7.046394351991931,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00015915043583599183,
      "loss": 4.3916,
      "mean_token_accuracy": 0.2264675334095955,
      "num_tokens": 15437729.0,
      "step": 6990
    },
    {
      "entropy": 5.153252506256104,
      "epoch": 7.05143721633888,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001588105711312336,
      "loss": 4.5012,
      "mean_token_accuracy": 0.22351074814796448,
      "num_tokens": 15449155.0,
      "step": 6995
    },
    {
      "entropy": 5.128206872940064,
      "epoch": 7.056480080685829,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015847106754590055,
      "loss": 4.4222,
      "mean_token_accuracy": 0.22596587538719176,
      "num_tokens": 15460502.0,
      "step": 7000
    },
    {
      "epoch": 7.056480080685829,
      "eval_entropy": 5.1260758950813425,
      "eval_loss": 5.692974090576172,
      "eval_mean_token_accuracy": 0.15377418531943823,
      "eval_num_tokens": 15460502.0,
      "eval_runtime": 2.6311,
      "eval_samples_per_second": 1192.644,
      "eval_steps_per_second": 149.366,
      "step": 7000
    },
    {
      "entropy": 5.154039812088013,
      "epoch": 7.061522945032778,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001581319261351777,
      "loss": 4.438,
      "mean_token_accuracy": 0.22144186943769456,
      "num_tokens": 15472035.0,
      "step": 7005
    },
    {
      "entropy": 5.143575811386109,
      "epoch": 7.0665658093797274,
      "grad_norm": 1.234375,
      "learning_rate": 0.00015779314795312426,
      "loss": 4.3715,
      "mean_token_accuracy": 0.2265063241124153,
      "num_tokens": 15482567.0,
      "step": 7010
    },
    {
      "entropy": 5.171265363693237,
      "epoch": 7.071608673726677,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001574547340526707,
      "loss": 4.5001,
      "mean_token_accuracy": 0.21982353925704956,
      "num_tokens": 15493785.0,
      "step": 7015
    },
    {
      "entropy": 5.175794553756714,
      "epoch": 7.076651538073626,
      "grad_norm": 1.21875,
      "learning_rate": 0.00015711668548561527,
      "loss": 4.4172,
      "mean_token_accuracy": 0.22556657642126082,
      "num_tokens": 15505321.0,
      "step": 7020
    },
    {
      "entropy": 5.15221757888794,
      "epoch": 7.081694402420575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00015677900330262075,
      "loss": 4.4676,
      "mean_token_accuracy": 0.22193102687597274,
      "num_tokens": 15517188.0,
      "step": 7025
    },
    {
      "entropy": 5.171244668960571,
      "epoch": 7.086737266767524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015644168855321098,
      "loss": 4.4828,
      "mean_token_accuracy": 0.22104574590921403,
      "num_tokens": 15528496.0,
      "step": 7030
    },
    {
      "entropy": 5.19063720703125,
      "epoch": 7.091780131114473,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00015610474228576818,
      "loss": 4.451,
      "mean_token_accuracy": 0.21990178674459457,
      "num_tokens": 15540717.0,
      "step": 7035
    },
    {
      "entropy": 5.166711854934692,
      "epoch": 7.096822995461422,
      "grad_norm": 1.140625,
      "learning_rate": 0.00015576816554752892,
      "loss": 4.3575,
      "mean_token_accuracy": 0.23132373690605162,
      "num_tokens": 15552878.0,
      "step": 7040
    },
    {
      "entropy": 5.2299281597137455,
      "epoch": 7.101865859808371,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001554319593845816,
      "loss": 4.4799,
      "mean_token_accuracy": 0.21541013121604918,
      "num_tokens": 15564285.0,
      "step": 7045
    },
    {
      "entropy": 5.182926845550537,
      "epoch": 7.10690872415532,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001550961248418627,
      "loss": 4.4681,
      "mean_token_accuracy": 0.22308433353900908,
      "num_tokens": 15574802.0,
      "step": 7050
    },
    {
      "entropy": 5.088316631317139,
      "epoch": 7.1119515885022695,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015476066296315376,
      "loss": 4.3952,
      "mean_token_accuracy": 0.2274264082312584,
      "num_tokens": 15585188.0,
      "step": 7055
    },
    {
      "entropy": 5.118604230880737,
      "epoch": 7.116994452849219,
      "grad_norm": 1.15625,
      "learning_rate": 0.000154425574791078,
      "loss": 4.4566,
      "mean_token_accuracy": 0.22074566632509232,
      "num_tokens": 15597043.0,
      "step": 7060
    },
    {
      "entropy": 5.1144310474395756,
      "epoch": 7.122037317196168,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00015409086136709723,
      "loss": 4.3862,
      "mean_token_accuracy": 0.22935310751199722,
      "num_tokens": 15608292.0,
      "step": 7065
    },
    {
      "entropy": 5.164602756500244,
      "epoch": 7.127080181543117,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001537565237315084,
      "loss": 4.4068,
      "mean_token_accuracy": 0.22729408442974092,
      "num_tokens": 15617898.0,
      "step": 7070
    },
    {
      "entropy": 5.207625865936279,
      "epoch": 7.132123045890066,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015342256292344065,
      "loss": 4.4727,
      "mean_token_accuracy": 0.2152162551879883,
      "num_tokens": 15630047.0,
      "step": 7075
    },
    {
      "entropy": 5.198468351364136,
      "epoch": 7.137165910237014,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015308897998085187,
      "loss": 4.4039,
      "mean_token_accuracy": 0.2323444202542305,
      "num_tokens": 15640326.0,
      "step": 7080
    },
    {
      "entropy": 5.17194709777832,
      "epoch": 7.142208774583963,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015275577594052564,
      "loss": 4.4015,
      "mean_token_accuracy": 0.22807519882917404,
      "num_tokens": 15650415.0,
      "step": 7085
    },
    {
      "entropy": 5.122431850433349,
      "epoch": 7.147251638930912,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001524229518380676,
      "loss": 4.4228,
      "mean_token_accuracy": 0.22701407521963118,
      "num_tokens": 15659725.0,
      "step": 7090
    },
    {
      "entropy": 5.171591424942017,
      "epoch": 7.1522945032778615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001520905087079029,
      "loss": 4.4696,
      "mean_token_accuracy": 0.22302038520574569,
      "num_tokens": 15670601.0,
      "step": 7095
    },
    {
      "entropy": 5.112852668762207,
      "epoch": 7.157337367624811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00015175844758327239,
      "loss": 4.3975,
      "mean_token_accuracy": 0.23161530792713164,
      "num_tokens": 15683118.0,
      "step": 7100
    },
    {
      "entropy": 5.131816720962524,
      "epoch": 7.16238023197176,
      "grad_norm": 1.25,
      "learning_rate": 0.00015142676949622973,
      "loss": 4.4812,
      "mean_token_accuracy": 0.22077066004276275,
      "num_tokens": 15694384.0,
      "step": 7105
    },
    {
      "entropy": 5.101582002639771,
      "epoch": 7.167423096318709,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015109547547763802,
      "loss": 4.4734,
      "mean_token_accuracy": 0.21831965446472168,
      "num_tokens": 15706767.0,
      "step": 7110
    },
    {
      "entropy": 5.152289819717407,
      "epoch": 7.172465960665658,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015076456655716675,
      "loss": 4.4409,
      "mean_token_accuracy": 0.2213535025715828,
      "num_tokens": 15718409.0,
      "step": 7115
    },
    {
      "entropy": 5.1540703773498535,
      "epoch": 7.177508825012607,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015043404376328856,
      "loss": 4.4315,
      "mean_token_accuracy": 0.2257857531309128,
      "num_tokens": 15728531.0,
      "step": 7120
    },
    {
      "entropy": 5.166474008560181,
      "epoch": 7.182551689359556,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015010390812327578,
      "loss": 4.472,
      "mean_token_accuracy": 0.22256550043821335,
      "num_tokens": 15739572.0,
      "step": 7125
    },
    {
      "entropy": 5.160050868988037,
      "epoch": 7.187594553706505,
      "grad_norm": 1.375,
      "learning_rate": 0.0001497741606631977,
      "loss": 4.4579,
      "mean_token_accuracy": 0.21896929889917374,
      "num_tokens": 15749046.0,
      "step": 7130
    },
    {
      "entropy": 5.122409534454346,
      "epoch": 7.1926374180534545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001494448024079169,
      "loss": 4.4528,
      "mean_token_accuracy": 0.22109250128269195,
      "num_tokens": 15760445.0,
      "step": 7135
    },
    {
      "entropy": 5.175435590744018,
      "epoch": 7.197680282400404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014911583438108656,
      "loss": 4.4409,
      "mean_token_accuracy": 0.22281454503536224,
      "num_tokens": 15771775.0,
      "step": 7140
    },
    {
      "entropy": 5.132323312759399,
      "epoch": 7.202723146747353,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014878725760514682,
      "loss": 4.4372,
      "mean_token_accuracy": 0.21821482479572296,
      "num_tokens": 15781369.0,
      "step": 7145
    },
    {
      "entropy": 5.136046838760376,
      "epoch": 7.207766011094302,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014845907310132202,
      "loss": 4.4831,
      "mean_token_accuracy": 0.22071818113327027,
      "num_tokens": 15791931.0,
      "step": 7150
    },
    {
      "entropy": 5.207011270523071,
      "epoch": 7.212808875441251,
      "grad_norm": 1.1875,
      "learning_rate": 0.00014813128188961695,
      "loss": 4.4347,
      "mean_token_accuracy": 0.223936328291893,
      "num_tokens": 15803673.0,
      "step": 7155
    },
    {
      "entropy": 5.190221166610717,
      "epoch": 7.2178517397882,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014780388498881447,
      "loss": 4.4921,
      "mean_token_accuracy": 0.21330800503492356,
      "num_tokens": 15815122.0,
      "step": 7160
    },
    {
      "entropy": 5.175816774368286,
      "epoch": 7.222894604135149,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001474768834164715,
      "loss": 4.4761,
      "mean_token_accuracy": 0.22186025679111482,
      "num_tokens": 15825714.0,
      "step": 7165
    },
    {
      "entropy": 5.1677710056304935,
      "epoch": 7.227937468482098,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014715027818891656,
      "loss": 4.3526,
      "mean_token_accuracy": 0.23623055666685105,
      "num_tokens": 15836803.0,
      "step": 7170
    },
    {
      "entropy": 5.16889705657959,
      "epoch": 7.2329803328290465,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001468240703212463,
      "loss": 4.4729,
      "mean_token_accuracy": 0.21961220502853393,
      "num_tokens": 15847529.0,
      "step": 7175
    },
    {
      "entropy": 5.205008840560913,
      "epoch": 7.238023197175996,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014649826082732226,
      "loss": 4.5142,
      "mean_token_accuracy": 0.21354184448719024,
      "num_tokens": 15857172.0,
      "step": 7180
    },
    {
      "entropy": 5.128279113769532,
      "epoch": 7.243066061522945,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00014617285071976777,
      "loss": 4.4485,
      "mean_token_accuracy": 0.22419991791248323,
      "num_tokens": 15867798.0,
      "step": 7185
    },
    {
      "entropy": 5.130161094665527,
      "epoch": 7.248108925869894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00014584784100996512,
      "loss": 4.4842,
      "mean_token_accuracy": 0.21619490534067154,
      "num_tokens": 15879429.0,
      "step": 7190
    },
    {
      "entropy": 5.171583032608032,
      "epoch": 7.253151790216843,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001455232327080518,
      "loss": 4.4328,
      "mean_token_accuracy": 0.2223067492246628,
      "num_tokens": 15890045.0,
      "step": 7195
    },
    {
      "entropy": 5.231035852432251,
      "epoch": 7.258194654563792,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014519902682291795,
      "loss": 4.5349,
      "mean_token_accuracy": 0.21420277804136276,
      "num_tokens": 15901044.0,
      "step": 7200
    },
    {
      "entropy": 5.192950248718262,
      "epoch": 7.263237518910741,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00014487522436220304,
      "loss": 4.4586,
      "mean_token_accuracy": 0.2219088077545166,
      "num_tokens": 15912260.0,
      "step": 7205
    },
    {
      "entropy": 5.106403970718384,
      "epoch": 7.26828038325769,
      "grad_norm": 1.28125,
      "learning_rate": 0.00014455182633229258,
      "loss": 4.4144,
      "mean_token_accuracy": 0.23532620817422867,
      "num_tokens": 15923305.0,
      "step": 7210
    },
    {
      "entropy": 5.171741533279419,
      "epoch": 7.273323247604639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00014422883373831497,
      "loss": 4.4742,
      "mean_token_accuracy": 0.2140638366341591,
      "num_tokens": 15933878.0,
      "step": 7215
    },
    {
      "entropy": 5.180574083328247,
      "epoch": 7.278366111951589,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014390624758413872,
      "loss": 4.4826,
      "mean_token_accuracy": 0.21907410621643067,
      "num_tokens": 15944849.0,
      "step": 7220
    },
    {
      "entropy": 5.129748296737671,
      "epoch": 7.283408976298538,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00014358406887236914,
      "loss": 4.4847,
      "mean_token_accuracy": 0.2200422018766403,
      "num_tokens": 15956297.0,
      "step": 7225
    },
    {
      "entropy": 5.151044464111328,
      "epoch": 7.288451840645487,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014326229860434488,
      "loss": 4.4505,
      "mean_token_accuracy": 0.2238429605960846,
      "num_tokens": 15967455.0,
      "step": 7230
    },
    {
      "entropy": 5.184419631958008,
      "epoch": 7.293494704992436,
      "grad_norm": 1.125,
      "learning_rate": 0.0001429409377801355,
      "loss": 4.5445,
      "mean_token_accuracy": 0.21651486307382584,
      "num_tokens": 15978959.0,
      "step": 7235
    },
    {
      "entropy": 5.159758567810059,
      "epoch": 7.298537569339385,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014261998739853785,
