ind-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407/checkpoint-11000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.620689655172415,
  "eval_steps": 500,
  "global_step": 11000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.113325738906861,
      "epoch": 0.003918495297805642,
      "grad_norm": 1.21875,
      "learning_rate": 2e-06,
      "loss": 6.8669,
      "mean_token_accuracy": 0.09941852316260338,
      "num_tokens": 9174.0,
      "step": 5
    },
    {
      "entropy": 7.112642478942871,
      "epoch": 0.007836990595611285,
      "grad_norm": 1.2109375,
      "learning_rate": 4.5e-06,
      "loss": 6.8445,
      "mean_token_accuracy": 0.09862779080867767,
      "num_tokens": 17790.0,
      "step": 10
    },
    {
      "entropy": 7.158182430267334,
      "epoch": 0.011755485893416929,
      "grad_norm": 1.234375,
      "learning_rate": 7e-06,
      "loss": 6.9709,
      "mean_token_accuracy": 0.0985221654176712,
      "num_tokens": 25850.0,
      "step": 15
    },
    {
      "entropy": 7.190377140045166,
      "epoch": 0.01567398119122257,
      "grad_norm": 1.3203125,
      "learning_rate": 9.5e-06,
      "loss": 7.0374,
      "mean_token_accuracy": 0.09146212115883827,
      "num_tokens": 35194.0,
      "step": 20
    },
    {
      "entropy": 7.184538173675537,
      "epoch": 0.019592476489028215,
      "grad_norm": 1.234375,
      "learning_rate": 1.2e-05,
      "loss": 6.9985,
      "mean_token_accuracy": 0.09104539602994918,
      "num_tokens": 44218.0,
      "step": 25
    },
    {
      "entropy": 7.137639093399048,
      "epoch": 0.023510971786833857,
      "grad_norm": 1.28125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.942,
      "mean_token_accuracy": 0.09604569971561432,
      "num_tokens": 53397.0,
      "step": 30
    },
    {
      "entropy": 7.152150678634643,
      "epoch": 0.0274294670846395,
      "grad_norm": 1.1796875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.8465,
      "mean_token_accuracy": 0.09854738116264343,
      "num_tokens": 62749.0,
      "step": 35
    },
    {
      "entropy": 7.131413555145263,
      "epoch": 0.03134796238244514,
      "grad_norm": 1.15625,
      "learning_rate": 1.95e-05,
      "loss": 6.8427,
      "mean_token_accuracy": 0.10017006173729896,
      "num_tokens": 71721.0,
      "step": 40
    },
    {
      "entropy": 7.150160312652588,
      "epoch": 0.03526645768025078,
      "grad_norm": 1.2265625,
      "learning_rate": 2.2e-05,
      "loss": 6.8265,
      "mean_token_accuracy": 0.1048971325159073,
      "num_tokens": 79844.0,
      "step": 45
    },
    {
      "entropy": 7.14600887298584,
      "epoch": 0.03918495297805643,
      "grad_norm": 1.140625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.8045,
      "mean_token_accuracy": 0.10030318275094033,
      "num_tokens": 88866.0,
      "step": 50
    },
    {
      "entropy": 7.112334775924682,
      "epoch": 0.04310344827586207,
      "grad_norm": 1.28125,
      "learning_rate": 2.7e-05,
      "loss": 6.8137,
      "mean_token_accuracy": 0.10082174241542816,
      "num_tokens": 97918.0,
      "step": 55
    },
    {
      "entropy": 7.0919252872467045,
      "epoch": 0.047021943573667714,
      "grad_norm": 1.109375,
      "learning_rate": 2.95e-05,
      "loss": 6.8353,
      "mean_token_accuracy": 0.10005640611052513,
      "num_tokens": 107043.0,
      "step": 60
    },
    {
      "entropy": 7.190604305267334,
      "epoch": 0.050940438871473356,
      "grad_norm": 1.2421875,
      "learning_rate": 3.2e-05,
      "loss": 6.8968,
      "mean_token_accuracy": 0.0979295790195465,
      "num_tokens": 116000.0,
      "step": 65
    },
    {
      "entropy": 7.273019075393677,
      "epoch": 0.054858934169279,
      "grad_norm": 1.1640625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.9081,
      "mean_token_accuracy": 0.10225948318839073,
      "num_tokens": 124559.0,
      "step": 70
    },
    {
      "entropy": 7.21258225440979,
      "epoch": 0.05877742946708464,
      "grad_norm": 1.2734375,
      "learning_rate": 3.7e-05,
      "loss": 6.8128,
      "mean_token_accuracy": 0.10783266574144364,
      "num_tokens": 132868.0,
      "step": 75
    },
    {
      "entropy": 7.171340751647949,
      "epoch": 0.06269592476489028,
      "grad_norm": 1.15625,
      "learning_rate": 3.95e-05,
      "loss": 6.9077,
      "mean_token_accuracy": 0.09100360721349716,
      "num_tokens": 141286.0,
      "step": 80
    },
    {
      "entropy": 7.0822936534881595,
      "epoch": 0.06661442006269593,
      "grad_norm": 1.15625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.8652,
      "mean_token_accuracy": 0.10414223074913025,
      "num_tokens": 150406.0,
      "step": 85
    },
    {
      "entropy": 7.177085208892822,
      "epoch": 0.07053291536050156,
      "grad_norm": 1.296875,
      "learning_rate": 4.45e-05,
      "loss": 6.913,
      "mean_token_accuracy": 0.09742300063371659,
      "num_tokens": 158770.0,
      "step": 90
    },
    {
      "entropy": 7.174734401702881,
      "epoch": 0.07445141065830721,
      "grad_norm": 1.296875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.8071,
      "mean_token_accuracy": 0.1039510190486908,
      "num_tokens": 167763.0,
      "step": 95
    },
    {
      "entropy": 7.1777242660522464,
      "epoch": 0.07836990595611286,
      "grad_norm": 1.03125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.8164,
      "mean_token_accuracy": 0.10320222228765488,
      "num_tokens": 177306.0,
      "step": 100
    },
    {
      "entropy": 7.103335762023926,
      "epoch": 0.0822884012539185,
      "grad_norm": 1.09375,
      "learning_rate": 5.2e-05,
      "loss": 6.8125,
      "mean_token_accuracy": 0.10012169480323792,
      "num_tokens": 186014.0,
      "step": 105
    },
    {
      "entropy": 7.209170961380005,
      "epoch": 0.08620689655172414,
      "grad_norm": 1.2421875,
      "learning_rate": 5.45e-05,
      "loss": 6.7311,
      "mean_token_accuracy": 0.10817711427807808,
      "num_tokens": 194122.0,
      "step": 110
    },
    {
      "entropy": 7.15381441116333,
      "epoch": 0.09012539184952978,
      "grad_norm": 1.1328125,
      "learning_rate": 5.7e-05,
      "loss": 6.7795,
      "mean_token_accuracy": 0.10609495490789414,
      "num_tokens": 203097.0,
      "step": 115
    },
    {
      "entropy": 7.090649890899658,
      "epoch": 0.09404388714733543,
      "grad_norm": 1.1484375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.7688,
      "mean_token_accuracy": 0.10584712252020836,
      "num_tokens": 211413.0,
      "step": 120
    },
    {
      "entropy": 7.0910121440887455,
      "epoch": 0.09796238244514106,
      "grad_norm": 1.125,
      "learning_rate": 6.2e-05,
      "loss": 6.8318,