      "loss": 4.4303,
      "mean_token_accuracy": 0.22529436498880387,
      "num_tokens": 15990509.0,
      "step": 7240
    },
    {
      "entropy": 5.1821935176849365,
      "epoch": 7.303580433686334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00014229944845707315,
      "loss": 4.4858,
      "mean_token_accuracy": 0.21260163187980652,
      "num_tokens": 16001680.0,
      "step": 7245
    },
    {
      "entropy": 5.207985830307007,
      "epoch": 7.308623298033283,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014197932195198374,
      "loss": 4.5309,
      "mean_token_accuracy": 0.22141532152891158,
      "num_tokens": 16016121.0,
      "step": 7250
    },
    {
      "entropy": 5.230870389938355,
      "epoch": 7.313666162380232,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014165960887823026,
      "loss": 4.4612,
      "mean_token_accuracy": 0.22050490528345107,
      "num_tokens": 16026640.0,
      "step": 7255
    },
    {
      "entropy": 5.153852653503418,
      "epoch": 7.318709026727181,
      "grad_norm": 1.21875,
      "learning_rate": 0.00014134031022948817,
      "loss": 4.4394,
      "mean_token_accuracy": 0.22102881073951722,
      "num_tokens": 16037298.0,
      "step": 7260
    },
    {
      "entropy": 5.158794593811035,
      "epoch": 7.32375189107413,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014102142699814525,
      "loss": 4.4273,
      "mean_token_accuracy": 0.22627291828393936,
      "num_tokens": 16048897.0,
      "step": 7265
    },
    {
      "entropy": 5.173964357376098,
      "epoch": 7.328794755421079,
      "grad_norm": 1.171875,
      "learning_rate": 0.00014070296017529773,
      "loss": 4.5055,
      "mean_token_accuracy": 0.21460242718458175,
      "num_tokens": 16061015.0,
      "step": 7270
    },
    {
      "entropy": 5.1031571388244625,
      "epoch": 7.333837619768028,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014038491075074805,
      "loss": 4.4262,
      "mean_token_accuracy": 0.22371019124984742,
      "num_tokens": 16071550.0,
      "step": 7275
    },
    {
      "entropy": 5.122872591018677,
      "epoch": 7.338880484114977,
      "grad_norm": 1.203125,
      "learning_rate": 0.00014006727971300098,
      "loss": 4.4357,
      "mean_token_accuracy": 0.2270269811153412,
      "num_tokens": 16082956.0,
      "step": 7280
    },
    {
      "entropy": 5.158475923538208,
      "epoch": 7.343923348461926,
      "grad_norm": 1.25,
      "learning_rate": 0.00013975006804926128,
      "loss": 4.436,
      "mean_token_accuracy": 0.22386692762374877,
      "num_tokens": 16093985.0,
      "step": 7285
    },
    {
      "entropy": 5.193248271942139,
      "epoch": 7.348966212808875,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00013943327674542988,
      "loss": 4.5557,
      "mean_token_accuracy": 0.21329813599586486,
      "num_tokens": 16106646.0,
      "step": 7290
    },
    {
      "entropy": 5.189702272415161,
      "epoch": 7.354009077155824,
      "grad_norm": 1.25,
      "learning_rate": 0.00013911690678610164,
      "loss": 4.5104,
      "mean_token_accuracy": 0.21881428956985474,
      "num_tokens": 16117525.0,
      "step": 7295
    },
    {
      "entropy": 5.171935415267944,
      "epoch": 7.3590519415027735,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013880095915456167,
      "loss": 4.532,
      "mean_token_accuracy": 0.20862729996442794,
      "num_tokens": 16128744.0,
      "step": 7300
    },
    {
      "entropy": 5.121890068054199,
      "epoch": 7.364094805849723,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013848543483278256,
      "loss": 4.4475,
      "mean_token_accuracy": 0.22047294527292252,
      "num_tokens": 16140126.0,
      "step": 7305
    },
    {
      "entropy": 5.159297513961792,
      "epoch": 7.369137670196672,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000138170334801421,
      "loss": 4.5044,
      "mean_token_accuracy": 0.2239419013261795,
      "num_tokens": 16152684.0,
      "step": 7310
    },
    {
      "entropy": 5.150891256332398,
      "epoch": 7.374180534543621,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00013785566003981538,
      "loss": 4.4735,
      "mean_token_accuracy": 0.22464701533317566,
      "num_tokens": 16164396.0,
      "step": 7315
    },
    {
      "entropy": 5.192217540740967,
      "epoch": 7.37922339889057,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001375414115259819,
      "loss": 4.5352,
      "mean_token_accuracy": 0.2187423139810562,
      "num_tokens": 16175857.0,
      "step": 7320
    },
    {
      "entropy": 5.178657865524292,
      "epoch": 7.384266263237519,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001372275902366122,
      "loss": 4.4092,
      "mean_token_accuracy": 0.2222050741314888,
      "num_tokens": 16188607.0,
      "step": 7325
    },
    {
      "entropy": 5.163939380645752,
      "epoch": 7.389309127584468,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013691419714707026,
      "loss": 4.4283,
      "mean_token_accuracy": 0.22540317177772523,
      "num_tokens": 16199513.0,
      "step": 7330
    },
    {
      "entropy": 5.141021013259888,
      "epoch": 7.394351991931417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00013660123323138893,
      "loss": 4.4647,
      "mean_token_accuracy": 0.21500949561595917,
      "num_tokens": 16209934.0,
      "step": 7335
    },
    {
      "entropy": 5.164689207077027,
      "epoch": 7.3993948562783665,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013628869946226738,
      "loss": 4.4461,
      "mean_token_accuracy": 0.22271998673677446,
      "num_tokens": 16219722.0,
      "step": 7340
    },
    {
      "entropy": 5.206972360610962,
      "epoch": 7.404437720625316,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001359765968110676,
      "loss": 4.5189,
      "mean_token_accuracy": 0.22015084475278854,
      "num_tokens": 16230414.0,
      "step": 7345
    },
    {
      "entropy": 5.184099769592285,
      "epoch": 7.409480584972265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013566492624781198,
      "loss": 4.4847,
      "mean_token_accuracy": 0.2165430262684822,
      "num_tokens": 16241563.0,
      "step": 7350
    },
    {
      "entropy": 5.057126712799072,
      "epoch": 7.414523449319213,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001353536887411797,
      "loss": 4.3118,
      "mean_token_accuracy": 0.22799203544855118,
      "num_tokens": 16253012.0,
      "step": 7355
    },
    {
      "entropy": 5.182149648666382,
      "epoch": 7.419566313666162,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013504288525850416,
      "loss": 4.4406,
      "mean_token_accuracy": 0.22110387235879897,
      "num_tokens": 16263642.0,
      "step": 7360
    },
    {
      "entropy": 5.132106924057007,
      "epoch": 7.424609178013111,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013473251676576975,
      "loss": 4.4392,
      "mean_token_accuracy": 0.21796586960554123,
      "num_tokens": 16274270.0,
      "step": 7365
    },
    {
      "entropy": 5.149462556838989,
      "epoch": 7.42965204236006,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000134422584227609,
      "loss": 4.5284,
      "mean_token_accuracy": 0.2227826938033104,
      "num_tokens": 16285894.0,
      "step": 7370
    },
    {
      "entropy": 5.170673274993897,
      "epoch": 7.434694906707009,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001341130886072992,
      "loss": 4.5348,
      "mean_token_accuracy": 0.21409969329833983,
      "num_tokens": 16298822.0,
      "step": 7375
    },
    {
      "entropy": 5.122301864624023,
      "epoch": 7.4397377710539585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013380403086676014,
      "loss": 4.4418,
      "mean_token_accuracy": 0.21916188299655914,
      "num_tokens": 16310410.0,
      "step": 7380
    },
    {
      "entropy": 5.145479965209961,
      "epoch": 7.444780635400908,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013349541196655006,
      "loss": 4.4965,
      "mean_token_accuracy": 0.2199232444167137,
      "num_tokens": 16320474.0,
      "step": 7385
    },
    {
      "entropy": 5.248195552825928,
      "epoch": 7.449823499747857,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00013318723286586388,
      "loss": 4.5115,
      "mean_token_accuracy": 0.2200802594423294,
      "num_tokens": 16331354.0,
      "step": 7390
    },
    {
      "entropy": 5.161272430419922,
      "epoch": 7.454866364094806,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001328794945225293,
      "loss": 4.4831,
      "mean_token_accuracy": 0.22407786101102828,
      "num_tokens": 16342599.0,
      "step": 7395
    },
    {
      "entropy": 5.193317127227783,
      "epoch": 7.459909228441755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00013257219789300417,
      "loss": 4.4994,
      "mean_token_accuracy": 0.2260016530752182,
      "num_tokens": 16353556.0,
      "step": 7400
    },
    {
      "entropy": 5.18008975982666,
      "epoch": 7.464952092788704,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001322653439323735,
      "loss": 4.4541,
      "mean_token_accuracy": 0.22499921172857285,
      "num_tokens": 16365633.0,
      "step": 7405
    },
    {
      "entropy": 5.1750298023223875,
      "epoch": 7.469994957135653,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013195893359434658,
      "loss": 4.3897,
      "mean_token_accuracy": 0.2277309462428093,
      "num_tokens": 16377401.0,
      "step": 7410
    },
    {
      "entropy": 5.1744671821594235,
      "epoch": 7.475037821482602,
      "grad_norm": 1.203125,
      "learning_rate": 0.00013165296783125357,
      "loss": 4.4239,
      "mean_token_accuracy": 0.22332353591918946,
      "num_tokens": 16389018.0,
      "step": 7415
    },
    {
      "entropy": 5.175772953033447,
      "epoch": 7.480080685829551,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00013134744759404332,
      "loss": 4.5092,
      "mean_token_accuracy": 0.2188422128558159,
      "num_tokens": 16400467.0,
      "step": 7420
    },
    {
      "entropy": 5.178360462188721,
      "epoch": 7.4851235501765006,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001310423738322797,
      "loss": 4.4842,
      "mean_token_accuracy": 0.2137891545891762,
      "num_tokens": 16411114.0,
      "step": 7425
    },
    {
      "entropy": 5.178935956954956,
      "epoch": 7.49016641452345,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013073774749413903,
      "loss": 4.4635,
      "mean_token_accuracy": 0.2264242798089981,
      "num_tokens": 16421762.0,
      "step": 7430
    },
    {
      "entropy": 5.163376760482788,
      "epoch": 7.495209278870399,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00013043356952640697,
      "loss": 4.4892,
      "mean_token_accuracy": 0.21581490337848663,
      "num_tokens": 16432700.0,
      "step": 7435
    },
    {
      "entropy": 5.094449806213379,
      "epoch": 7.500252143217347,
      "grad_norm": 1.265625,
      "learning_rate": 0.00013012984087447567,
      "loss": 4.4738,
      "mean_token_accuracy": 0.21627049446105956,
      "num_tokens": 16443256.0,
      "step": 7440
    },
    {
      "entropy": 5.148663902282715,
      "epoch": 7.505295007564296,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00012982656248234088,
      "loss": 4.4778,
      "mean_token_accuracy": 0.22331043928861619,
      "num_tokens": 16454078.0,
      "step": 7445
    },
    {
      "entropy": 5.179423809051514,
      "epoch": 7.510337871911245,
      "grad_norm": 1.25,
      "learning_rate": 0.00012952373529259863,
      "loss": 4.4614,
      "mean_token_accuracy": 0.22324355840682983,
      "num_tokens": 16464322.0,
      "step": 7450
    },
    {
      "entropy": 5.194389915466308,
      "epoch": 7.515380736258194,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012922136024644298,
      "loss": 4.5494,
      "mean_token_accuracy": 0.21684356778860092,
      "num_tokens": 16475326.0,
      "step": 7455
    },
    {
      "entropy": 5.243834590911865,
      "epoch": 7.5204236006051435,
      "grad_norm": 1.296875,
      "learning_rate": 0.00012891943828366247,
      "loss": 4.543,
      "mean_token_accuracy": 0.22003503739833832,
      "num_tokens": 16486406.0,
      "step": 7460
    },
    {
      "entropy": 5.2015456199646,
      "epoch": 7.525466464952093,
      "grad_norm": 1.171875,
      "learning_rate": 0.00012861797034263763,
      "loss": 4.5044,
      "mean_token_accuracy": 0.21789226680994034,
      "num_tokens": 16498299.0,
      "step": 7465
    },
    {
      "entropy": 5.119556951522827,
      "epoch": 7.530509329299042,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001283169573603376,
      "loss": 4.4683,
      "mean_token_accuracy": 0.22565991282463074,
      "num_tokens": 16508401.0,
      "step": 7470
    },
    {
      "entropy": 5.202124309539795,
      "epoch": 7.535552193645991,
      "grad_norm": 1.1875,
      "learning_rate": 0.00012801640027231792,
      "loss": 4.4785,
      "mean_token_accuracy": 0.21772000193595886,
      "num_tokens": 16519075.0,
      "step": 7475
    },
    {
      "entropy": 5.124085474014282,
      "epoch": 7.54059505799294,
      "grad_norm": 1.203125,
      "learning_rate": 0.00012771630001271668,
      "loss": 4.494,
      "mean_token_accuracy": 0.2212168887257576,
      "num_tokens": 16531024.0,
      "step": 7480
    },
    {
      "entropy": 5.14650673866272,
      "epoch": 7.545637922339889,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012741665751425253,
      "loss": 4.4744,
      "mean_token_accuracy": 0.21856363117694855,
      "num_tokens": 16542167.0,
      "step": 7485
    },
    {
      "entropy": 5.198973703384399,
      "epoch": 7.550680786686838,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012711747370822127,
      "loss": 4.5509,
      "mean_token_accuracy": 0.2134639248251915,
      "num_tokens": 16554421.0,
      "step": 7490
    },
    {
      "entropy": 5.185412502288818,
      "epoch": 7.555723651033787,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012681874952449315,
      "loss": 4.4739,
      "mean_token_accuracy": 0.2211108162999153,
      "num_tokens": 16564614.0,
      "step": 7495
    },
    {
      "entropy": 5.143761110305786,
      "epoch": 7.560766515380736,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001265204858915097,
      "loss": 4.4028,
      "mean_token_accuracy": 0.23240901082754134,
      "num_tokens": 16575347.0,
      "step": 7500
    },
    {
      "epoch": 7.560766515380736,
      "eval_entropy": 5.1159592572666,
      "eval_loss": 5.695576190948486,
      "eval_mean_token_accuracy": 0.15420451354586137,
      "eval_num_tokens": 16575347.0,
      "eval_runtime": 2.6408,
      "eval_samples_per_second": 1188.266,
      "eval_steps_per_second": 148.817,
      "step": 7500
    },
    {
      "entropy": 5.192011308670044,
      "epoch": 7.5658093797276855,
      "grad_norm": 1.25,
      "learning_rate": 0.00012622268373628127,
      "loss": 4.4676,
      "mean_token_accuracy": 0.21707640141248702,
      "num_tokens": 16585104.0,
      "step": 7505
    },
    {
      "entropy": 5.145976305007935,
      "epoch": 7.570852244074635,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00012592534398438369,
      "loss": 4.4701,
      "mean_token_accuracy": 0.21700835227966309,
      "num_tokens": 16595659.0,
      "step": 7510
    },
    {
      "entropy": 5.143455457687378,
      "epoch": 7.575895108421584,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012562846755995586,
      "loss": 4.4193,
      "mean_token_accuracy": 0.21693531274795533,
      "num_tokens": 16605473.0,
      "step": 7515
    },
    {
      "entropy": 5.170410013198852,
      "epoch": 7.580937972768533,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012533205538569654,
      "loss": 4.4961,
      "mean_token_accuracy": 0.21224465072155,
      "num_tokens": 16617723.0,
      "step": 7520
    },
    {
      "entropy": 5.1171355724334715,
      "epoch": 7.585980837115482,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001250361083828617,
      "loss": 4.4419,
      "mean_token_accuracy": 0.21874863058328628,
      "num_tokens": 16628166.0,
      "step": 7525
    },
    {
      "entropy": 5.104258966445923,
      "epoch": 7.591023701462431,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001247406274712613,
      "loss": 4.462,
      "mean_token_accuracy": 0.2164518028497696,
      "num_tokens": 16641261.0,
      "step": 7530
    },
    {
      "entropy": 5.13037428855896,
      "epoch": 7.59606656580938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00012444561356925697,
      "loss": 4.3703,
      "mean_token_accuracy": 0.2370496928691864,
      "num_tokens": 16651991.0,
      "step": 7535
    },
    {
      "entropy": 5.1764068603515625,
      "epoch": 7.6011094301563285,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012415106759375855,
      "loss": 4.4772,
      "mean_token_accuracy": 0.21864103078842162,
      "num_tokens": 16663727.0,
      "step": 7540
    },
    {
      "entropy": 5.179099416732788,
      "epoch": 7.606152294503278,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00012385699046022188,
      "loss": 4.5061,
      "mean_token_accuracy": 0.2169162377715111,
      "num_tokens": 16675594.0,
      "step": 7545
    },
    {
      "entropy": 5.129764175415039,
      "epoch": 7.611195158850227,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012356338308264537,
      "loss": 4.3977,
      "mean_token_accuracy": 0.2263380393385887,
      "num_tokens": 16685027.0,
      "step": 7550
    },
    {
      "entropy": 5.169068431854248,
      "epoch": 7.616238023197176,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001232702463735676,
      "loss": 4.4386,
      "mean_token_accuracy": 0.21943617165088652,
      "num_tokens": 16695017.0,
      "step": 7555
    },
    {
      "entropy": 5.158508729934693,
      "epoch": 7.621280887544125,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012297758124406427,
      "loss": 4.452,
      "mean_token_accuracy": 0.22499199211597443,
      "num_tokens": 16705033.0,
      "step": 7560
    },
    {
      "entropy": 5.122021770477295,
      "epoch": 7.626323751891074,
      "grad_norm": 1.25,
      "learning_rate": 0.00012268538860374517,
      "loss": 4.4856,
      "mean_token_accuracy": 0.22727629244327546,
      "num_tokens": 16715743.0,
      "step": 7565
    },
    {
      "entropy": 5.173369693756103,
      "epoch": 7.631366616238023,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012239366936075196,
      "loss": 4.5226,
      "mean_token_accuracy": 0.212329763174057,
      "num_tokens": 16726719.0,
      "step": 7570
    },
    {
      "entropy": 5.1458171844482425,
      "epoch": 7.636409480584972,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00012210242442175462,
      "loss": 4.4225,
      "mean_token_accuracy": 0.23534375727176665,
      "num_tokens": 16737221.0,
      "step": 7575
    },
    {
      "entropy": 5.137267303466797,
      "epoch": 7.641452344931921,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012181165469194924,
      "loss": 4.48,
      "mean_token_accuracy": 0.21692835688591003,
      "num_tokens": 16747871.0,
      "step": 7580
    },
    {
      "entropy": 5.197782945632935,
      "epoch": 7.6464952092788705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001215213610750548,
      "loss": 4.5645,
      "mean_token_accuracy": 0.20769049525260924,
      "num_tokens": 16760249.0,
      "step": 7585
    },
    {
      "entropy": 5.185040807723999,
      "epoch": 7.65153807362582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012123154447331064,
      "loss": 4.5068,
      "mean_token_accuracy": 0.2112462341785431,
      "num_tokens": 16770800.0,
      "step": 7590
    },
    {
      "entropy": 5.183054780960083,
      "epoch": 7.656580937972769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00012094220578747332,
      "loss": 4.4755,
      "mean_token_accuracy": 0.2154254510998726,
      "num_tokens": 16781586.0,
      "step": 7595
    },
    {
      "entropy": 5.2510663032531735,
      "epoch": 7.661623802319718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001206533459168143,
      "loss": 4.5562,
      "mean_token_accuracy": 0.2146369382739067,
      "num_tokens": 16792903.0,
      "step": 7600
    },
    {
      "entropy": 5.168661594390869,
      "epoch": 7.666666666666667,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00012036496575911656,
      "loss": 4.5257,
      "mean_token_accuracy": 0.2144796833395958,
      "num_tokens": 16805118.0,
      "step": 7605
    },
    {
      "entropy": 5.150777339935303,
      "epoch": 7.671709531013616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00012007706621067232,
      "loss": 4.4949,
      "mean_token_accuracy": 0.21809468418359756,
      "num_tokens": 16815550.0,
      "step": 7610
    },
    {
      "entropy": 5.174894523620606,
      "epoch": 7.676752395360564,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011978964816628008,
      "loss": 4.4411,
      "mean_token_accuracy": 0.2191233828663826,
      "num_tokens": 16827012.0,
      "step": 7615
    },
    {
      "entropy": 5.152797222137451,
      "epoch": 7.681795259707513,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011950271251924178,
      "loss": 4.4278,
      "mean_token_accuracy": 0.22323757261037827,
      "num_tokens": 16837543.0,
      "step": 7620
    },
    {
      "entropy": 5.129453039169311,
      "epoch": 7.6868381240544625,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011921626016135991,
      "loss": 4.3924,
      "mean_token_accuracy": 0.227545328438282,
      "num_tokens": 16848691.0,
      "step": 7625
    },
    {
      "entropy": 5.137750434875488,
      "epoch": 7.691880988401412,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00011893029198293512,
      "loss": 4.4203,
      "mean_token_accuracy": 0.22665021568536758,
      "num_tokens": 16859909.0,
      "step": 7630
    },
    {
      "entropy": 5.132231283187866,
      "epoch": 7.696923852748361,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011864480887276301,
      "loss": 4.4628,
      "mean_token_accuracy": 0.21817567348480224,
      "num_tokens": 16869564.0,
      "step": 7635
    },
    {
      "entropy": 5.144774675369263,
      "epoch": 7.70196671709531,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001183598117181317,
      "loss": 4.4866,
      "mean_token_accuracy": 0.22621842175722123,
      "num_tokens": 16879362.0,
      "step": 7640
    },
    {
      "entropy": 5.1308088302612305,
      "epoch": 7.707009581442259,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011807530140481896,
      "loss": 4.4445,
      "mean_token_accuracy": 0.22603902071714402,
      "num_tokens": 16889360.0,
      "step": 7645
    },
    {
      "entropy": 5.140509748458863,
      "epoch": 7.712052445789208,
      "grad_norm": 1.109375,
      "learning_rate": 0.00011779127881708946,
      "loss": 4.4705,
      "mean_token_accuracy": 0.21892292350530623,
      "num_tokens": 16902018.0,
      "step": 7650
    },
    {
      "entropy": 5.196710395812988,
      "epoch": 7.717095310136157,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001175077448376918,
      "loss": 4.5371,
      "mean_token_accuracy": 0.20978907495737076,
      "num_tokens": 16913473.0,
      "step": 7655
    },
    {
      "entropy": 5.087620258331299,
      "epoch": 7.722138174483106,
      "grad_norm": 1.1875,
      "learning_rate": 0.0001172247003478562,
      "loss": 4.4096,
      "mean_token_accuracy": 0.22685015052556992,
      "num_tokens": 16924974.0,
      "step": 7660
    },
    {
      "entropy": 5.120499610900879,
      "epoch": 7.7271810388300555,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011694214622729157,
      "loss": 4.4378,
      "mean_token_accuracy": 0.2218290939927101,
      "num_tokens": 16934878.0,
      "step": 7665
    },
    {
      "entropy": 5.133027458190918,
      "epoch": 7.732223903177005,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011666008335418241,
      "loss": 4.4904,
      "mean_token_accuracy": 0.21519976258277893,
      "num_tokens": 16946142.0,
      "step": 7670
    },
    {
      "entropy": 5.156450843811035,
      "epoch": 7.737266767523954,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001163785126051868,
      "loss": 4.4631,
      "mean_token_accuracy": 0.22532116025686263,
      "num_tokens": 16956312.0,
      "step": 7675
    },
    {
      "entropy": 5.088868951797485,
      "epoch": 7.742309631870903,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001160974348554331,
      "loss": 4.388,
      "mean_token_accuracy": 0.22821759432554245,
      "num_tokens": 16967008.0,
      "step": 7680
    },
    {
      "entropy": 5.132564544677734,
      "epoch": 7.747352496217852,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011581685097851747,
      "loss": 4.464,
      "mean_token_accuracy": 0.22269868552684785,
      "num_tokens": 16978136.0,
      "step": 7685
    },
    {
      "entropy": 5.162061405181885,
      "epoch": 7.752395360564801,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011553676184650096,
      "loss": 4.4272,
      "mean_token_accuracy": 0.22734332829713821,
      "num_tokens": 16988274.0,
      "step": 7690
    },
    {
      "entropy": 5.157920074462891,
      "epoch": 7.75743822491175,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011525716832990722,
      "loss": 4.4608,
      "mean_token_accuracy": 0.22140037268400192,
      "num_tokens": 16998418.0,
      "step": 7695
    },
    {
      "entropy": 5.155846357345581,
      "epoch": 7.762481089258699,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011497807129771916,
      "loss": 4.4606,
      "mean_token_accuracy": 0.22475731372833252,
      "num_tokens": 17008949.0,
      "step": 7700
    },
    {
      "entropy": 5.182293558120728,
      "epoch": 7.767523953605648,
      "grad_norm": 1.265625,
      "learning_rate": 0.00011469947161737689,
      "loss": 4.4721,
      "mean_token_accuracy": 0.2222769871354103,
      "num_tokens": 17019315.0,
      "step": 7705
    },
    {
      "entropy": 5.177806425094604,
      "epoch": 7.7725668179525975,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011442137015477464,
      "loss": 4.4345,
      "mean_token_accuracy": 0.21971192359924316,
      "num_tokens": 17028942.0,
      "step": 7710
    },
    {
      "entropy": 5.147392654418946,
      "epoch": 7.777609682299547,
      "grad_norm": 1.3125,
      "learning_rate": 0.00011414376777425823,
      "loss": 4.4711,
      "mean_token_accuracy": 0.22350502461194993,
      "num_tokens": 17039302.0,
      "step": 7715
    },
    {
      "entropy": 5.142403841018677,
      "epoch": 7.782652546646495,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011386666533862214,
      "loss": 4.5133,
      "mean_token_accuracy": 0.21598893851041795,
      "num_tokens": 17051895.0,
      "step": 7720
    },
    {
      "entropy": 5.103192758560181,
      "epoch": 7.787695410993444,
      "grad_norm": 1.109375,
      "learning_rate": 0.00011359006370910727,
      "loss": 4.3813,
      "mean_token_accuracy": 0.22458279579877855,
      "num_tokens": 17062528.0,
      "step": 7725
    },
    {
      "entropy": 5.210125350952149,
      "epoch": 7.792738275340393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011331396374539771,
      "loss": 4.5502,
      "mean_token_accuracy": 0.21473172456026077,
      "num_tokens": 17074521.0,
      "step": 7730
    },
    {
      "entropy": 5.1501213073730465,
      "epoch": 7.797781139687342,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011303836630561861,
      "loss": 4.4802,
      "mean_token_accuracy": 0.22016585916280745,
      "num_tokens": 17085590.0,
      "step": 7735
    },
    {
      "entropy": 5.248780059814453,
      "epoch": 7.802824004034291,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011276327224633311,
      "loss": 4.5837,
      "mean_token_accuracy": 0.2107292726635933,
      "num_tokens": 17097165.0,
      "step": 7740
    },
    {
      "entropy": 5.145720672607422,
      "epoch": 7.80786686838124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011248868242254006,
      "loss": 4.4548,
      "mean_token_accuracy": 0.2215493157505989,
      "num_tokens": 17108612.0,