      "mean_token_accuracy": 0.1045131392776966,
      "num_tokens": 220550.0,
      "step": 125
    },
    {
      "entropy": 7.155442237854004,
      "epoch": 0.10188087774294671,
      "grad_norm": 1.2109375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.7659,
      "mean_token_accuracy": 0.10155851393938065,
      "num_tokens": 229197.0,
      "step": 130
    },
    {
      "entropy": 7.165218687057495,
      "epoch": 0.10579937304075235,
      "grad_norm": 1.1796875,
      "learning_rate": 6.7e-05,
      "loss": 6.8843,
      "mean_token_accuracy": 0.09892990812659264,
      "num_tokens": 238479.0,
      "step": 135
    },
    {
      "entropy": 7.259754037857055,
      "epoch": 0.109717868338558,
      "grad_norm": 1.2890625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.8008,
      "mean_token_accuracy": 0.10250233411788941,
      "num_tokens": 246318.0,
      "step": 140
    },
    {
      "entropy": 7.158761405944825,
      "epoch": 0.11363636363636363,
      "grad_norm": 1.1875,
      "learning_rate": 7.2e-05,
      "loss": 6.8664,
      "mean_token_accuracy": 0.10097494721412659,
      "num_tokens": 254783.0,
      "step": 145
    },
    {
      "entropy": 7.123447751998901,
      "epoch": 0.11755485893416928,
      "grad_norm": 1.265625,
      "learning_rate": 7.45e-05,
      "loss": 6.7508,
      "mean_token_accuracy": 0.10182412713766098,
      "num_tokens": 263416.0,
      "step": 150
    },
    {
      "entropy": 7.246314954757691,
      "epoch": 0.12147335423197492,
      "grad_norm": 1.390625,
      "learning_rate": 7.7e-05,
      "loss": 6.8447,
      "mean_token_accuracy": 0.1027430385351181,
      "num_tokens": 271930.0,
      "step": 155
    },
    {
      "entropy": 7.18838324546814,
      "epoch": 0.12539184952978055,
      "grad_norm": 1.1328125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.7962,
      "mean_token_accuracy": 0.10592841804027557,
      "num_tokens": 280546.0,
      "step": 160
    },
    {
      "entropy": 7.097708988189697,
      "epoch": 0.12931034482758622,
      "grad_norm": 1.2890625,
      "learning_rate": 8.2e-05,
      "loss": 6.8186,
      "mean_token_accuracy": 0.10792352184653282,
      "num_tokens": 288813.0,
      "step": 165
    },
    {
      "entropy": 7.2543150901794435,
      "epoch": 0.13322884012539185,
      "grad_norm": 1.21875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.7815,
      "mean_token_accuracy": 0.10464645773172379,
      "num_tokens": 297966.0,
      "step": 170
    },
    {
      "entropy": 7.13132381439209,
      "epoch": 0.1371473354231975,
      "grad_norm": 1.2109375,
      "learning_rate": 8.7e-05,
      "loss": 6.9155,
      "mean_token_accuracy": 0.10553643256425857,
      "num_tokens": 307135.0,
      "step": 175
    },
    {
      "entropy": 7.204241514205933,
      "epoch": 0.14106583072100312,
      "grad_norm": 1.1171875,
      "learning_rate": 8.95e-05,
      "loss": 6.8318,
      "mean_token_accuracy": 0.10002151355147362,
      "num_tokens": 315546.0,
      "step": 180
    },
    {
      "entropy": 7.187180662155152,
      "epoch": 0.14498432601880878,
      "grad_norm": 1.21875,
      "learning_rate": 9.2e-05,
      "loss": 6.8224,
      "mean_token_accuracy": 0.10502876788377762,
      "num_tokens": 323930.0,
      "step": 185
    },
    {
      "entropy": 7.087406158447266,
      "epoch": 0.14890282131661442,
      "grad_norm": 1.328125,
      "learning_rate": 9.45e-05,
      "loss": 6.7248,
      "mean_token_accuracy": 0.11350480541586876,
      "num_tokens": 332419.0,
      "step": 190
    },
    {
      "entropy": 7.082424688339233,
      "epoch": 0.15282131661442006,
      "grad_norm": 1.265625,
      "learning_rate": 9.7e-05,
      "loss": 6.8184,
      "mean_token_accuracy": 0.10288615301251411,
      "num_tokens": 341362.0,
      "step": 195
    },
    {
      "entropy": 7.1997878074646,
      "epoch": 0.15673981191222572,
      "grad_norm": 1.2890625,
      "learning_rate": 9.95e-05,
      "loss": 6.7508,
      "mean_token_accuracy": 0.10806922689080238,
      "num_tokens": 349395.0,
      "step": 200
    },
    {
      "entropy": 7.0235200881958,
      "epoch": 0.16065830721003135,
      "grad_norm": 1.40625,
      "learning_rate": 0.000102,
      "loss": 6.8101,
      "mean_token_accuracy": 0.099128620326519,
      "num_tokens": 358413.0,
      "step": 205
    },
    {
      "entropy": 7.297369050979614,
      "epoch": 0.164576802507837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.8827,
      "mean_token_accuracy": 0.10295988991856575,
      "num_tokens": 366489.0,
      "step": 210
    },
    {
      "entropy": 7.091110467910767,
      "epoch": 0.16849529780564262,
      "grad_norm": 1.203125,
      "learning_rate": 0.000107,
      "loss": 6.7112,
      "mean_token_accuracy": 0.10220234468579292,
      "num_tokens": 375335.0,
      "step": 215
    },
    {
      "entropy": 7.204977703094483,
      "epoch": 0.1724137931034483,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001095,
      "loss": 6.7757,
      "mean_token_accuracy": 0.10802004486322403,
      "num_tokens": 384276.0,
      "step": 220
    },
    {
      "entropy": 7.145157670974731,
      "epoch": 0.17633228840125392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000112,
      "loss": 6.8694,
      "mean_token_accuracy": 0.09782575145363807,
      "num_tokens": 393571.0,
      "step": 225
    },
    {
      "entropy": 7.1306267261505125,
      "epoch": 0.18025078369905956,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001145,
      "loss": 6.8154,
      "mean_token_accuracy": 0.09901047199964523,
      "num_tokens": 401865.0,
      "step": 230
    },
    {
      "entropy": 7.0903645038604735,
      "epoch": 0.1841692789968652,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.7693,
      "mean_token_accuracy": 0.10383733138442039,
      "num_tokens": 410518.0,
      "step": 235
    },
    {
      "entropy": 7.176298904418945,
      "epoch": 0.18808777429467086,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.772,
      "mean_token_accuracy": 0.10165056958794594,
      "num_tokens": 419769.0,
      "step": 240
    },
    {
      "entropy": 7.052699327468872,
      "epoch": 0.1920062695924765,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000122,
      "loss": 6.7062,
      "mean_token_accuracy": 0.11240611821413041,
      "num_tokens": 428204.0,
      "step": 245
    },
    {
      "entropy": 7.229116773605346,
      "epoch": 0.19592476489028213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001245,
      "loss": 6.8173,
      "mean_token_accuracy": 0.10085652396082878,
      "num_tokens": 436594.0,
      "step": 250
    },
    {
      "entropy": 6.918967628479004,