      "step": 7745
    },
    {
      "entropy": 5.216763019561768,
      "epoch": 7.81290973272819,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011221459768767074,
      "loss": 4.4811,
      "mean_token_accuracy": 0.2267795592546463,
      "num_tokens": 17118775.0,
      "step": 7750
    },
    {
      "entropy": 5.176859188079834,
      "epoch": 7.817952597075139,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011194101889358702,
      "loss": 4.5238,
      "mean_token_accuracy": 0.21635256707668304,
      "num_tokens": 17130403.0,
      "step": 7755
    },
    {
      "entropy": 5.175843667984009,
      "epoch": 7.822995461422088,
      "grad_norm": 1.25,
      "learning_rate": 0.00011166794689057794,
      "loss": 4.4964,
      "mean_token_accuracy": 0.21365673393011092,
      "num_tokens": 17141859.0,
      "step": 7760
    },
    {
      "entropy": 5.106728267669678,
      "epoch": 7.828038325769037,
      "grad_norm": 1.28125,
      "learning_rate": 0.00011139538252735765,
      "loss": 4.4707,
      "mean_token_accuracy": 0.22205861955881118,
      "num_tokens": 17152801.0,
      "step": 7765
    },
    {
      "entropy": 5.173954105377197,
      "epoch": 7.833081190115986,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011112332665106251,
      "loss": 4.4761,
      "mean_token_accuracy": 0.22187758535146712,
      "num_tokens": 17163151.0,
      "step": 7770
    },
    {
      "entropy": 5.2153829574584964,
      "epoch": 7.838124054462935,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00011085178010724845,
      "loss": 4.4648,
      "mean_token_accuracy": 0.21620999723672868,
      "num_tokens": 17173322.0,
      "step": 7775
    },
    {
      "entropy": 5.156828689575195,
      "epoch": 7.843166918809884,
      "grad_norm": 1.25,
      "learning_rate": 0.00011058074373988842,
      "loss": 4.4342,
      "mean_token_accuracy": 0.21983254253864287,
      "num_tokens": 17183270.0,
      "step": 7780
    },
    {
      "entropy": 5.1314410209655765,
      "epoch": 7.848209783156833,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001103102183913696,
      "loss": 4.4708,
      "mean_token_accuracy": 0.21938934922218323,
      "num_tokens": 17193823.0,
      "step": 7785
    },
    {
      "entropy": 5.147124528884888,
      "epoch": 7.8532526475037825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00011004020490249116,
      "loss": 4.5002,
      "mean_token_accuracy": 0.22147620022296904,
      "num_tokens": 17204971.0,
      "step": 7790
    },
    {
      "entropy": 5.144163131713867,
      "epoch": 7.858295511850731,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010977070411246107,
      "loss": 4.4785,
      "mean_token_accuracy": 0.21953538060188293,
      "num_tokens": 17216137.0,
      "step": 7795
    },
    {
      "entropy": 5.180009984970093,
      "epoch": 7.86333837619768,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010950171685889413,
      "loss": 4.5006,
      "mean_token_accuracy": 0.2164258912205696,
      "num_tokens": 17226402.0,
      "step": 7800
    },
    {
      "entropy": 5.149246406555176,
      "epoch": 7.868381240544629,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010923324397780887,
      "loss": 4.3887,
      "mean_token_accuracy": 0.22766632437705994,
      "num_tokens": 17237484.0,
      "step": 7805
    },
    {
      "entropy": 5.16761965751648,
      "epoch": 7.873424104891578,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00010896528630362531,
      "loss": 4.4308,
      "mean_token_accuracy": 0.22397062331438064,
      "num_tokens": 17247896.0,
      "step": 7810
    },
    {
      "entropy": 5.185039234161377,
      "epoch": 7.878466969238527,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010869784466916193,
      "loss": 4.4841,
      "mean_token_accuracy": 0.2158147543668747,
      "num_tokens": 17258442.0,
      "step": 7815
    },
    {
      "entropy": 5.217952919006348,
      "epoch": 7.883509833585476,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00010843091990563365,
      "loss": 4.5471,
      "mean_token_accuracy": 0.21175887286663056,
      "num_tokens": 17270166.0,
      "step": 7820
    },
    {
      "entropy": 5.156439685821534,
      "epoch": 7.888552697932425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010816451284264864,
      "loss": 4.5433,
      "mean_token_accuracy": 0.2114953577518463,
      "num_tokens": 17280404.0,
      "step": 7825
    },
    {
      "entropy": 5.143426609039307,
      "epoch": 7.8935955622793745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010789862430820626,
      "loss": 4.5145,
      "mean_token_accuracy": 0.21795062422752381,
      "num_tokens": 17292419.0,
      "step": 7830
    },
    {
      "entropy": 5.153320837020874,
      "epoch": 7.898638426626324,
      "grad_norm": 1.25,
      "learning_rate": 0.00010763325512869422,
      "loss": 4.4877,
      "mean_token_accuracy": 0.21953888684511186,
      "num_tokens": 17304152.0,
      "step": 7835
    },
    {
      "entropy": 5.145594358444214,
      "epoch": 7.903681290973273,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001073684061288861,
      "loss": 4.4352,
      "mean_token_accuracy": 0.21777669936418534,
      "num_tokens": 17315235.0,
      "step": 7840
    },
    {
      "entropy": 5.195118761062622,
      "epoch": 7.908724155320222,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010710407813193859,
      "loss": 4.588,
      "mean_token_accuracy": 0.2187290832400322,
      "num_tokens": 17327291.0,
      "step": 7845
    },
    {
      "entropy": 5.206640958786011,
      "epoch": 7.913767019667171,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010684027195938936,
      "loss": 4.5154,
      "mean_token_accuracy": 0.2184297800064087,
      "num_tokens": 17337596.0,
      "step": 7850
    },
    {
      "entropy": 5.174378347396851,
      "epoch": 7.91880988401412,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001065769884311539,
      "loss": 4.5049,
      "mean_token_accuracy": 0.21329114735126495,
      "num_tokens": 17348980.0,
      "step": 7855
    },
    {
      "entropy": 5.231718301773071,
      "epoch": 7.923852748361069,
      "grad_norm": 1.421875,
      "learning_rate": 0.00010631422836552354,
      "loss": 4.5615,
      "mean_token_accuracy": 0.21745679527521133,
      "num_tokens": 17360261.0,
      "step": 7860
    },
    {
      "entropy": 5.158034992218018,
      "epoch": 7.928895612708018,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010605199257916269,
      "loss": 4.4794,
      "mean_token_accuracy": 0.21889538019895555,
      "num_tokens": 17371370.0,
      "step": 7865
    },
    {
      "entropy": 5.1622003555297855,
      "epoch": 7.9339384770549675,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010579028188710619,
      "loss": 4.4387,
      "mean_token_accuracy": 0.23299529999494553,
      "num_tokens": 17383728.0,
      "step": 7870
    },
    {
      "entropy": 5.1969305038452145,
      "epoch": 7.938981341401917,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010552909710275693,
      "loss": 4.4969,
      "mean_token_accuracy": 0.22433869540691376,
      "num_tokens": 17394627.0,
      "step": 7875
    },
    {
      "entropy": 5.2154473781585695,
      "epoch": 7.944024205748866,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010526843903788319,
      "loss": 4.511,
      "mean_token_accuracy": 0.2184775084257126,
      "num_tokens": 17405072.0,
      "step": 7880
    },
    {
      "entropy": 5.205847597122192,
      "epoch": 7.949067070095815,
      "grad_norm": 1.203125,
      "learning_rate": 0.00010500830850261634,
      "loss": 4.5334,
      "mean_token_accuracy": 0.2222561553120613,
      "num_tokens": 17416703.0,
      "step": 7885
    },
    {
      "entropy": 5.120602750778199,
      "epoch": 7.954109934442764,
      "grad_norm": 1.25,
      "learning_rate": 0.00010474870630544797,
      "loss": 4.4481,
      "mean_token_accuracy": 0.22238091081380845,
      "num_tokens": 17429419.0,
      "step": 7890
    },
    {
      "entropy": 5.178711652755737,
      "epoch": 7.959152798789713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010448963325322772,
      "loss": 4.5236,
      "mean_token_accuracy": 0.2175870418548584,
      "num_tokens": 17440935.0,
      "step": 7895
    },
    {
      "entropy": 5.211516046524048,
      "epoch": 7.964195663136661,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010423109015116065,
      "loss": 4.5802,
      "mean_token_accuracy": 0.21506956815719605,
      "num_tokens": 17451403.0,
      "step": 7900
    },
    {
      "entropy": 5.169587182998657,
      "epoch": 7.96923852748361,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00010397307780280477,
      "loss": 4.4796,
      "mean_token_accuracy": 0.22592357993125917,
      "num_tokens": 17462071.0,
      "step": 7905
    },
    {
      "entropy": 5.1502608299255375,
      "epoch": 7.9742813918305595,
      "grad_norm": 1.125,
      "learning_rate": 0.00010371559701006825,
      "loss": 4.5008,
      "mean_token_accuracy": 0.22126431316137313,
      "num_tokens": 17474333.0,
      "step": 7910
    },
    {
      "entropy": 5.132789945602417,
      "epoch": 7.979324256177509,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010345864857320747,
      "loss": 4.4239,
      "mean_token_accuracy": 0.22141513973474503,
      "num_tokens": 17485305.0,
      "step": 7915
    },
    {
      "entropy": 5.159731674194336,
      "epoch": 7.984367120524458,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00010320223329082396,
      "loss": 4.5219,
      "mean_token_accuracy": 0.2124677672982216,
      "num_tokens": 17497221.0,
      "step": 7920
    },
    {
      "entropy": 5.1624650955200195,
      "epoch": 7.989409984871407,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001029463519598624,
      "loss": 4.5022,
      "mean_token_accuracy": 0.21461506485939025,
      "num_tokens": 17509476.0,
      "step": 7925
    },
    {
      "entropy": 5.221511316299439,
      "epoch": 7.994452849218356,
      "grad_norm": 1.234375,
      "learning_rate": 0.00010269100537560789,
      "loss": 4.4517,
      "mean_token_accuracy": 0.22929217666387558,
      "num_tokens": 17521035.0,
      "step": 7930
    },
    {
      "entropy": 5.1760015964508055,
      "epoch": 7.999495713565305,
      "grad_norm": 1.265625,
      "learning_rate": 0.00010243619433168352,
      "loss": 4.4715,
      "mean_token_accuracy": 0.21471783220767976,
      "num_tokens": 17532650.0,
      "step": 7935
    },
    {
      "entropy": 5.129750304751926,
      "epoch": 8.00403429147756,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010218191962004781,
      "loss": 4.3059,
      "mean_token_accuracy": 0.23605398337046304,
      "num_tokens": 17541059.0,
      "step": 7940
    },
    {
      "entropy": 5.120468616485596,
      "epoch": 8.009077155824508,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00010192818203099258,
      "loss": 4.295,
      "mean_token_accuracy": 0.23822735100984574,
      "num_tokens": 17551117.0,
      "step": 7945
    },
    {
      "entropy": 5.145382499694824,
      "epoch": 8.014120020171458,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001016749823531399,
      "loss": 4.3856,
      "mean_token_accuracy": 0.2273772746324539,
      "num_tokens": 17563736.0,
      "step": 7950
    },
    {
      "entropy": 5.170421409606933,
      "epoch": 8.019162884518407,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010142232137344036,
      "loss": 4.4174,
      "mean_token_accuracy": 0.22626265585422517,
      "num_tokens": 17575774.0,
      "step": 7955
    },
    {
      "entropy": 5.192155027389527,
      "epoch": 8.024205748865356,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010117019987717005,
      "loss": 4.3998,
      "mean_token_accuracy": 0.23100073933601378,
      "num_tokens": 17586319.0,
      "step": 7960
    },
    {
      "entropy": 5.171562767028808,
      "epoch": 8.029248613212305,
      "grad_norm": 1.1875,
      "learning_rate": 0.00010091861864792848,
      "loss": 4.3721,
      "mean_token_accuracy": 0.23185830265283586,
      "num_tokens": 17597142.0,
      "step": 7965
    },
    {
      "entropy": 5.11940369606018,
      "epoch": 8.034291477559254,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001006675784676358,
      "loss": 4.4025,
      "mean_token_accuracy": 0.22572533041238785,
      "num_tokens": 17608184.0,
      "step": 7970
    },
    {
      "entropy": 5.144017076492309,
      "epoch": 8.039334341906203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00010041708011653081,
      "loss": 4.3858,
      "mean_token_accuracy": 0.2287905380129814,
      "num_tokens": 17620148.0,
      "step": 7975
    },
    {
      "entropy": 5.168688440322876,
      "epoch": 8.044377206253152,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010016712437316799,
      "loss": 4.3847,
      "mean_token_accuracy": 0.23021430522203445,
      "num_tokens": 17630686.0,
      "step": 7980
    },
    {
      "entropy": 5.076414632797241,
      "epoch": 8.049420070600101,
      "grad_norm": 1.1796875,
      "learning_rate": 9.991771201441569e-05,
      "loss": 4.2996,
      "mean_token_accuracy": 0.2414441868662834,
      "num_tokens": 17643521.0,
      "step": 7985
    },
    {
      "entropy": 5.202349615097046,
      "epoch": 8.05446293494705,
      "grad_norm": 1.21875,
      "learning_rate": 9.966884381545322e-05,
      "loss": 4.4074,
      "mean_token_accuracy": 0.22451456785202026,
      "num_tokens": 17653843.0,
      "step": 7990
    },
    {
      "entropy": 5.164351844787598,
      "epoch": 8.059505799294,
      "grad_norm": 1.2734375,
      "learning_rate": 9.942052054976872e-05,
      "loss": 4.3414,
      "mean_token_accuracy": 0.22360394895076752,
      "num_tokens": 17665566.0,
      "step": 7995
    },
    {
      "entropy": 5.158666896820068,
      "epoch": 8.064548663640949,
      "grad_norm": 1.1953125,
      "learning_rate": 9.917274298915665e-05,
      "loss": 4.4564,
      "mean_token_accuracy": 0.22265976518392563,
      "num_tokens": 17678249.0,
      "step": 8000
    },
    {
      "epoch": 8.064548663640949,