      "epoch": 0.19984326018808776,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000127,
      "loss": 6.6093,
      "mean_token_accuracy": 0.1123976968228817,
      "num_tokens": 444645.0,
      "step": 255
    },
    {
      "entropy": 7.119955444335938,
      "epoch": 0.20376175548589343,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001295,
      "loss": 6.6713,
      "mean_token_accuracy": 0.1125674806535244,
      "num_tokens": 453016.0,
      "step": 260
    },
    {
      "entropy": 7.06598129272461,
      "epoch": 0.20768025078369906,
      "grad_norm": 1.171875,
      "learning_rate": 0.000132,
      "loss": 6.7522,
      "mean_token_accuracy": 0.1047289289534092,
      "num_tokens": 462116.0,
      "step": 265
    },
    {
      "entropy": 7.1544452667236325,
      "epoch": 0.2115987460815047,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.7091,
      "mean_token_accuracy": 0.10634818449616432,
      "num_tokens": 470807.0,
      "step": 270
    },
    {
      "entropy": 7.103619909286499,
      "epoch": 0.21551724137931033,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8065,
      "mean_token_accuracy": 0.10270731151103973,
      "num_tokens": 479592.0,
      "step": 275
    },
    {
      "entropy": 7.1156768798828125,
      "epoch": 0.219435736677116,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001395,
      "loss": 6.7446,
      "mean_token_accuracy": 0.10210504084825515,
      "num_tokens": 488107.0,
      "step": 280
    },
    {
      "entropy": 7.0720561981201175,
      "epoch": 0.22335423197492163,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7655,
      "mean_token_accuracy": 0.1056784711778164,
      "num_tokens": 496775.0,
      "step": 285
    },
    {
      "entropy": 7.222696924209595,
      "epoch": 0.22727272727272727,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001445,
      "loss": 6.7897,
      "mean_token_accuracy": 0.10648871958255768,
      "num_tokens": 505415.0,
      "step": 290
    },
    {
      "entropy": 7.079438734054565,
      "epoch": 0.23119122257053293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000147,
      "loss": 6.7795,
      "mean_token_accuracy": 0.1061191700398922,
      "num_tokens": 514447.0,
      "step": 295
    },
    {
      "entropy": 7.112887191772461,
      "epoch": 0.23510971786833856,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001495,
      "loss": 6.7863,
      "mean_token_accuracy": 0.10735970661044121,
      "num_tokens": 522998.0,
      "step": 300
    },
    {
      "entropy": 7.187893390655518,
      "epoch": 0.2390282131661442,
      "grad_norm": 1.375,
      "learning_rate": 0.000152,
      "loss": 6.7606,
      "mean_token_accuracy": 0.10097135305404663,
      "num_tokens": 531542.0,
      "step": 305
    },
    {
      "entropy": 7.002828359603882,
      "epoch": 0.24294670846394983,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.6867,
      "mean_token_accuracy": 0.10434283241629601,
      "num_tokens": 540143.0,
      "step": 310
    },
    {
      "entropy": 7.160926628112793,
      "epoch": 0.2468652037617555,
      "grad_norm": 1.21875,
      "learning_rate": 0.000157,
      "loss": 6.8173,
      "mean_token_accuracy": 0.10131673216819763,
      "num_tokens": 549156.0,
      "step": 315
    },
    {
      "entropy": 7.030450010299683,
      "epoch": 0.2507836990595611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001595,
      "loss": 6.6447,
      "mean_token_accuracy": 0.1034945398569107,
      "num_tokens": 557522.0,
      "step": 320
    },
    {
      "entropy": 7.135869073867798,
      "epoch": 0.2547021943573668,
      "grad_norm": 1.28125,
      "learning_rate": 0.000162,
      "loss": 6.7145,
      "mean_token_accuracy": 0.109796904027462,
      "num_tokens": 566650.0,
      "step": 325
    },
    {
      "entropy": 6.977855205535889,
      "epoch": 0.25862068965517243,
      "grad_norm": 1.421875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.6739,
      "mean_token_accuracy": 0.1077032022178173,
      "num_tokens": 576219.0,
      "step": 330
    },
    {
      "entropy": 7.109291315078735,
      "epoch": 0.26253918495297807,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8204,
      "mean_token_accuracy": 0.1049183763563633,
      "num_tokens": 584304.0,
      "step": 335
    },
    {
      "entropy": 7.060112428665161,
      "epoch": 0.2664576802507837,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7326,
      "mean_token_accuracy": 0.10800091400742531,
      "num_tokens": 593163.0,
      "step": 340
    },
    {
      "entropy": 7.065881586074829,
      "epoch": 0.27037617554858934,
      "grad_norm": 1.5,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.6391,
      "mean_token_accuracy": 0.10378878638148308,
      "num_tokens": 602077.0,
      "step": 345
    },
    {
      "entropy": 7.035525178909301,
      "epoch": 0.274294670846395,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.7544,
      "mean_token_accuracy": 0.10282572582364083,
      "num_tokens": 610009.0,
      "step": 350
    },
    {
      "entropy": 7.141838598251343,
      "epoch": 0.2782131661442006,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000177,
      "loss": 6.8191,
      "mean_token_accuracy": 0.10182501152157783,
      "num_tokens": 619282.0,
      "step": 355
    },
    {
      "entropy": 7.106745004653931,
      "epoch": 0.28213166144200624,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001795,
      "loss": 6.7578,
      "mean_token_accuracy": 0.10329684540629387,
      "num_tokens": 628138.0,
      "step": 360
    },
    {
      "entropy": 7.018217658996582,
      "epoch": 0.28605015673981193,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000182,
      "loss": 6.579,
      "mean_token_accuracy": 0.11497451290488243,
      "num_tokens": 637021.0,
      "step": 365
    },
    {
      "entropy": 7.0183099746704105,
      "epoch": 0.28996865203761757,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001845,
      "loss": 6.7904,
      "mean_token_accuracy": 0.09623993113636971,
      "num_tokens": 646703.0,
      "step": 370
    },
    {
      "entropy": 7.033835697174072,
      "epoch": 0.2938871473354232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000187,
      "loss": 6.6457,
      "mean_token_accuracy": 0.10587730780243873,
      "num_tokens": 655616.0,
      "step": 375
    },
    {
      "entropy": 7.035226058959961,
      "epoch": 0.29780564263322884,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001895,
      "loss": 6.6675,
      "mean_token_accuracy": 0.11625474393367767,
      "num_tokens": 663783.0,
      "step": 380
    },
    {
      "entropy": 7.009380054473877,
      "epoch": 0.3017241379310345,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000192,