      "eval_entropy": 5.102120893298821,
      "eval_loss": 5.695427417755127,
      "eval_mean_token_accuracy": 0.15428788666127596,
      "eval_num_tokens": 17678249.0,
      "eval_runtime": 2.6437,
      "eval_samples_per_second": 1186.954,
      "eval_steps_per_second": 148.653,
      "step": 8000
    },
    {
      "entropy": 5.151681041717529,
      "epoch": 8.069591527987898,
      "grad_norm": 1.421875,
      "learning_rate": 9.892551190371529e-05,
      "loss": 4.3732,
      "mean_token_accuracy": 0.23386379033327104,
      "num_tokens": 17688437.0,
      "step": 8005
    },
    {
      "entropy": 5.152360200881958,
      "epoch": 8.074634392334847,
      "grad_norm": 1.2109375,
      "learning_rate": 9.867882806184469e-05,
      "loss": 4.3809,
      "mean_token_accuracy": 0.22618581354618073,
      "num_tokens": 17698192.0,
      "step": 8010
    },
    {
      "entropy": 5.0930005550384525,
      "epoch": 8.079677256681796,
      "grad_norm": 1.1953125,
      "learning_rate": 9.843269223024376e-05,
      "loss": 4.3357,
      "mean_token_accuracy": 0.23865833431482314,
      "num_tokens": 17709126.0,
      "step": 8015
    },
    {
      "entropy": 5.167470693588257,
      "epoch": 8.084720121028745,
      "grad_norm": 1.40625,
      "learning_rate": 9.818710517390841e-05,
      "loss": 4.4071,
      "mean_token_accuracy": 0.2226140171289444,
      "num_tokens": 17719127.0,
      "step": 8020
    },
    {
      "entropy": 5.157788467407227,
      "epoch": 8.089762985375694,
      "grad_norm": 1.21875,
      "learning_rate": 9.794206765612893e-05,
      "loss": 4.3882,
      "mean_token_accuracy": 0.22911464869976045,
      "num_tokens": 17730656.0,
      "step": 8025
    },
    {
      "entropy": 5.170150184631348,
      "epoch": 8.094805849722642,
      "grad_norm": 1.4453125,
      "learning_rate": 9.76975804384876e-05,
      "loss": 4.3577,
      "mean_token_accuracy": 0.23315648287534713,
      "num_tokens": 17741224.0,
      "step": 8030
    },
    {
      "entropy": 5.143690061569214,
      "epoch": 8.09984871406959,
      "grad_norm": 1.15625,
      "learning_rate": 9.745364428085618e-05,
      "loss": 4.3776,
      "mean_token_accuracy": 0.22909916788339615,
      "num_tokens": 17751398.0,
      "step": 8035
    },
    {
      "entropy": 5.136434507369995,
      "epoch": 8.10489157841654,
      "grad_norm": 1.34375,
      "learning_rate": 9.721025994139404e-05,
      "loss": 4.3647,
      "mean_token_accuracy": 0.2297883689403534,
      "num_tokens": 17761218.0,
      "step": 8040
    },
    {
      "entropy": 5.156273555755615,
      "epoch": 8.109934442763489,
      "grad_norm": 1.3203125,
      "learning_rate": 9.696742817654519e-05,
      "loss": 4.4212,
      "mean_token_accuracy": 0.22611196637153624,
      "num_tokens": 17771401.0,
      "step": 8045
    },
    {
      "entropy": 5.176234006881714,
      "epoch": 8.114977307110438,
      "grad_norm": 1.3125,
      "learning_rate": 9.672514974103643e-05,
      "loss": 4.3847,
      "mean_token_accuracy": 0.228081014752388,
      "num_tokens": 17781067.0,
      "step": 8050
    },
    {
      "entropy": 5.1388654708862305,
      "epoch": 8.120020171457387,
      "grad_norm": 1.2890625,
      "learning_rate": 9.648342538787467e-05,
      "loss": 4.369,
      "mean_token_accuracy": 0.23767980188131332,
      "num_tokens": 17791327.0,
      "step": 8055
    },
    {
      "entropy": 5.133828353881836,
      "epoch": 8.125063035804336,
      "grad_norm": 1.3984375,
      "learning_rate": 9.62422558683449e-05,
      "loss": 4.3501,
      "mean_token_accuracy": 0.2262016698718071,
      "num_tokens": 17800963.0,
      "step": 8060
    },
    {
      "entropy": 5.173811769485473,
      "epoch": 8.130105900151285,
      "grad_norm": 1.34375,
      "learning_rate": 9.600164193200742e-05,
      "loss": 4.3892,
      "mean_token_accuracy": 0.23020390868186952,
      "num_tokens": 17811738.0,
      "step": 8065
    },
    {
      "entropy": 5.121002292633056,
      "epoch": 8.135148764498235,
      "grad_norm": 1.2734375,
      "learning_rate": 9.576158432669599e-05,
      "loss": 4.3273,
      "mean_token_accuracy": 0.23602456152439116,
      "num_tokens": 17822912.0,
      "step": 8070
    },
    {
      "entropy": 5.157894945144653,
      "epoch": 8.140191628845184,
      "grad_norm": 1.2578125,
      "learning_rate": 9.552208379851507e-05,
      "loss": 4.4248,
      "mean_token_accuracy": 0.22711092084646226,
      "num_tokens": 17834528.0,
      "step": 8075
    },
    {
      "entropy": 5.161261463165284,
      "epoch": 8.145234493192133,
      "grad_norm": 1.2421875,
      "learning_rate": 9.528314109183788e-05,
      "loss": 4.4713,
      "mean_token_accuracy": 0.22533195167779924,
      "num_tokens": 17846675.0,
      "step": 8080
    },
    {
      "entropy": 5.128841209411621,
      "epoch": 8.150277357539082,
      "grad_norm": 1.3046875,
      "learning_rate": 9.504475694930392e-05,
      "loss": 4.3902,
      "mean_token_accuracy": 0.2320045292377472,
      "num_tokens": 17857722.0,
      "step": 8085
    },
    {
      "entropy": 5.148091650009155,
      "epoch": 8.155320221886031,
      "grad_norm": 1.390625,
      "learning_rate": 9.480693211181654e-05,
      "loss": 4.3977,
      "mean_token_accuracy": 0.22392394691705703,
      "num_tokens": 17867592.0,
      "step": 8090
    },
    {
      "entropy": 5.198516893386841,
      "epoch": 8.16036308623298,
      "grad_norm": 1.4296875,
      "learning_rate": 9.456966731854089e-05,
      "loss": 4.3707,
      "mean_token_accuracy": 0.22869945615530013,
      "num_tokens": 17879023.0,
      "step": 8095
    },
    {
      "entropy": 5.18330864906311,
      "epoch": 8.16540595057993,
      "grad_norm": 1.2421875,
      "learning_rate": 9.43329633069013e-05,
      "loss": 4.4263,
      "mean_token_accuracy": 0.22802911698818207,
      "num_tokens": 17891509.0,
      "step": 8100
    },
    {
      "entropy": 5.158009624481201,
      "epoch": 8.170448814926878,
      "grad_norm": 1.3359375,
      "learning_rate": 9.409682081257947e-05,
      "loss": 4.4777,
      "mean_token_accuracy": 0.22269847691059114,
      "num_tokens": 17903680.0,
      "step": 8105
    },
    {
      "entropy": 5.183296966552734,
      "epoch": 8.175491679273827,
      "grad_norm": 1.2421875,
      "learning_rate": 9.386124056951154e-05,
      "loss": 4.4018,
      "mean_token_accuracy": 0.23517338186502457,
      "num_tokens": 17916474.0,
      "step": 8110
    },
    {
      "entropy": 5.162954425811767,
      "epoch": 8.180534543620777,
      "grad_norm": 1.1953125,
      "learning_rate": 9.362622330988641e-05,
      "loss": 4.4537,
      "mean_token_accuracy": 0.2249269336462021,
      "num_tokens": 17929186.0,
      "step": 8115
    },
    {
      "entropy": 5.12801308631897,
      "epoch": 8.185577407967726,
      "grad_norm": 1.296875,
      "learning_rate": 9.339176976414317e-05,
      "loss": 4.3828,
      "mean_token_accuracy": 0.23194323480129242,
      "num_tokens": 17939720.0,
      "step": 8120
    },
    {
      "entropy": 5.175691938400268,
      "epoch": 8.190620272314675,
      "grad_norm": 1.171875,
      "learning_rate": 9.315788066096892e-05,
      "loss": 4.3851,
      "mean_token_accuracy": 0.23168755322694778,
      "num_tokens": 17949410.0,
      "step": 8125
    },
    {
      "entropy": 5.160462379455566,
      "epoch": 8.195663136661624,
      "grad_norm": 1.453125,
      "learning_rate": 9.292455672729619e-05,
      "loss": 4.4133,
      "mean_token_accuracy": 0.22186147570610046,
      "num_tokens": 17959715.0,
      "step": 8130
    },
    {
      "entropy": 5.092800903320312,
      "epoch": 8.200706001008573,
      "grad_norm": 1.3046875,
      "learning_rate": 9.269179868830135e-05,
      "loss": 4.3022,
      "mean_token_accuracy": 0.2425137057900429,
      "num_tokens": 17971166.0,
      "step": 8135
    },
    {
      "entropy": 5.1288636207580565,
      "epoch": 8.205748865355522,
      "grad_norm": 1.078125,
      "learning_rate": 9.24596072674016e-05,
      "loss": 4.3717,
      "mean_token_accuracy": 0.23047407269477843,
      "num_tokens": 17983629.0,
      "step": 8140
    },
    {
      "entropy": 5.193580913543701,
      "epoch": 8.210791729702471,
      "grad_norm": 1.21875,
      "learning_rate": 9.222798318625327e-05,
      "loss": 4.4132,
      "mean_token_accuracy": 0.22606464177370073,
      "num_tokens": 17994949.0,
      "step": 8145
    },
    {
      "entropy": 5.1623023509979244,
      "epoch": 8.21583459404942,
      "grad_norm": 1.234375,
      "learning_rate": 9.19969271647494e-05,
      "loss": 4.3959,
      "mean_token_accuracy": 0.2293732687830925,
      "num_tokens": 18005221.0,
      "step": 8150
    },
    {
      "entropy": 5.177295589447022,
      "epoch": 8.22087745839637,
      "grad_norm": 1.21875,
      "learning_rate": 9.176643992101749e-05,
      "loss": 4.4086,
      "mean_token_accuracy": 0.22230487316846848,
      "num_tokens": 18015919.0,
      "step": 8155
    },
    {
      "entropy": 5.15458083152771,
      "epoch": 8.225920322743319,
      "grad_norm": 1.3984375,
      "learning_rate": 9.153652217141703e-05,
      "loss": 4.3247,
      "mean_token_accuracy": 0.23471898287534715,
      "num_tokens": 18025948.0,
      "step": 8160
    },
    {
      "entropy": 5.198016548156739,
      "epoch": 8.230963187090268,
      "grad_norm": 1.3359375,
      "learning_rate": 9.130717463053787e-05,
      "loss": 4.5391,
      "mean_token_accuracy": 0.22256282418966294,
      "num_tokens": 18037619.0,
      "step": 8165
    },
    {
      "entropy": 5.161934471130371,
      "epoch": 8.236006051437217,
      "grad_norm": 1.265625,
      "learning_rate": 9.10783980111973e-05,
      "loss": 4.4054,
      "mean_token_accuracy": 0.22485827803611755,
      "num_tokens": 18048528.0,
      "step": 8170
    },
    {
      "entropy": 5.223745822906494,
      "epoch": 8.241048915784166,
      "grad_norm": 1.375,
      "learning_rate": 9.085019302443842e-05,
      "loss": 4.4421,
      "mean_token_accuracy": 0.22950258404016494,
      "num_tokens": 18060423.0,
      "step": 8175
    },
    {
      "entropy": 5.184373092651367,
      "epoch": 8.246091780131115,
      "grad_norm": 1.25,
      "learning_rate": 9.062256037952757e-05,
      "loss": 4.4185,
      "mean_token_accuracy": 0.2204434722661972,
      "num_tokens": 18071281.0,
      "step": 8180
    },
    {
      "entropy": 5.130430603027344,
      "epoch": 8.251134644478064,
      "grad_norm": 1.2734375,
      "learning_rate": 9.039550078395234e-05,
      "loss": 4.3899,
      "mean_token_accuracy": 0.22606073021888734,
      "num_tokens": 18082320.0,
      "step": 8185
    },
    {
      "entropy": 5.177854537963867,
      "epoch": 8.256177508825013,
      "grad_norm": 1.390625,
      "learning_rate": 9.016901494341906e-05,
      "loss": 4.3764,
      "mean_token_accuracy": 0.22474264949560166,
      "num_tokens": 18093557.0,
      "step": 8190
    },
    {
      "entropy": 5.167903280258178,
      "epoch": 8.261220373171962,
      "grad_norm": 1.2578125,
      "learning_rate": 8.994310356185098e-05,
      "loss": 4.3902,
      "mean_token_accuracy": 0.22610028833150864,
      "num_tokens": 18104515.0,
      "step": 8195
    },
    {
      "entropy": 5.192443561553955,
      "epoch": 8.266263237518912,
      "grad_norm": 1.296875,
      "learning_rate": 8.971776734138601e-05,
      "loss": 4.429,
      "mean_token_accuracy": 0.2279272571206093,
      "num_tokens": 18118065.0,
      "step": 8200
    },
    {
      "entropy": 5.199595499038696,
      "epoch": 8.27130610186586,
      "grad_norm": 1.3359375,
      "learning_rate": 8.949300698237414e-05,
      "loss": 4.4488,
      "mean_token_accuracy": 0.22315744012594224,
      "num_tokens": 18129001.0,
      "step": 8205
    },
    {
      "entropy": 5.165442276000976,
      "epoch": 8.276348966212808,
      "grad_norm": 1.28125,
      "learning_rate": 8.926882318337586e-05,
      "loss": 4.4837,
      "mean_token_accuracy": 0.22171541750431062,
      "num_tokens": 18140987.0,
      "step": 8210
    },
    {
      "entropy": 5.086031866073609,
      "epoch": 8.281391830559757,
      "grad_norm": 1.4375,
      "learning_rate": 8.904521664115956e-05,
      "loss": 4.2608,
      "mean_token_accuracy": 0.24366332292556764,
      "num_tokens": 18151860.0,
      "step": 8215
    },
    {
      "entropy": 5.144732475280762,
      "epoch": 8.286434694906706,
      "grad_norm": 1.2890625,
      "learning_rate": 8.882218805069958e-05,
      "loss": 4.3241,
      "mean_token_accuracy": 0.23358970880508423,
      "num_tokens": 18162347.0,
      "step": 8220
    },
    {
      "entropy": 5.161101865768432,
      "epoch": 8.291477559253655,
      "grad_norm": 1.3359375,
      "learning_rate": 8.859973810517383e-05,
      "loss": 4.3321,
      "mean_token_accuracy": 0.22890213280916213,
      "num_tokens": 18172880.0,
      "step": 8225
    },
    {
      "entropy": 5.131391191482544,
      "epoch": 8.296520423600604,
      "grad_norm": 1.375,
      "learning_rate": 8.8377867495962e-05,
      "loss": 4.4168,
      "mean_token_accuracy": 0.2276117518544197,
      "num_tokens": 18184813.0,
      "step": 8230
    },
    {
      "entropy": 5.165047550201416,
      "epoch": 8.301563287947554,
      "grad_norm": 1.2109375,
      "learning_rate": 8.815657691264288e-05,
      "loss": 4.3502,
      "mean_token_accuracy": 0.23218773156404496,
      "num_tokens": 18196669.0,
      "step": 8235
    },
    {
      "entropy": 5.114106559753418,
      "epoch": 8.306606152294503,
      "grad_norm": 1.40625,
      "learning_rate": 8.793586704299284e-05,
      "loss": 4.358,
      "mean_token_accuracy": 0.23516918271780013,
      "num_tokens": 18208456.0,
      "step": 8240
    },
    {
      "entropy": 5.112807655334473,
      "epoch": 8.311649016641452,
      "grad_norm": 1.1875,
      "learning_rate": 8.77157385729832e-05,
      "loss": 4.3276,
      "mean_token_accuracy": 0.23378737270832062,
      "num_tokens": 18219087.0,
      "step": 8245
    },
    {
      "entropy": 5.187710762023926,
      "epoch": 8.316691880988401,
      "grad_norm": 1.296875,
      "learning_rate": 8.749619218677842e-05,
      "loss": 4.3875,
      "mean_token_accuracy": 0.23108946084976195,
      "num_tokens": 18229494.0,
      "step": 8250
    },
    {
      "entropy": 5.120807838439942,
      "epoch": 8.32173474533535,
      "grad_norm": 1.265625,
      "learning_rate": 8.727722856673357e-05,
      "loss": 4.3768,
      "mean_token_accuracy": 0.234053897857666,
      "num_tokens": 18240914.0,
      "step": 8255
    },
    {
      "entropy": 5.19225926399231,
      "epoch": 8.3267776096823,
      "grad_norm": 1.546875,
      "learning_rate": 8.705884839339279e-05,
      "loss": 4.4352,
      "mean_token_accuracy": 0.22599344998598098,
      "num_tokens": 18250659.0,
      "step": 8260
    },
    {
      "entropy": 5.189092445373535,
      "epoch": 8.331820474029248,
      "grad_norm": 1.5546875,
      "learning_rate": 8.684105234548658e-05,
      "loss": 4.3889,
      "mean_token_accuracy": 0.2267034411430359,
      "num_tokens": 18261450.0,
      "step": 8265
    },
    {
      "entropy": 5.180127334594727,
      "epoch": 8.336863338376197,
      "grad_norm": 1.2734375,
      "learning_rate": 8.66238410999301e-05,
      "loss": 4.369,
      "mean_token_accuracy": 0.226296067237854,
      "num_tokens": 18272210.0,
      "step": 8270
    },
    {
      "entropy": 5.128127908706665,
      "epoch": 8.341906202723147,
      "grad_norm": 1.3828125,
      "learning_rate": 8.640721533182092e-05,
      "loss": 4.3814,
      "mean_token_accuracy": 0.2264604836702347,
      "num_tokens": 18283006.0,
      "step": 8275
    },
    {
      "entropy": 5.097172737121582,
      "epoch": 8.346949067070096,
      "grad_norm": 1.25,
      "learning_rate": 8.619117571443702e-05,
      "loss": 4.3605,
      "mean_token_accuracy": 0.2335774153470993,
      "num_tokens": 18293959.0,
      "step": 8280
    },
    {
      "entropy": 5.145702743530274,
      "epoch": 8.351991931417045,
      "grad_norm": 1.3828125,
      "learning_rate": 8.597572291923438e-05,
      "loss": 4.4868,
      "mean_token_accuracy": 0.2240848273038864,
      "num_tokens": 18304003.0,
      "step": 8285
    },
    {
      "entropy": 5.092479562759399,
      "epoch": 8.357034795763994,
      "grad_norm": 1.125,
      "learning_rate": 8.576085761584534e-05,
      "loss": 4.3551,
      "mean_token_accuracy": 0.24194653183221818,
      "num_tokens": 18315784.0,
      "step": 8290
    },
    {
      "entropy": 5.126586771011352,
      "epoch": 8.362077660110943,
      "grad_norm": 1.3359375,
      "learning_rate": 8.554658047207616e-05,
      "loss": 4.2923,
      "mean_token_accuracy": 0.2393563225865364,
      "num_tokens": 18327645.0,
      "step": 8295
    },
    {
      "entropy": 5.1454023838043215,
      "epoch": 8.367120524457892,
      "grad_norm": 1.328125,
      "learning_rate": 8.533289215390516e-05,
      "loss": 4.3936,
      "mean_token_accuracy": 0.22914001792669297,
      "num_tokens": 18338134.0,
      "step": 8300
    },
    {
      "entropy": 5.179714822769165,
      "epoch": 8.372163388804841,
      "grad_norm": 1.2890625,
      "learning_rate": 8.51197933254806e-05,
      "loss": 4.4377,
      "mean_token_accuracy": 0.22828578650951387,
      "num_tokens": 18349736.0,
      "step": 8305
    },
    {
      "entropy": 5.171978330612182,
      "epoch": 8.37720625315179,
      "grad_norm": 1.328125,
      "learning_rate": 8.490728464911849e-05,
      "loss": 4.409,
      "mean_token_accuracy": 0.22720164954662322,
      "num_tokens": 18361874.0,
      "step": 8310
    },
    {
      "entropy": 5.1254970073699955,
      "epoch": 8.38224911749874,
      "grad_norm": 1.28125,
      "learning_rate": 8.469536678530085e-05,
      "loss": 4.3806,
      "mean_token_accuracy": 0.22909559011459352,
      "num_tokens": 18373360.0,
      "step": 8315
    },
    {
      "entropy": 5.085026502609253,
      "epoch": 8.387291981845689,
      "grad_norm": 1.375,
      "learning_rate": 8.448404039267308e-05,
      "loss": 4.3269,
      "mean_token_accuracy": 0.233455154299736,
      "num_tokens": 18384139.0,
      "step": 8320
    },
    {
      "entropy": 5.1967545509338375,
      "epoch": 8.392334846192638,
      "grad_norm": 1.3671875,
      "learning_rate": 8.427330612804266e-05,
      "loss": 4.4492,
      "mean_token_accuracy": 0.2272382378578186,
      "num_tokens": 18396093.0,
      "step": 8325
    },
    {
      "entropy": 5.139287328720092,
      "epoch": 8.397377710539587,
      "grad_norm": 1.3203125,
      "learning_rate": 8.406316464637639e-05,
      "loss": 4.4113,
      "mean_token_accuracy": 0.22990965694189072,
      "num_tokens": 18407591.0,
      "step": 8330
    },
    {
      "entropy": 5.181426239013672,
      "epoch": 8.402420574886536,
      "grad_norm": 1.34375,
      "learning_rate": 8.385361660079896e-05,
      "loss": 4.4355,
      "mean_token_accuracy": 0.22903767973184586,
      "num_tokens": 18418248.0,
      "step": 8335
    },
    {
      "entropy": 5.162497186660767,
      "epoch": 8.407463439233485,
      "grad_norm": 1.1953125,
      "learning_rate": 8.364466264259046e-05,
      "loss": 4.3681,
      "mean_token_accuracy": 0.2261652484536171,
      "num_tokens": 18428474.0,
      "step": 8340
    },
    {
      "entropy": 5.174172258377075,
      "epoch": 8.412506303580434,
      "grad_norm": 1.28125,
      "learning_rate": 8.343630342118471e-05,
      "loss": 4.4223,
      "mean_token_accuracy": 0.22251778542995454,
      "num_tokens": 18439378.0,
      "step": 8345
    },
    {
      "entropy": 5.122017431259155,
      "epoch": 8.417549167927383,
      "grad_norm": 1.3359375,
      "learning_rate": 8.322853958416685e-05,
      "loss": 4.3735,
      "mean_token_accuracy": 0.23093120008707047,
      "num_tokens": 18450229.0,
      "step": 8350
    },
    {
      "entropy": 5.1595457077026365,
      "epoch": 8.422592032274332,
      "grad_norm": 1.3515625,
      "learning_rate": 8.30213717772718e-05,
      "loss": 4.3959,
      "mean_token_accuracy": 0.22536484003067017,
      "num_tokens": 18460960.0,
      "step": 8355
    },
    {
      "entropy": 5.148212051391601,
      "epoch": 8.427634896621282,
      "grad_norm": 1.2734375,
      "learning_rate": 8.281480064438176e-05,
      "loss": 4.4216,
      "mean_token_accuracy": 0.2262904390692711,
      "num_tokens": 18472029.0,
      "step": 8360
    },
    {
      "entropy": 5.145326471328735,
      "epoch": 8.43267776096823,
      "grad_norm": 1.203125,
      "learning_rate": 8.260882682752464e-05,
      "loss": 4.3971,
      "mean_token_accuracy": 0.224821537733078,
      "num_tokens": 18483610.0,
      "step": 8365
    },
    {
      "entropy": 5.172882223129273,
      "epoch": 8.43772062531518,
      "grad_norm": 1.2578125,
      "learning_rate": 8.24034509668718e-05,
      "loss": 4.438,
      "mean_token_accuracy": 0.22512471824884414,
      "num_tokens": 18494328.0,
      "step": 8370
    },
    {
      "entropy": 5.1417724609375,
      "epoch": 8.442763489662129,
      "grad_norm": 1.328125,
      "learning_rate": 8.219867370073623e-05,
      "loss": 4.4127,
      "mean_token_accuracy": 0.22132449597120285,
      "num_tokens": 18505432.0,
      "step": 8375
    },
    {
      "entropy": 5.1651874542236325,
      "epoch": 8.447806354009078,
      "grad_norm": 1.171875,
      "learning_rate": 8.199449566557029e-05,
      "loss": 4.4598,
      "mean_token_accuracy": 0.21518996655941008,
      "num_tokens": 18517878.0,
      "step": 8380
    },
    {
      "entropy": 5.189282560348511,
      "epoch": 8.452849218356025,
      "grad_norm": 1.2734375,
      "learning_rate": 8.179091749596409e-05,
      "loss": 4.3899,
      "mean_token_accuracy": 0.23130193948745728,
      "num_tokens": 18529156.0,
      "step": 8385
    },
    {
      "entropy": 5.152287244796753,
      "epoch": 8.457892082702974,
      "grad_norm": 1.171875,
      "learning_rate": 8.158793982464322e-05,
      "loss": 4.4144,
      "mean_token_accuracy": 0.2277046650648117,
      "num_tokens": 18540634.0,
      "step": 8390
    },
    {
      "entropy": 5.145708703994751,
      "epoch": 8.462934947049924,
      "grad_norm": 1.4140625,
      "learning_rate": 8.138556328246698e-05,
      "loss": 4.3789,
      "mean_token_accuracy": 0.22981030493974686,
      "num_tokens": 18551006.0,
      "step": 8395
    },
    {
      "entropy": 5.1357622146606445,
      "epoch": 8.467977811396873,
      "grad_norm": 1.359375,
      "learning_rate": 8.118378849842634e-05,
      "loss": 4.37,
      "mean_token_accuracy": 0.23288295567035674,
      "num_tokens": 18561933.0,
      "step": 8400
    },
    {
      "entropy": 5.188577365875244,
      "epoch": 8.473020675743822,
      "grad_norm": 1.2734375,
      "learning_rate": 8.098261609964204e-05,
      "loss": 4.4997,
      "mean_token_accuracy": 0.2246645763516426,
      "num_tokens": 18574104.0,
      "step": 8405
    },
    {
      "entropy": 5.146593809127808,
      "epoch": 8.478063540090771,
      "grad_norm": 1.3203125,
      "learning_rate": 8.078204671136245e-05,
      "loss": 4.4613,
      "mean_token_accuracy": 0.2280785247683525,
      "num_tokens": 18584268.0,
      "step": 8410
    },
    {
      "entropy": 5.163978195190429,
      "epoch": 8.48310640443772,
      "grad_norm": 1.3046875,
      "learning_rate": 8.05820809569619e-05,
      "loss": 4.424,
      "mean_token_accuracy": 0.22420844733715056,
      "num_tokens": 18594440.0,
      "step": 8415
    },
    {
      "entropy": 5.183492040634155,
      "epoch": 8.48814926878467,
      "grad_norm": 1.2265625,
      "learning_rate": 8.03827194579386e-05,
      "loss": 4.4405,
      "mean_token_accuracy": 0.2248072072863579,
      "num_tokens": 18605366.0,
      "step": 8420
    },
    {
      "entropy": 5.222967672348022,
      "epoch": 8.493192133131618,
      "grad_norm": 1.4140625,
      "learning_rate": 8.01839628339126e-05,
      "loss": 4.4431,
      "mean_token_accuracy": 0.23022955358028413,
      "num_tokens": 18616154.0,
      "step": 8425
    },
    {
      "entropy": 5.162531757354737,
      "epoch": 8.498234997478567,
      "grad_norm": 1.3125,
      "learning_rate": 7.998581170262412e-05,
      "loss": 4.4125,
      "mean_token_accuracy": 0.2245088189840317,
      "num_tokens": 18626054.0,
      "step": 8430
    },
    {
      "entropy": 5.196127653121948,
      "epoch": 8.503277861825516,
      "grad_norm": 1.25,
      "learning_rate": 7.978826667993141e-05,
      "loss": 4.45,
      "mean_token_accuracy": 0.22675908505916595,
      "num_tokens": 18637680.0,
      "step": 8435
    },
    {
      "entropy": 5.197119140625,
      "epoch": 8.508320726172466,
      "grad_norm": 1.234375,
      "learning_rate": 7.95913283798091e-05,
      "loss": 4.4185,
      "mean_token_accuracy": 0.22625104635953902,
      "num_tokens": 18647713.0,
      "step": 8440
    },
    {
      "entropy": 5.245308113098145,
      "epoch": 8.513363590519415,
      "grad_norm": 1.2265625,
      "learning_rate": 7.93949974143458e-05,
      "loss": 4.4958,
      "mean_token_accuracy": 0.22607841789722444,
      "num_tokens": 18660003.0,
      "step": 8445
    },
    {
      "entropy": 5.164935684204101,
      "epoch": 8.518406454866364,
      "grad_norm": 1.2734375,
      "learning_rate": 7.919927439374278e-05,
      "loss": 4.3882,
      "mean_token_accuracy": 0.23165587186813355,
      "num_tokens": 18671949.0,
      "step": 8450
    },
    {
      "entropy": 5.150627708435058,
      "epoch": 8.523449319213313,
      "grad_norm": 1.2421875,
      "learning_rate": 7.900415992631162e-05,
      "loss": 4.3871,
      "mean_token_accuracy": 0.23084062784910203,
      "num_tokens": 18682344.0,
      "step": 8455
    },
    {
      "entropy": 5.126788759231568,
      "epoch": 8.528492183560262,
      "grad_norm": 1.21875,
      "learning_rate": 7.880965461847261e-05,
      "loss": 4.3641,
      "mean_token_accuracy": 0.2370360180735588,
      "num_tokens": 18694088.0,
      "step": 8460
    },
    {
      "entropy": 5.192434072494507,
      "epoch": 8.533535047907211,
      "grad_norm": 1.3671875,
      "learning_rate": 7.86157590747528e-05,
      "loss": 4.4626,
      "mean_token_accuracy": 0.21858568787574767,
      "num_tokens": 18705317.0,
      "step": 8465
    },
    {
      "entropy": 5.1963306903839115,
      "epoch": 8.53857791225416,
      "grad_norm": 1.3125,
      "learning_rate": 7.8422473897784e-05,
      "loss": 4.4908,
      "mean_token_accuracy": 0.22368544936180115,
      "num_tokens": 18715569.0,
      "step": 8470
    },
    {
      "entropy": 5.1619666576385494,
      "epoch": 8.54362077660111,
      "grad_norm": 1.296875,
      "learning_rate": 7.822979968830089e-05,
      "loss": 4.3681,
      "mean_token_accuracy": 0.23752204775810243,
      "num_tokens": 18726037.0,
      "step": 8475
    },
    {
      "entropy": 5.201433610916138,
      "epoch": 8.548663640948059,
      "grad_norm": 1.3046875,
      "learning_rate": 7.803773704513953e-05,
      "loss": 4.4238,
      "mean_token_accuracy": 0.22067201286554336,
      "num_tokens": 18736425.0,
      "step": 8480
    },
    {
      "entropy": 5.143138265609741,
      "epoch": 8.553706505295008,
      "grad_norm": 1.3203125,
      "learning_rate": 7.784628656523488e-05,
      "loss": 4.4001,
      "mean_token_accuracy": 0.22622804045677186,
      "num_tokens": 18746171.0,
      "step": 8485
    },
    {
      "entropy": 5.180696678161621,
      "epoch": 8.558749369641957,
      "grad_norm": 1.1484375,
      "learning_rate": 7.76554488436195e-05,
      "loss": 4.3839,
      "mean_token_accuracy": 0.22663193196058273,
      "num_tokens": 18757509.0,
      "step": 8490
    },
    {
      "entropy": 5.175520706176758,
      "epoch": 8.563792233988906,
      "grad_norm": 1.484375,
      "learning_rate": 7.746522447342148e-05,
      "loss": 4.4263,
      "mean_token_accuracy": 0.23478810489177704,
      "num_tokens": 18767416.0,
      "step": 8495
    },
    {
      "entropy": 5.1783312320709225,
      "epoch": 8.568835098335855,
      "grad_norm": 1.15625,
      "learning_rate": 7.727561404586255e-05,
      "loss": 4.4459,
      "mean_token_accuracy": 0.2255529448390007,
      "num_tokens": 18779497.0,
      "step": 8500
    },
    {
      "epoch": 8.568835098335855,
      "eval_entropy": 5.114228140610169,
      "eval_loss": 5.696971893310547,
      "eval_mean_token_accuracy": 0.1545760200830513,
      "eval_num_tokens": 18779497.0,
      "eval_runtime": 2.6336,
      "eval_samples_per_second": 1191.544,