      "loss": 6.6976,
      "mean_token_accuracy": 0.10993565618991852,
      "num_tokens": 671855.0,
      "step": 385
    },
    {
      "entropy": 7.081801080703736,
      "epoch": 0.3056426332288401,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001945,
      "loss": 6.706,
      "mean_token_accuracy": 0.10679346546530724,
      "num_tokens": 680981.0,
      "step": 390
    },
    {
      "entropy": 6.960446166992187,
      "epoch": 0.30956112852664575,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.6116,
      "mean_token_accuracy": 0.10894792973995208,
      "num_tokens": 689294.0,
      "step": 395
    },
    {
      "entropy": 6.911626148223877,
      "epoch": 0.31347962382445144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.5734,
      "mean_token_accuracy": 0.1067522719502449,
      "num_tokens": 697703.0,
      "step": 400
    },
    {
      "entropy": 7.081985807418823,
      "epoch": 0.31739811912225707,
      "grad_norm": 1.3125,
      "learning_rate": 0.000202,
      "loss": 6.7298,
      "mean_token_accuracy": 0.10676686987280845,
      "num_tokens": 706792.0,
      "step": 405
    },
    {
      "entropy": 7.004686450958252,
      "epoch": 0.3213166144200627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6674,
      "mean_token_accuracy": 0.11164869293570519,
      "num_tokens": 715864.0,
      "step": 410
    },
    {
      "entropy": 7.0087442874908445,
      "epoch": 0.32523510971786834,
      "grad_norm": 1.375,
      "learning_rate": 0.000207,
      "loss": 6.7022,
      "mean_token_accuracy": 0.10990188345313072,
      "num_tokens": 723921.0,
      "step": 415
    },
    {
      "entropy": 7.068296241760254,
      "epoch": 0.329153605015674,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002095,
      "loss": 6.7359,
      "mean_token_accuracy": 0.10363818407058716,
      "num_tokens": 732797.0,
      "step": 420
    },
    {
      "entropy": 6.906872224807739,
      "epoch": 0.3330721003134796,
      "grad_norm": 1.328125,
      "learning_rate": 0.000212,
      "loss": 6.5814,
      "mean_token_accuracy": 0.11549990624189377,
      "num_tokens": 741248.0,
      "step": 425
    },
    {
      "entropy": 7.061840009689331,
      "epoch": 0.33699059561128525,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002145,
      "loss": 6.6781,
      "mean_token_accuracy": 0.10647747218608856,
      "num_tokens": 749766.0,
      "step": 430
    },
    {
      "entropy": 6.947984743118286,
      "epoch": 0.3409090909090909,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6594,
      "mean_token_accuracy": 0.10216243714094161,
      "num_tokens": 758695.0,
      "step": 435
    },
    {
      "entropy": 7.034923553466797,
      "epoch": 0.3448275862068966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002195,
      "loss": 6.6444,
      "mean_token_accuracy": 0.10249690040946006,
      "num_tokens": 767624.0,
      "step": 440
    },
    {
      "entropy": 6.945315933227539,
      "epoch": 0.3487460815047022,
      "grad_norm": 1.421875,
      "learning_rate": 0.000222,
      "loss": 6.6862,
      "mean_token_accuracy": 0.10265796408057212,
      "num_tokens": 776616.0,
      "step": 445
    },
    {
      "entropy": 6.97764482498169,
      "epoch": 0.35266457680250785,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002245,
      "loss": 6.6966,
      "mean_token_accuracy": 0.10413416102528572,
      "num_tokens": 786147.0,
      "step": 450
    },
    {
      "entropy": 7.103307485580444,
      "epoch": 0.3565830721003135,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7967,
      "mean_token_accuracy": 0.10401362031698227,
      "num_tokens": 795213.0,
      "step": 455
    },
    {
      "entropy": 6.968210601806641,
      "epoch": 0.3605015673981191,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6085,
      "mean_token_accuracy": 0.10945433229207993,
      "num_tokens": 803118.0,
      "step": 460
    },
    {
      "entropy": 6.9675391674041744,
      "epoch": 0.36442006269592475,
      "grad_norm": 1.25,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5691,
      "mean_token_accuracy": 0.11808251142501831,
      "num_tokens": 811358.0,
      "step": 465
    },
    {
      "entropy": 6.9308342933654785,
      "epoch": 0.3683385579937304,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.578,
      "mean_token_accuracy": 0.11157155334949494,
      "num_tokens": 819653.0,
      "step": 470
    },
    {
      "entropy": 6.824059677124024,
      "epoch": 0.3722570532915361,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000237,
      "loss": 6.6342,
      "mean_token_accuracy": 0.10801615715026855,
      "num_tokens": 828462.0,
      "step": 475
    },
    {
      "entropy": 7.107431173324585,
      "epoch": 0.3761755485893417,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002395,
      "loss": 6.7107,
      "mean_token_accuracy": 0.10842569470405579,
      "num_tokens": 836204.0,
      "step": 480
    },
    {
      "entropy": 6.875178289413452,
      "epoch": 0.38009404388714735,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000242,
      "loss": 6.6054,
      "mean_token_accuracy": 0.11096373572945595,
      "num_tokens": 845032.0,
      "step": 485
    },
    {
      "entropy": 6.9599480628967285,
      "epoch": 0.384012539184953,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002445,
      "loss": 6.6359,
      "mean_token_accuracy": 0.11337224692106247,
      "num_tokens": 853324.0,
      "step": 490
    },
    {
      "entropy": 6.926474952697754,
      "epoch": 0.3879310344827586,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000247,
      "loss": 6.5964,
      "mean_token_accuracy": 0.12043852508068084,
      "num_tokens": 861640.0,
      "step": 495
    },
    {
      "entropy": 6.9969902515411375,
      "epoch": 0.39184952978056425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002495,
      "loss": 6.7759,
      "mean_token_accuracy": 0.10493798106908798,
      "num_tokens": 870758.0,
      "step": 500
    },
    {
      "epoch": 0.39184952978056425,
      "eval_entropy": 6.9411013653112015,
      "eval_loss": 6.728126049041748,
      "eval_mean_token_accuracy": 0.10773137585772563,
      "eval_num_tokens": 870758.0,
      "eval_runtime": 2.9294,
      "eval_samples_per_second": 1407.095,
      "eval_steps_per_second": 176.143,
      "step": 500
    },
    {
      "entropy": 6.992229700088501,
      "epoch": 0.3957680250783699,
      "grad_norm": 1.46875,
      "learning_rate": 0.000252,
      "loss": 6.6957,
      "mean_token_accuracy": 0.10136032626032829,
      "num_tokens": 879968.0,
      "step": 505
    },
    {
      "entropy": 6.883435440063477,
      "epoch": 0.3996865203761755,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002545,