      "eval_steps_per_second": 149.228,
      "step": 8500
    },
    {
      "entropy": 5.198412370681763,
      "epoch": 8.573877962682804,
      "grad_norm": 1.265625,
      "learning_rate": 7.708661815025619e-05,
      "loss": 4.4608,
      "mean_token_accuracy": 0.22483488619327546,
      "num_tokens": 18790524.0,
      "step": 8505
    },
    {
      "entropy": 5.1512613773345945,
      "epoch": 8.578920827029753,
      "grad_norm": 1.3046875,
      "learning_rate": 7.689823737400617e-05,
      "loss": 4.3919,
      "mean_token_accuracy": 0.22180528938770294,
      "num_tokens": 18801656.0,
      "step": 8510
    },
    {
      "entropy": 5.164031934738159,
      "epoch": 8.583963691376702,
      "grad_norm": 1.40625,
      "learning_rate": 7.671047230260412e-05,
      "loss": 4.4194,
      "mean_token_accuracy": 0.22609245777130127,
      "num_tokens": 18812234.0,
      "step": 8515
    },
    {
      "entropy": 5.157821464538574,
      "epoch": 8.589006555723651,
      "grad_norm": 1.203125,
      "learning_rate": 7.652332351962836e-05,
      "loss": 4.3988,
      "mean_token_accuracy": 0.2258038565516472,
      "num_tokens": 18822727.0,
      "step": 8520
    },
    {
      "entropy": 5.209461259841919,
      "epoch": 8.5940494200706,
      "grad_norm": 1.359375,
      "learning_rate": 7.633679160674144e-05,
      "loss": 4.4556,
      "mean_token_accuracy": 0.221569986641407,
      "num_tokens": 18833510.0,
      "step": 8525
    },
    {
      "entropy": 5.211916065216064,
      "epoch": 8.59909228441755,
      "grad_norm": 1.375,
      "learning_rate": 7.615087714368902e-05,
      "loss": 4.4346,
      "mean_token_accuracy": 0.22653616666793824,
      "num_tokens": 18845660.0,
      "step": 8530
    },
    {
      "entropy": 5.139431476593018,
      "epoch": 8.604135148764499,
      "grad_norm": 1.3515625,
      "learning_rate": 7.59655807082975e-05,
      "loss": 4.2986,
      "mean_token_accuracy": 0.24049947857856752,
      "num_tokens": 18855801.0,
      "step": 8535
    },
    {
      "entropy": 5.195802831649781,
      "epoch": 8.609178013111448,
      "grad_norm": 1.265625,
      "learning_rate": 7.578090287647242e-05,
      "loss": 4.4673,
      "mean_token_accuracy": 0.22765713334083557,
      "num_tokens": 18867402.0,
      "step": 8540
    },
    {
      "entropy": 5.222857141494751,
      "epoch": 8.614220877458397,
      "grad_norm": 1.34375,
      "learning_rate": 7.559684422219683e-05,
      "loss": 4.5466,
      "mean_token_accuracy": 0.2140881225466728,
      "num_tokens": 18878833.0,
      "step": 8545
    },
    {
      "entropy": 5.159449100494385,
      "epoch": 8.619263741805346,
      "grad_norm": 1.359375,
      "learning_rate": 7.541340531752916e-05,
      "loss": 4.4194,
      "mean_token_accuracy": 0.22314474433660508,
      "num_tokens": 18890002.0,
      "step": 8550
    },
    {
      "entropy": 5.150857257843017,
      "epoch": 8.624306606152295,
      "grad_norm": 1.3359375,
      "learning_rate": 7.523058673260183e-05,
      "loss": 4.4056,
      "mean_token_accuracy": 0.23074856847524644,
      "num_tokens": 18901312.0,
      "step": 8555
    },
    {
      "entropy": 5.174584865570068,
      "epoch": 8.629349470499244,
      "grad_norm": 1.3515625,
      "learning_rate": 7.50483890356192e-05,
      "loss": 4.383,
      "mean_token_accuracy": 0.22923626005649567,
      "num_tokens": 18910750.0,
      "step": 8560
    },
    {
      "entropy": 5.127251768112183,
      "epoch": 8.634392334846194,
      "grad_norm": 1.34375,
      "learning_rate": 7.486681279285598e-05,
      "loss": 4.3739,
      "mean_token_accuracy": 0.23146091252565384,
      "num_tokens": 18923143.0,
      "step": 8565
    },
    {
      "entropy": 5.176032543182373,
      "epoch": 8.63943519919314,
      "grad_norm": 1.328125,
      "learning_rate": 7.468585856865522e-05,
      "loss": 4.4004,
      "mean_token_accuracy": 0.228582701086998,
      "num_tokens": 18934569.0,
      "step": 8570
    },
    {
      "entropy": 5.119245862960815,
      "epoch": 8.64447806354009,
      "grad_norm": 1.34375,
      "learning_rate": 7.45055269254269e-05,
      "loss": 4.4096,
      "mean_token_accuracy": 0.2215472161769867,
      "num_tokens": 18946359.0,
      "step": 8575
    },
    {
      "entropy": 5.143156480789185,
      "epoch": 8.649520927887039,
      "grad_norm": 1.2265625,
      "learning_rate": 7.432581842364585e-05,
      "loss": 4.4081,
      "mean_token_accuracy": 0.22314869463443757,
      "num_tokens": 18957769.0,
      "step": 8580
    },
    {
      "entropy": 5.154001379013062,
      "epoch": 8.654563792233988,
      "grad_norm": 1.234375,
      "learning_rate": 7.414673362185031e-05,
      "loss": 4.3608,
      "mean_token_accuracy": 0.23021700084209443,
      "num_tokens": 18970816.0,
      "step": 8585
    },
    {
      "entropy": 5.150351333618164,
      "epoch": 8.659606656580937,
      "grad_norm": 1.359375,
      "learning_rate": 7.396827307663988e-05,
      "loss": 4.3902,
      "mean_token_accuracy": 0.22305918186903,
      "num_tokens": 18981447.0,
      "step": 8590
    },
    {
      "entropy": 5.198001909255981,
      "epoch": 8.664649520927886,
      "grad_norm": 1.296875,
      "learning_rate": 7.379043734267421e-05,
      "loss": 4.4524,
      "mean_token_accuracy": 0.2172454282641411,
      "num_tokens": 18992132.0,
      "step": 8595
    },
    {
      "entropy": 5.145677614212036,
      "epoch": 8.669692385274836,
      "grad_norm": 1.328125,
      "learning_rate": 7.36132269726707e-05,
      "loss": 4.3984,
      "mean_token_accuracy": 0.2245606243610382,
      "num_tokens": 19002993.0,
      "step": 8600
    },
    {
      "entropy": 5.184981918334961,
      "epoch": 8.674735249621785,
      "grad_norm": 1.203125,
      "learning_rate": 7.343664251740342e-05,
      "loss": 4.4237,
      "mean_token_accuracy": 0.22328116595745087,
      "num_tokens": 19014255.0,
      "step": 8605
    },
    {
      "entropy": 5.10263729095459,
      "epoch": 8.679778113968734,
      "grad_norm": 1.3828125,
      "learning_rate": 7.326068452570078e-05,
      "loss": 4.3493,
      "mean_token_accuracy": 0.22412345558404922,
      "num_tokens": 19024200.0,
      "step": 8610
    },
    {
      "entropy": 5.202133274078369,
      "epoch": 8.684820978315683,
      "grad_norm": 1.46875,
      "learning_rate": 7.30853535444444e-05,
      "loss": 4.3595,
      "mean_token_accuracy": 0.22763469219207763,
      "num_tokens": 19033705.0,
      "step": 8615
    },
    {
      "entropy": 5.152081871032715,
      "epoch": 8.689863842662632,
      "grad_norm": 1.28125,
      "learning_rate": 7.29106501185669e-05,
      "loss": 4.4715,
      "mean_token_accuracy": 0.2131981298327446,
      "num_tokens": 19045452.0,
      "step": 8620
    },
    {
      "entropy": 5.143366813659668,
      "epoch": 8.694906707009581,
      "grad_norm": 1.15625,
      "learning_rate": 7.273657479105068e-05,
      "loss": 4.4072,
      "mean_token_accuracy": 0.22738747000694276,
      "num_tokens": 19057934.0,
      "step": 8625
    },
    {
      "entropy": 5.09853229522705,
      "epoch": 8.69994957135653,
      "grad_norm": 1.328125,
      "learning_rate": 7.256312810292576e-05,
      "loss": 4.3282,
      "mean_token_accuracy": 0.23469046503305435,
      "num_tokens": 19067989.0,
      "step": 8630
    },
    {
      "entropy": 5.144941997528076,
      "epoch": 8.70499243570348,
      "grad_norm": 1.5234375,
      "learning_rate": 7.23903105932685e-05,
      "loss": 4.3499,
      "mean_token_accuracy": 0.23052453398704528,
      "num_tokens": 19078070.0,
      "step": 8635
    },
    {
      "entropy": 5.127585077285767,
      "epoch": 8.710035300050428,
      "grad_norm": 1.234375,
      "learning_rate": 7.221812279919978e-05,
      "loss": 4.4184,
      "mean_token_accuracy": 0.228779336810112,
      "num_tokens": 19089391.0,
      "step": 8640
    },
    {
      "entropy": 5.107707738876343,
      "epoch": 8.715078164397378,
      "grad_norm": 1.375,
      "learning_rate": 7.20465652558831e-05,
      "loss": 4.3867,
      "mean_token_accuracy": 0.23164014816284179,
      "num_tokens": 19100101.0,
      "step": 8645
    },
    {
      "entropy": 5.068668079376221,
      "epoch": 8.720121028744327,
      "grad_norm": 1.078125,
      "learning_rate": 7.187563849652332e-05,
      "loss": 4.3386,
      "mean_token_accuracy": 0.23821428269147873,
      "num_tokens": 19112108.0,
      "step": 8650
    },
    {
      "entropy": 5.117764568328857,
      "epoch": 8.725163893091276,
      "grad_norm": 1.34375,
      "learning_rate": 7.170534305236479e-05,
      "loss": 4.3221,
      "mean_token_accuracy": 0.2398919716477394,
      "num_tokens": 19122375.0,
      "step": 8655
    },
    {
      "entropy": 5.113310813903809,
      "epoch": 8.730206757438225,
      "grad_norm": 1.140625,
      "learning_rate": 7.153567945268963e-05,
      "loss": 4.4132,
      "mean_token_accuracy": 0.22768631875514983,
      "num_tokens": 19134863.0,
      "step": 8660
    },
    {
      "entropy": 5.158536052703857,
      "epoch": 8.735249621785174,
      "grad_norm": 1.15625,
      "learning_rate": 7.136664822481624e-05,
      "loss": 4.4379,
      "mean_token_accuracy": 0.2164291113615036,
      "num_tokens": 19145898.0,
      "step": 8665
    },
    {
      "entropy": 5.182373142242431,
      "epoch": 8.740292486132123,
      "grad_norm": 1.3203125,
      "learning_rate": 7.119824989409761e-05,
      "loss": 4.3946,
      "mean_token_accuracy": 0.23052615523338318,
      "num_tokens": 19156280.0,
      "step": 8670
    },
    {
      "entropy": 5.170133638381958,
      "epoch": 8.745335350479072,
      "grad_norm": 1.28125,
      "learning_rate": 7.103048498391954e-05,
      "loss": 4.3998,
      "mean_token_accuracy": 0.23702604323625565,
      "num_tokens": 19167494.0,
      "step": 8675
    },
    {
      "entropy": 5.1524364948272705,
      "epoch": 8.750378214826021,
      "grad_norm": 1.296875,
      "learning_rate": 7.086335401569935e-05,
      "loss": 4.4144,
      "mean_token_accuracy": 0.21907378882169723,
      "num_tokens": 19179350.0,
      "step": 8680
    },
    {
      "entropy": 5.141133260726929,
      "epoch": 8.75542107917297,
      "grad_norm": 1.375,
      "learning_rate": 7.06968575088838e-05,
      "loss": 4.3956,
      "mean_token_accuracy": 0.23540458977222442,
      "num_tokens": 19189776.0,
      "step": 8685
    },
    {
      "entropy": 5.116704034805298,
      "epoch": 8.76046394351992,
      "grad_norm": 1.265625,
      "learning_rate": 7.053099598094804e-05,
      "loss": 4.3255,
      "mean_token_accuracy": 0.23617296516895295,
      "num_tokens": 19201297.0,
      "step": 8690
    },
    {
      "entropy": 5.117445373535157,
      "epoch": 8.765506807866869,
      "grad_norm": 1.203125,
      "learning_rate": 7.036576994739343e-05,
      "loss": 4.4418,
      "mean_token_accuracy": 0.21940775364637374,
      "num_tokens": 19212505.0,
      "step": 8695
    },
    {
      "entropy": 5.107796335220337,
      "epoch": 8.770549672213818,
      "grad_norm": 1.328125,
      "learning_rate": 7.020117992174638e-05,
      "loss": 4.3527,
      "mean_token_accuracy": 0.23021068572998046,
      "num_tokens": 19223956.0,
      "step": 8700
    },
    {
      "entropy": 5.153073501586914,
      "epoch": 8.775592536560767,
      "grad_norm": 1.25,
      "learning_rate": 7.003722641555635e-05,
      "loss": 4.4349,
      "mean_token_accuracy": 0.22498720288276672,
      "num_tokens": 19234690.0,
      "step": 8705
    },
    {
      "entropy": 5.151028108596802,
      "epoch": 8.780635400907716,
      "grad_norm": 1.2734375,
      "learning_rate": 6.987390993839476e-05,
      "loss": 4.3366,
      "mean_token_accuracy": 0.24349051117897033,
      "num_tokens": 19245127.0,
      "step": 8710
    },
    {
      "entropy": 5.137298631668091,
      "epoch": 8.785678265254665,
      "grad_norm": 1.3125,
      "learning_rate": 6.971123099785281e-05,
      "loss": 4.3637,
      "mean_token_accuracy": 0.23706451207399368,
      "num_tokens": 19255587.0,
      "step": 8715
    },
    {
      "entropy": 5.19605450630188,
      "epoch": 8.790721129601614,
      "grad_norm": 1.1640625,
      "learning_rate": 6.954919009954061e-05,
      "loss": 4.5076,
      "mean_token_accuracy": 0.2219483584165573,
      "num_tokens": 19267862.0,
      "step": 8720
    },
    {
      "entropy": 5.141691780090332,
      "epoch": 8.795763993948563,
      "grad_norm": 1.3125,
      "learning_rate": 6.938778774708488e-05,
      "loss": 4.3639,
      "mean_token_accuracy": 0.22846208661794662,
      "num_tokens": 19278106.0,
      "step": 8725
    },
    {
      "entropy": 5.140495920181275,
      "epoch": 8.800806858295513,
      "grad_norm": 1.3046875,
      "learning_rate": 6.922702444212792e-05,
      "loss": 4.427,
      "mean_token_accuracy": 0.2231375440955162,
      "num_tokens": 19288832.0,
      "step": 8730
    },
    {
      "entropy": 5.114597272872925,
      "epoch": 8.805849722642462,
      "grad_norm": 1.3515625,
      "learning_rate": 6.906690068432568e-05,
      "loss": 4.3281,
      "mean_token_accuracy": 0.23094843924045563,
      "num_tokens": 19300048.0,
      "step": 8735
    },
    {
      "entropy": 5.147270727157593,
      "epoch": 8.81089258698941,
      "grad_norm": 1.234375,
      "learning_rate": 6.89074169713466e-05,
      "loss": 4.4878,
      "mean_token_accuracy": 0.22836720645427705,
      "num_tokens": 19311882.0,
      "step": 8740
    },
    {
      "entropy": 5.188826274871826,
      "epoch": 8.815935451336358,
      "grad_norm": 1.28125,
      "learning_rate": 6.87485737988697e-05,
      "loss": 4.3972,
      "mean_token_accuracy": 0.22525349110364914,
      "num_tokens": 19323195.0,
      "step": 8745
    },
    {
      "entropy": 5.074718999862671,
      "epoch": 8.820978315683309,
      "grad_norm": 1.09375,
      "learning_rate": 6.859037166058323e-05,
      "loss": 4.3302,
      "mean_token_accuracy": 0.23585810214281083,
      "num_tokens": 19334520.0,
      "step": 8750
    },
    {
      "entropy": 5.11641526222229,
      "epoch": 8.826021180030256,
      "grad_norm": 1.34375,
      "learning_rate": 6.843281104818313e-05,
      "loss": 4.3985,
      "mean_token_accuracy": 0.22927930057048798,
      "num_tokens": 19345550.0,
      "step": 8755
    },
    {
      "entropy": 5.13916015625,
      "epoch": 8.831064044377205,
      "grad_norm": 1.2890625,
      "learning_rate": 6.827589245137135e-05,
      "loss": 4.3868,
      "mean_token_accuracy": 0.22718503177165986,
      "num_tokens": 19355822.0,
      "step": 8760
    },
    {
      "entropy": 5.123105859756469,
      "epoch": 8.836106908724155,
      "grad_norm": 1.3203125,
      "learning_rate": 6.811961635785456e-05,
      "loss": 4.3964,
      "mean_token_accuracy": 0.2223242536187172,
      "num_tokens": 19365974.0,
      "step": 8765
    },
    {
      "entropy": 5.203814315795898,
      "epoch": 8.841149773071104,
      "grad_norm": 1.453125,
      "learning_rate": 6.796398325334238e-05,
      "loss": 4.4095,
      "mean_token_accuracy": 0.23311451226472854,
      "num_tokens": 19376341.0,
      "step": 8770
    },
    {
      "entropy": 5.194783353805542,
      "epoch": 8.846192637418053,
      "grad_norm": 1.1953125,
      "learning_rate": 6.78089936215462e-05,
      "loss": 4.4203,
      "mean_token_accuracy": 0.2223714143037796,
      "num_tokens": 19387027.0,
      "step": 8775
    },
    {
      "entropy": 5.170251655578613,
      "epoch": 8.851235501765002,
      "grad_norm": 1.3359375,
      "learning_rate": 6.765464794417723e-05,
      "loss": 4.393,
      "mean_token_accuracy": 0.22215620428323746,
      "num_tokens": 19397455.0,
      "step": 8780
    },
    {
      "entropy": 5.205050230026245,
      "epoch": 8.856278366111951,
      "grad_norm": 1.2578125,
      "learning_rate": 6.750094670094558e-05,
      "loss": 4.4649,
      "mean_token_accuracy": 0.2185963973402977,
      "num_tokens": 19407311.0,
      "step": 8785
    },
    {
      "entropy": 5.158578157424927,
      "epoch": 8.8613212304589,
      "grad_norm": 1.234375,
      "learning_rate": 6.734789036955813e-05,
      "loss": 4.4291,
      "mean_token_accuracy": 0.23185295462608338,
      "num_tokens": 19418948.0,
      "step": 8790
    },
    {
      "entropy": 5.162908983230591,
      "epoch": 8.86636409480585,
      "grad_norm": 1.3125,
      "learning_rate": 6.719547942571755e-05,
      "loss": 4.3744,
      "mean_token_accuracy": 0.2284393325448036,
      "num_tokens": 19431223.0,
      "step": 8795
    },
    {
      "entropy": 5.132456207275391,
      "epoch": 8.871406959152798,
      "grad_norm": 1.2109375,
      "learning_rate": 6.704371434312048e-05,
      "loss": 4.3741,
      "mean_token_accuracy": 0.23618563860654831,
      "num_tokens": 19444685.0,
      "step": 8800
    },
    {
      "entropy": 5.150185632705688,
      "epoch": 8.876449823499748,
      "grad_norm": 1.3203125,
      "learning_rate": 6.689259559345641e-05,
      "loss": 4.3711,
      "mean_token_accuracy": 0.22950804382562637,
      "num_tokens": 19454952.0,
      "step": 8805
    },
    {
      "entropy": 5.104757690429688,
      "epoch": 8.881492687846697,
      "grad_norm": 1.1640625,
      "learning_rate": 6.674212364640577e-05,
      "loss": 4.3864,
      "mean_token_accuracy": 0.2232255220413208,
      "num_tokens": 19465841.0,
      "step": 8810
    },
    {
      "entropy": 5.171678733825684,
      "epoch": 8.886535552193646,
      "grad_norm": 1.3125,
      "learning_rate": 6.659229896963899e-05,
      "loss": 4.4889,
      "mean_token_accuracy": 0.2239171639084816,
      "num_tokens": 19477483.0,
      "step": 8815
    },
    {
      "entropy": 5.1386473178863525,
      "epoch": 8.891578416540595,
      "grad_norm": 1.1796875,
      "learning_rate": 6.644312202881452e-05,
      "loss": 4.4171,
      "mean_token_accuracy": 0.22187758535146712,
      "num_tokens": 19489660.0,
      "step": 8820
    },
    {
      "entropy": 5.1552976131439205,
      "epoch": 8.896621280887544,
      "grad_norm": 1.2109375,
      "learning_rate": 6.629459328757781e-05,
      "loss": 4.4137,
      "mean_token_accuracy": 0.22751563042402267,
      "num_tokens": 19500634.0,
      "step": 8825
    },
    {
      "entropy": 5.141992902755737,
      "epoch": 8.901664145234493,
      "grad_norm": 1.4140625,
      "learning_rate": 6.614671320755953e-05,
      "loss": 4.3845,
      "mean_token_accuracy": 0.22838451862335205,
      "num_tokens": 19511687.0,
      "step": 8830
    },
    {
      "entropy": 5.155193662643432,
      "epoch": 8.906707009581442,
      "grad_norm": 1.125,
      "learning_rate": 6.599948224837446e-05,
      "loss": 4.3838,
      "mean_token_accuracy": 0.22182397097349166,
      "num_tokens": 19523920.0,
      "step": 8835
    },
    {
      "entropy": 5.174787282943726,
      "epoch": 8.911749873928391,
      "grad_norm": 1.21875,
      "learning_rate": 6.585290086761972e-05,
      "loss": 4.4087,
      "mean_token_accuracy": 0.2346874475479126,
      "num_tokens": 19534759.0,
      "step": 8840
    },
    {
      "entropy": 5.16672830581665,
      "epoch": 8.91679273827534,
      "grad_norm": 1.1875,
      "learning_rate": 6.570696952087368e-05,
      "loss": 4.4005,
      "mean_token_accuracy": 0.22461390048265456,
      "num_tokens": 19545855.0,
      "step": 8845
    },
    {
      "entropy": 5.146234130859375,
      "epoch": 8.92183560262229,
      "grad_norm": 1.1328125,
      "learning_rate": 6.556168866169435e-05,
      "loss": 4.4217,
      "mean_token_accuracy": 0.22424035966396333,
      "num_tokens": 19556267.0,
      "step": 8850
    },
    {
      "entropy": 5.132492637634277,
      "epoch": 8.926878466969239,
      "grad_norm": 1.296875,
      "learning_rate": 6.541705874161795e-05,
      "loss": 4.387,
      "mean_token_accuracy": 0.2302074357867241,
      "num_tokens": 19567321.0,
      "step": 8855
    },
    {
      "entropy": 5.156880521774292,
      "epoch": 8.931921331316188,
      "grad_norm": 1.28125,
      "learning_rate": 6.527308021015771e-05,
      "loss": 4.3801,
      "mean_token_accuracy": 0.22845539450645447,
      "num_tokens": 19578017.0,
      "step": 8860
    },
    {
      "entropy": 5.188018798828125,
      "epoch": 8.936964195663137,
      "grad_norm": 1.296875,
      "learning_rate": 6.512975351480207e-05,
      "loss": 4.4143,
      "mean_token_accuracy": 0.22367270439863204,
      "num_tokens": 19589544.0,
      "step": 8865
    },
    {
      "entropy": 5.174517869949341,
      "epoch": 8.942007060010086,
      "grad_norm": 1.2578125,
      "learning_rate": 6.498707910101383e-05,
      "loss": 4.474,
      "mean_token_accuracy": 0.2193455159664154,
      "num_tokens": 19599782.0,
      "step": 8870
    },
    {
      "entropy": 5.222671461105347,
      "epoch": 8.947049924357035,
      "grad_norm": 1.203125,
      "learning_rate": 6.484505741222824e-05,
      "loss": 4.5064,
      "mean_token_accuracy": 0.21572977155447007,
      "num_tokens": 19611582.0,
      "step": 8875
    },
    {
      "entropy": 5.190328741073609,
      "epoch": 8.952092788703984,
      "grad_norm": 1.5546875,
      "learning_rate": 6.470368888985215e-05,
      "loss": 4.3888,
      "mean_token_accuracy": 0.23282261192798615,
      "num_tokens": 19620381.0,
      "step": 8880
    },
    {
      "entropy": 5.189480972290039,
      "epoch": 8.957135653050933,
      "grad_norm": 1.5234375,
      "learning_rate": 6.456297397326202e-05,
      "loss": 4.4479,
      "mean_token_accuracy": 0.21640007942914963,
      "num_tokens": 19631808.0,
      "step": 8885
    },
    {
      "entropy": 5.143616437911987,
      "epoch": 8.962178517397883,
      "grad_norm": 1.2890625,
      "learning_rate": 6.442291309980321e-05,
      "loss": 4.3892,
      "mean_token_accuracy": 0.22327383011579513,
      "num_tokens": 19643209.0,
      "step": 8890
    },
    {
      "entropy": 5.156438636779785,
      "epoch": 8.967221381744832,
      "grad_norm": 1.171875,
      "learning_rate": 6.4283506704788e-05,
      "loss": 4.4135,
      "mean_token_accuracy": 0.22170841842889785,
      "num_tokens": 19654505.0,
      "step": 8895
    },
    {
      "entropy": 5.1455460548400875,
      "epoch": 8.97226424609178,
      "grad_norm": 1.2890625,
      "learning_rate": 6.414475522149481e-05,
      "loss": 4.3896,
      "mean_token_accuracy": 0.22402234375476837,
      "num_tokens": 19665566.0,
      "step": 8900
    },
    {
      "entropy": 5.116247892379761,
      "epoch": 8.97730711043873,
      "grad_norm": 1.296875,
      "learning_rate": 6.400665908116631e-05,
      "loss": 4.4014,
      "mean_token_accuracy": 0.2224682718515396,
      "num_tokens": 19677222.0,
      "step": 8905
    },
    {
      "entropy": 5.102776432037354,
      "epoch": 8.982349974785679,
      "grad_norm": 1.21875,
      "learning_rate": 6.386921871300865e-05,
      "loss": 4.3896,
      "mean_token_accuracy": 0.2324525833129883,
      "num_tokens": 19689413.0,
      "step": 8910
    },
    {
      "entropy": 5.182309961318969,
      "epoch": 8.987392839132628,
      "grad_norm": 1.28125,
      "learning_rate": 6.373243454418952e-05,
      "loss": 4.417,
      "mean_token_accuracy": 0.22739481776952744,
      "num_tokens": 19701417.0,
      "step": 8915
    },
    {
      "entropy": 5.1530022621154785,
      "epoch": 8.992435703479577,
      "grad_norm": 1.4453125,
      "learning_rate": 6.35963069998374e-05,
      "loss": 4.4137,
      "mean_token_accuracy": 0.23058675825595856,
      "num_tokens": 19711042.0,
      "step": 8920
    },
    {
      "entropy": 5.174558210372925,
      "epoch": 8.997478567826526,
      "grad_norm": 1.484375,
      "learning_rate": 6.346083650303975e-05,
      "loss": 4.3634,
      "mean_token_accuracy": 0.2283007651567459,
      "num_tokens": 19720160.0,
      "step": 8925
    },
    {
      "entropy": 5.114773485395643,
      "epoch": 9.00201714573878,
      "grad_norm": 1.2109375,
      "learning_rate": 6.332602347484202e-05,
      "loss": 4.3749,
      "mean_token_accuracy": 0.23947841591305202,
      "num_tokens": 19728930.0,
      "step": 8930
    },
    {
      "entropy": 5.1582419872283936,
      "epoch": 9.007060010085729,
      "grad_norm": 1.1875,
      "learning_rate": 6.31918683342462e-05,
      "loss": 4.3781,
      "mean_token_accuracy": 0.230475752055645,
      "num_tokens": 19739541.0,
      "step": 8935
    },
    {
      "entropy": 5.18987545967102,
      "epoch": 9.012102874432678,
      "grad_norm": 1.1640625,
      "learning_rate": 6.305837149820959e-05,
      "loss": 4.4478,
      "mean_token_accuracy": 0.22493044584989547,
      "num_tokens": 19751494.0,
      "step": 8940
    },
    {
      "entropy": 5.214781904220581,
      "epoch": 9.017145738779627,
      "grad_norm": 1.2265625,
      "learning_rate": 6.29255333816434e-05,
      "loss": 4.4505,
      "mean_token_accuracy": 0.23081174939870835,
      "num_tokens": 19763075.0,
      "step": 8945
    },
    {
      "entropy": 5.18967113494873,
      "epoch": 9.022188603126576,
      "grad_norm": 1.359375,
      "learning_rate": 6.279335439741164e-05,
      "loss": 4.397,
      "mean_token_accuracy": 0.2241222620010376,
      "num_tokens": 19773363.0,
      "step": 8950
    },
    {
      "entropy": 5.155258846282959,
      "epoch": 9.027231467473525,
      "grad_norm": 1.2734375,
      "learning_rate": 6.266183495632954e-05,
      "loss": 4.3894,
      "mean_token_accuracy": 0.2371391162276268,
      "num_tokens": 19784480.0,
      "step": 8955
    },
    {
      "entropy": 5.118967533111572,
      "epoch": 9.032274331820474,
      "grad_norm": 1.390625,
      "learning_rate": 6.253097546716264e-05,
      "loss": 4.3477,
      "mean_token_accuracy": 0.2303277552127838,
      "num_tokens": 19795494.0,
      "step": 8960
    },
    {
      "entropy": 5.134348964691162,
      "epoch": 9.037317196167423,
      "grad_norm": 1.453125,
      "learning_rate": 6.240077633662519e-05,
      "loss": 4.2403,
      "mean_token_accuracy": 0.2444775640964508,
      "num_tokens": 19805132.0,
      "step": 8965
    },
    {
      "entropy": 5.191956663131714,
      "epoch": 9.042360060514373,
      "grad_norm": 1.359375,
      "learning_rate": 6.227123796937905e-05,
      "loss": 4.3008,
      "mean_token_accuracy": 0.23201472759246827,
      "num_tokens": 19814429.0,
      "step": 8970
    },
    {
      "entropy": 5.149028015136719,
      "epoch": 9.047402924861322,
      "grad_norm": 1.4453125,
      "learning_rate": 6.214236076803264e-05,
      "loss": 4.4346,
      "mean_token_accuracy": 0.21843936443328857,
      "num_tokens": 19825033.0,
      "step": 8975
    },
    {
      "entropy": 5.163254117965698,
      "epoch": 9.05244578920827,
      "grad_norm": 1.328125,
      "learning_rate": 6.201414513313907e-05,
      "loss": 4.3576,
      "mean_token_accuracy": 0.23096331357955932,
      "num_tokens": 19836157.0,
      "step": 8980
    },
    {
      "entropy": 5.179744911193848,
      "epoch": 9.05748865355522,
      "grad_norm": 1.265625,
      "learning_rate": 6.188659146319562e-05,
      "loss": 4.4237,
      "mean_token_accuracy": 0.22397961914539338,
      "num_tokens": 19847828.0,
      "step": 8985
    },
    {
      "entropy": 5.146542406082153,
      "epoch": 9.062531517902169,
      "grad_norm": 1.34375,
      "learning_rate": 6.175970015464196e-05,
      "loss": 4.4072,
      "mean_token_accuracy": 0.2306445375084877,
      "num_tokens": 19858341.0,
      "step": 8990
    },
    {
      "entropy": 5.15264663696289,
      "epoch": 9.067574382249118,
      "grad_norm": 1.1953125,
      "learning_rate": 6.163347160185929e-05,
      "loss": 4.3341,
      "mean_token_accuracy": 0.23283738344907762,
      "num_tokens": 19870023.0,
      "step": 8995
    },
    {
      "entropy": 5.128273391723633,
      "epoch": 9.072617246596067,
      "grad_norm": 1.3046875,
      "learning_rate": 6.150790619716873e-05,
      "loss": 4.3227,
      "mean_token_accuracy": 0.2352721333503723,
      "num_tokens": 19882554.0,
      "step": 9000
    },
    {
      "epoch": 9.072617246596067,
      "eval_entropy": 5.116135515936155,
      "eval_loss": 5.694945335388184,
      "eval_mean_token_accuracy": 0.15426953756854733,
      "eval_num_tokens": 19882554.0,
      "eval_runtime": 2.6443,
      "eval_samples_per_second": 1186.713,
      "eval_steps_per_second": 148.623,
      "step": 9000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9910,
  "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": 3.3197500204416e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}