      "loss": 6.6479,
      "mean_token_accuracy": 0.10431178212165833,
      "num_tokens": 888804.0,
      "step": 510
    },
    {
      "entropy": 6.971563863754272,
      "epoch": 0.4036050156739812,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000257,
      "loss": 6.6693,
      "mean_token_accuracy": 0.10408427342772483,
      "num_tokens": 898593.0,
      "step": 515
    },
    {
      "entropy": 6.925919246673584,
      "epoch": 0.40752351097178685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002595,
      "loss": 6.6901,
      "mean_token_accuracy": 0.10555115193128586,
      "num_tokens": 906873.0,
      "step": 520
    },
    {
      "entropy": 6.934886932373047,
      "epoch": 0.4114420062695925,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000262,
      "loss": 6.6698,
      "mean_token_accuracy": 0.11156188994646073,
      "num_tokens": 915080.0,
      "step": 525
    },
    {
      "entropy": 6.947892522811889,
      "epoch": 0.4153605015673981,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6785,
      "mean_token_accuracy": 0.10801282972097397,
      "num_tokens": 923741.0,
      "step": 530
    },
    {
      "entropy": 6.883993148803711,
      "epoch": 0.41927899686520376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.564,
      "mean_token_accuracy": 0.11001835316419602,
      "num_tokens": 932455.0,
      "step": 535
    },
    {
      "entropy": 6.902733898162841,
      "epoch": 0.4231974921630094,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.5815,
      "mean_token_accuracy": 0.11325104832649231,
      "num_tokens": 941064.0,
      "step": 540
    },
    {
      "entropy": 6.83945484161377,
      "epoch": 0.427115987460815,
      "grad_norm": 1.25,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6085,
      "mean_token_accuracy": 0.109104935079813,
      "num_tokens": 950261.0,
      "step": 545
    },
    {
      "entropy": 7.026263904571533,
      "epoch": 0.43103448275862066,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002745,
      "loss": 6.6813,
      "mean_token_accuracy": 0.10714460536837578,
      "num_tokens": 959151.0,
      "step": 550
    },
    {
      "entropy": 6.907926321029663,
      "epoch": 0.43495297805642635,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000277,
      "loss": 6.7019,
      "mean_token_accuracy": 0.1055656410753727,
      "num_tokens": 968441.0,
      "step": 555
    },
    {
      "entropy": 6.876136827468872,
      "epoch": 0.438871473354232,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002795,
      "loss": 6.5755,
      "mean_token_accuracy": 0.10419308096170425,
      "num_tokens": 977058.0,
      "step": 560
    },
    {
      "entropy": 6.904338932037353,
      "epoch": 0.4427899686520376,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5945,
      "mean_token_accuracy": 0.11105613335967064,
      "num_tokens": 986531.0,
      "step": 565
    },
    {
      "entropy": 6.768454742431641,
      "epoch": 0.44670846394984326,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002845,
      "loss": 6.4264,
      "mean_token_accuracy": 0.11743755117058755,
      "num_tokens": 995081.0,
      "step": 570
    },
    {
      "entropy": 6.789128589630127,
      "epoch": 0.4506269592476489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000287,
      "loss": 6.4896,
      "mean_token_accuracy": 0.11737141981720925,
      "num_tokens": 1003459.0,
      "step": 575
    },
    {
      "entropy": 6.81758337020874,
      "epoch": 0.45454545454545453,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002895,
      "loss": 6.5718,
      "mean_token_accuracy": 0.11360911354422569,
      "num_tokens": 1012420.0,
      "step": 580
    },
    {
      "entropy": 6.936664247512818,
      "epoch": 0.45846394984326017,
      "grad_norm": 1.375,
      "learning_rate": 0.000292,
      "loss": 6.6519,
      "mean_token_accuracy": 0.10682766661047935,
      "num_tokens": 1021198.0,
      "step": 585
    },
    {
      "entropy": 6.860663843154907,
      "epoch": 0.46238244514106586,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002945,
      "loss": 6.6291,
      "mean_token_accuracy": 0.11134118661284446,
      "num_tokens": 1030023.0,
      "step": 590
    },
    {
      "entropy": 6.792492532730103,
      "epoch": 0.4663009404388715,
      "grad_norm": 1.34375,
      "learning_rate": 0.000297,
      "loss": 6.4932,
      "mean_token_accuracy": 0.11575940474867821,
      "num_tokens": 1038824.0,
      "step": 595
    },
    {
      "entropy": 6.972049903869629,
      "epoch": 0.4702194357366771,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002995,
      "loss": 6.7682,
      "mean_token_accuracy": 0.10632524192333222,
      "num_tokens": 1048207.0,
      "step": 600
    },
    {
      "entropy": 7.017117214202881,
      "epoch": 0.47413793103448276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000302,
      "loss": 6.721,
      "mean_token_accuracy": 0.10762688145041466,
      "num_tokens": 1057354.0,
      "step": 605
    },
    {
      "entropy": 6.904824686050415,
      "epoch": 0.4780564263322884,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003045,
      "loss": 6.6649,
      "mean_token_accuracy": 0.10675990283489227,
      "num_tokens": 1066825.0,
      "step": 610
    },
    {
      "entropy": 6.836680459976196,
      "epoch": 0.48197492163009403,
      "grad_norm": 1.453125,
      "learning_rate": 0.000307,
      "loss": 6.6006,
      "mean_token_accuracy": 0.10550268962979317,
      "num_tokens": 1075752.0,
      "step": 615
    },
    {
      "entropy": 6.744200849533081,
      "epoch": 0.48589341692789967,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003095,
      "loss": 6.4958,
      "mean_token_accuracy": 0.11007569953799248,
      "num_tokens": 1084826.0,
      "step": 620
    },
    {
      "entropy": 6.978894758224487,
      "epoch": 0.4898119122257053,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000312,
      "loss": 6.6386,
      "mean_token_accuracy": 0.10888363420963287,
      "num_tokens": 1093425.0,
      "step": 625
    },
    {
      "entropy": 6.815886735916138,
      "epoch": 0.493730407523511,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003145,
      "loss": 6.6631,
      "mean_token_accuracy": 0.10534236505627632,
      "num_tokens": 1102619.0,
      "step": 630
    },
    {
      "entropy": 6.809355783462524,
      "epoch": 0.49764890282131663,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000317,
      "loss": 6.5432,
      "mean_token_accuracy": 0.11107134073972702,
      "num_tokens": 1111337.0,
      "step": 635
    },
    {
      "entropy": 6.7085394859313965,
      "epoch": 0.5015673981191222,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003195,
      "loss": 6.3882,
      "mean_token_accuracy": 0.1193773865699768,
      "num_tokens": 1119381.0,
      "step": 640
    },
    {
      "entropy": 6.69315505027771,
      "epoch": 0.5054858934169278,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000322,
      "loss": 6.4395,
      "mean_token_accuracy": 0.11702300012111663,
      "num_tokens": 1127825.0,
      "step": 645
    },
    {
      "entropy": 6.850753402709961,
      "epoch": 0.5094043887147336,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4943,
      "mean_token_accuracy": 0.10672505050897599,
      "num_tokens": 1136750.0,
      "step": 650
    },
    {
      "entropy": 6.7495848655700685,
      "epoch": 0.5133228840125392,
      "grad_norm": 1.25,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5289,
      "mean_token_accuracy": 0.11170963272452354,
      "num_tokens": 1145739.0,
      "step": 655
    },
    {
      "entropy": 6.791320037841797,
      "epoch": 0.5172413793103449,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5995,
      "mean_token_accuracy": 0.10925844609737397,
      "num_tokens": 1154415.0,
      "step": 660
    },
    {
      "entropy": 6.774902629852295,
      "epoch": 0.5211598746081505,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4739,
      "mean_token_accuracy": 0.11284942775964737,
      "num_tokens": 1162480.0,
      "step": 665
    },
    {
      "entropy": 6.759902763366699,
      "epoch": 0.5250783699059561,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5655,
      "mean_token_accuracy": 0.10794797912240028,
      "num_tokens": 1170774.0,
      "step": 670
    },
    {
      "entropy": 6.727387142181397,
      "epoch": 0.5289968652037618,
      "grad_norm": 1.171875,
      "learning_rate": 0.000337,
      "loss": 6.5016,
      "mean_token_accuracy": 0.1089845173060894,
      "num_tokens": 1179629.0,
      "step": 675
    },
    {
      "entropy": 6.78474268913269,
      "epoch": 0.5329153605015674,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003395,
      "loss": 6.6631,
      "mean_token_accuracy": 0.1076794520020485,
      "num_tokens": 1189111.0,
      "step": 680
    },
    {
      "entropy": 6.840711498260498,
      "epoch": 0.536833855799373,
      "grad_norm": 1.375,
      "learning_rate": 0.000342,
      "loss": 6.6419,
      "mean_token_accuracy": 0.10526446029543876,
      "num_tokens": 1198827.0,
      "step": 685
    },
    {
      "entropy": 6.920011568069458,
      "epoch": 0.5407523510971787,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5519,
      "mean_token_accuracy": 0.11411965638399124,
      "num_tokens": 1207089.0,
      "step": 690
    },
    {
      "entropy": 6.778568458557129,
      "epoch": 0.5446708463949843,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000347,
      "loss": 6.5436,
      "mean_token_accuracy": 0.10897781774401664,
      "num_tokens": 1216509.0,
      "step": 695
    },
    {
      "entropy": 6.596413469314575,
      "epoch": 0.54858934169279,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003495,
      "loss": 6.3856,
      "mean_token_accuracy": 0.1203030876815319,
      "num_tokens": 1225684.0,
      "step": 700
    },
    {
      "entropy": 6.760894346237182,
      "epoch": 0.5525078369905956,
      "grad_norm": 1.359375,
      "learning_rate": 0.000352,
      "loss": 6.4132,
      "mean_token_accuracy": 0.12549414113163948,
      "num_tokens": 1233675.0,
      "step": 705
    },
    {
      "entropy": 6.648054599761963,
      "epoch": 0.5564263322884012,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003545,
      "loss": 6.4268,
      "mean_token_accuracy": 0.11218728423118592,
      "num_tokens": 1242147.0,
      "step": 710
    },
    {
      "entropy": 6.716562032699585,
      "epoch": 0.5603448275862069,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000357,
      "loss": 6.5515,
      "mean_token_accuracy": 0.11361126005649566,
      "num_tokens": 1251127.0,
      "step": 715
    },
    {
      "entropy": 6.667671203613281,
      "epoch": 0.5642633228840125,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003595,
      "loss": 6.4954,
      "mean_token_accuracy": 0.11189050376415252,
      "num_tokens": 1260589.0,
      "step": 720
    },
    {
      "entropy": 6.848757553100586,
      "epoch": 0.5681818181818182,
      "grad_norm": 1.28125,
      "learning_rate": 0.000362,
      "loss": 6.5731,
      "mean_token_accuracy": 0.11687926128506661,
      "num_tokens": 1269281.0,
      "step": 725
    },
    {
      "entropy": 6.655617666244507,
      "epoch": 0.5721003134796239,
      "grad_norm": 1.25,
      "learning_rate": 0.0003645,
      "loss": 6.419,
      "mean_token_accuracy": 0.11433188989758492,
      "num_tokens": 1278033.0,
      "step": 730
    },
    {
      "entropy": 6.798342895507813,
      "epoch": 0.5760188087774295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000367,
      "loss": 6.5162,
      "mean_token_accuracy": 0.11682824715971947,
      "num_tokens": 1285847.0,
      "step": 735
    },
    {
      "entropy": 6.71445107460022,
      "epoch": 0.5799373040752351,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003695,
      "loss": 6.4871,
      "mean_token_accuracy": 0.11209994032979012,
      "num_tokens": 1294770.0,
      "step": 740
    },
    {
      "entropy": 6.870912408828735,
      "epoch": 0.5838557993730408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000372,
      "loss": 6.7032,
      "mean_token_accuracy": 0.10493360385298729,
      "num_tokens": 1304282.0,
      "step": 745
    },
    {
      "entropy": 6.731619119644165,
      "epoch": 0.5877742946708464,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003745,
      "loss": 6.5341,
      "mean_token_accuracy": 0.10778379514813423,
      "num_tokens": 1313218.0,
      "step": 750
    },
    {
      "entropy": 6.747414207458496,
      "epoch": 0.591692789968652,
      "grad_norm": 1.171875,
      "learning_rate": 0.000377,
      "loss": 6.5067,
      "mean_token_accuracy": 0.11040924564003944,
      "num_tokens": 1322093.0,
      "step": 755
    },
    {
      "entropy": 6.585576343536377,
      "epoch": 0.5956112852664577,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003795,
      "loss": 6.5403,
      "mean_token_accuracy": 0.11336176246404647,
      "num_tokens": 1330718.0,
      "step": 760
    },
    {
      "entropy": 6.7226659774780275,
      "epoch": 0.5995297805642633,
      "grad_norm": 1.5625,
      "learning_rate": 0.000382,
      "loss": 6.3939,
      "mean_token_accuracy": 0.11393325701355934,
      "num_tokens": 1339672.0,
      "step": 765
    },
    {
      "entropy": 6.641554880142212,
      "epoch": 0.603448275862069,
      "grad_norm": 1.25,
      "learning_rate": 0.0003845,
      "loss": 6.4174,
      "mean_token_accuracy": 0.11754084452986717,
      "num_tokens": 1348587.0,
      "step": 770
    },
    {
      "entropy": 6.603374004364014,
      "epoch": 0.6073667711598746,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4169,
      "mean_token_accuracy": 0.11932410523295403,
      "num_tokens": 1357597.0,
      "step": 775
    },
    {
      "entropy": 6.732632350921631,
      "epoch": 0.6112852664576802,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5225,
      "mean_token_accuracy": 0.11163635179400444,
      "num_tokens": 1366246.0,
      "step": 780
    },
    {
      "entropy": 6.643588590621948,
      "epoch": 0.6152037617554859,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4231,
      "mean_token_accuracy": 0.11742421090602875,
      "num_tokens": 1374664.0,
      "step": 785
    },
    {
      "entropy": 6.60569806098938,
      "epoch": 0.6191222570532915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4298,
      "mean_token_accuracy": 0.11620132252573967,
      "num_tokens": 1382765.0,
      "step": 790
    },
    {
      "entropy": 6.606018590927124,
      "epoch": 0.6230407523510971,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4046,
      "mean_token_accuracy": 0.11398342102766038,
      "num_tokens": 1391708.0,
      "step": 795
    },
    {
      "entropy": 6.664660692214966,
      "epoch": 0.6269592476489029,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003995,
      "loss": 6.5624,
      "mean_token_accuracy": 0.1093956932425499,
      "num_tokens": 1401373.0,
      "step": 800
    },
    {
      "entropy": 6.67238826751709,
      "epoch": 0.6308777429467085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000402,
      "loss": 6.3628,
      "mean_token_accuracy": 0.11816710904240608,
      "num_tokens": 1409702.0,
      "step": 805
    },
    {
      "entropy": 6.693083429336548,
      "epoch": 0.6347962382445141,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004045,
      "loss": 6.5834,
      "mean_token_accuracy": 0.10899265483021736,
      "num_tokens": 1419596.0,
      "step": 810
    },
    {
      "entropy": 6.636869716644287,
      "epoch": 0.6387147335423198,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5177,
      "mean_token_accuracy": 0.11091106310486794,
      "num_tokens": 1428437.0,
      "step": 815
    },
    {
      "entropy": 6.675769662857055,
      "epoch": 0.6426332288401254,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004095,
      "loss": 6.4288,
      "mean_token_accuracy": 0.12033025994896888,
      "num_tokens": 1436764.0,
      "step": 820
    },
    {
      "entropy": 6.624548244476318,
      "epoch": 0.646551724137931,
      "grad_norm": 1.1875,
      "learning_rate": 0.000412,
      "loss": 6.4725,
      "mean_token_accuracy": 0.11756742298603058,
      "num_tokens": 1445755.0,
      "step": 825
    },
    {
      "entropy": 6.693640947341919,
      "epoch": 0.6504702194357367,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004145,
      "loss": 6.5313,
      "mean_token_accuracy": 0.1114391803741455,
      "num_tokens": 1454550.0,
      "step": 830
    },
    {
      "entropy": 6.563589239120484,
      "epoch": 0.6543887147335423,
      "grad_norm": 1.21875,
      "learning_rate": 0.000417,
      "loss": 6.418,
      "mean_token_accuracy": 0.11557744145393371,
      "num_tokens": 1463039.0,
      "step": 835
    },
    {
      "entropy": 6.734728002548218,
      "epoch": 0.658307210031348,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004195,
      "loss": 6.478,
      "mean_token_accuracy": 0.11269574463367463,
      "num_tokens": 1472029.0,
      "step": 840
    },
    {
      "entropy": 6.571061801910401,
      "epoch": 0.6622257053291536,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000422,
      "loss": 6.5079,
      "mean_token_accuracy": 0.11439812406897545,
      "num_tokens": 1481588.0,
      "step": 845
    },
    {
      "entropy": 6.628999423980713,
      "epoch": 0.6661442006269592,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004245,
      "loss": 6.4059,
      "mean_token_accuracy": 0.11478642746806145,
      "num_tokens": 1489915.0,
      "step": 850
    },
    {
      "entropy": 6.672647428512573,
      "epoch": 0.6700626959247649,
      "grad_norm": 1.1875,
      "learning_rate": 0.000427,
      "loss": 6.4842,
      "mean_token_accuracy": 0.11243890598416328,
      "num_tokens": 1497920.0,
      "step": 855
    },
    {
      "entropy": 6.510155534744262,
      "epoch": 0.6739811912225705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004295,
      "loss": 6.3466,
      "mean_token_accuracy": 0.1146190918982029,
      "num_tokens": 1506965.0,
      "step": 860
    },
    {
      "entropy": 6.6185520648956295,
      "epoch": 0.6778996865203761,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000432,
      "loss": 6.4236,
      "mean_token_accuracy": 0.11795553267002105,
      "num_tokens": 1516028.0,
      "step": 865
    },
    {
      "entropy": 6.592772388458252,
      "epoch": 0.6818181818181818,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004345,
      "loss": 6.4644,
      "mean_token_accuracy": 0.11529229208827019,
      "num_tokens": 1524579.0,
      "step": 870
    },
    {
      "entropy": 6.539308547973633,
      "epoch": 0.6857366771159875,
      "grad_norm": 1.109375,
      "learning_rate": 0.000437,
      "loss": 6.3769,
      "mean_token_accuracy": 0.11740082949399948,
      "num_tokens": 1534285.0,
      "step": 875
    },
    {
      "entropy": 6.588557434082031,
      "epoch": 0.6896551724137931,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004395,
      "loss": 6.3787,
      "mean_token_accuracy": 0.11641007363796234,
      "num_tokens": 1543701.0,
      "step": 880
    },
    {
      "entropy": 6.673776912689209,
      "epoch": 0.6935736677115988,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000442,
      "loss": 6.6596,
      "mean_token_accuracy": 0.10156535282731056,
      "num_tokens": 1552715.0,
      "step": 885
    },
    {
      "entropy": 6.408450174331665,
      "epoch": 0.6974921630094044,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004445,
      "loss": 6.2622,
      "mean_token_accuracy": 0.11470296010375022,
      "num_tokens": 1561523.0,
      "step": 890
    },
    {
      "entropy": 6.55662088394165,
      "epoch": 0.70141065830721,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000447,
      "loss": 6.4914,
      "mean_token_accuracy": 0.11589494124054908,
      "num_tokens": 1570057.0,
      "step": 895
    },
    {
      "entropy": 6.568389081954956,
      "epoch": 0.7053291536050157,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3345,
      "mean_token_accuracy": 0.11779361218214035,
      "num_tokens": 1578795.0,
      "step": 900
    },
    {
      "entropy": 6.456902074813843,
      "epoch": 0.7092476489028213,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4116,
      "mean_token_accuracy": 0.11231630444526672,
      "num_tokens": 1588108.0,
      "step": 905
    },
    {
      "entropy": 6.604302167892456,
      "epoch": 0.713166144200627,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3924,
      "mean_token_accuracy": 0.12046928554773331,
      "num_tokens": 1596545.0,
      "step": 910
    },
    {
      "entropy": 6.536626100540161,
      "epoch": 0.7170846394984326,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3547,
      "mean_token_accuracy": 0.12420721054077148,
      "num_tokens": 1605098.0,
      "step": 915
    },
    {
      "entropy": 6.524212121963501,
      "epoch": 0.7210031347962382,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4001,
      "mean_token_accuracy": 0.12292874529957772,
      "num_tokens": 1613760.0,
      "step": 920
    },
    {
      "entropy": 6.498387098312378,
      "epoch": 0.7249216300940439,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000462,
      "loss": 6.427,
      "mean_token_accuracy": 0.11109715700149536,
      "num_tokens": 1622345.0,
      "step": 925
    },
    {
      "entropy": 6.643976163864136,
      "epoch": 0.7288401253918495,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004645,
      "loss": 6.4157,
      "mean_token_accuracy": 0.1193670779466629,
      "num_tokens": 1630706.0,
      "step": 930
    },
    {
      "entropy": 6.5410667896270756,
      "epoch": 0.7327586206896551,
      "grad_norm": 1.265625,
      "learning_rate": 0.000467,
      "loss": 6.4832,
      "mean_token_accuracy": 0.11565442010760307,
      "num_tokens": 1639880.0,
      "step": 935
    },
    {
      "entropy": 6.527542972564698,
      "epoch": 0.7366771159874608,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004695,
      "loss": 6.3374,
      "mean_token_accuracy": 0.11161092445254325,
      "num_tokens": 1648325.0,
      "step": 940
    },
    {
      "entropy": 6.35137448310852,
      "epoch": 0.7405956112852664,
      "grad_norm": 1.234375,
      "learning_rate": 0.000472,
      "loss": 6.2914,
      "mean_token_accuracy": 0.12073137611150742,
      "num_tokens": 1656951.0,
      "step": 945
    },
    {
      "entropy": 6.511531782150269,
      "epoch": 0.7445141065830722,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004745,
      "loss": 6.4261,
      "mean_token_accuracy": 0.11508626267313957,
      "num_tokens": 1665832.0,
      "step": 950
    },
    {
      "entropy": 6.508042287826538,
      "epoch": 0.7484326018808778,
      "grad_norm": 1.28125,
      "learning_rate": 0.000477,
      "loss": 6.3476,
      "mean_token_accuracy": 0.11256715059280395,
      "num_tokens": 1675116.0,
      "step": 955
    },
    {
      "entropy": 6.4066481590271,
      "epoch": 0.7523510971786834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004795,
      "loss": 6.3361,
      "mean_token_accuracy": 0.11710453629493714,
      "num_tokens": 1684195.0,
      "step": 960
    },
    {
      "entropy": 6.511843872070313,
      "epoch": 0.7562695924764891,
      "grad_norm": 1.140625,
      "learning_rate": 0.000482,
      "loss": 6.3273,
      "mean_token_accuracy": 0.1193136751651764,
      "num_tokens": 1693025.0,
      "step": 965
    },
    {
      "entropy": 6.339096546173096,
      "epoch": 0.7601880877742947,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004845,
      "loss": 6.3018,
      "mean_token_accuracy": 0.11799748837947846,
      "num_tokens": 1701210.0,
      "step": 970
    },
    {
      "entropy": 6.4773938179016115,
      "epoch": 0.7641065830721003,
      "grad_norm": 1.25,
      "learning_rate": 0.000487,
      "loss": 6.3955,
      "mean_token_accuracy": 0.10802683010697364,
      "num_tokens": 1711026.0,
      "step": 975
    },
    {
      "entropy": 6.531501007080078,
      "epoch": 0.768025078369906,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004895,
      "loss": 6.4253,
      "mean_token_accuracy": 0.116270412504673,
      "num_tokens": 1719441.0,
      "step": 980
    },
    {
      "entropy": 6.4567617893219,
      "epoch": 0.7719435736677116,
      "grad_norm": 1.1875,
      "learning_rate": 0.000492,
      "loss": 6.4179,
      "mean_token_accuracy": 0.11469775289297104,
      "num_tokens": 1728300.0,
      "step": 985
    },
    {
      "entropy": 6.554265642166138,
      "epoch": 0.7758620689655172,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004945,
      "loss": 6.3013,
      "mean_token_accuracy": 0.12283783331513405,
      "num_tokens": 1736364.0,
      "step": 990
    },
    {
      "entropy": 6.199279832839966,
      "epoch": 0.7797805642633229,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497,
      "loss": 6.1391,
      "mean_token_accuracy": 0.12946273759007454,
      "num_tokens": 1745043.0,
      "step": 995
    },
    {
      "entropy": 6.48647174835205,
      "epoch": 0.7836990595611285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995,
      "loss": 6.2994,
      "mean_token_accuracy": 0.12199065089225769,
      "num_tokens": 1754008.0,
      "step": 1000
    },
    {
      "epoch": 0.7836990595611285,
      "eval_entropy": 6.26603299148323,
      "eval_loss": 6.44874906539917,
      "eval_mean_token_accuracy": 0.11973919128176085,
      "eval_num_tokens": 1754008.0,
      "eval_runtime": 2.8517,
      "eval_samples_per_second": 1445.471,
      "eval_steps_per_second": 180.947,
      "step": 1000
    },
    {
      "entropy": 6.291542720794678,
      "epoch": 0.7876175548589341,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999998713243189,
      "loss": 6.3511,
      "mean_token_accuracy": 0.1147798478603363,
      "num_tokens": 1763592.0,
      "step": 1005
    },
    {
      "entropy": 6.429899978637695,
      "epoch": 0.7915360501567398,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999993485796164,
      "loss": 6.2139,
      "mean_token_accuracy": 0.12304601818323135,
      "num_tokens": 1773111.0,
      "step": 1010
    },
    {
      "entropy": 6.388966035842896,
      "epoch": 0.7954545454545454,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004999984237245962,
      "loss": 6.2628,
      "mean_token_accuracy": 0.12534007281064988,
      "num_tokens": 1781230.0,
      "step": 1015
    },
    {
      "entropy": 6.55175461769104,
      "epoch": 0.799373040752351,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004999970967609109,
      "loss": 6.5136,
      "mean_token_accuracy": 0.11764922365546227,
      "num_tokens": 1790324.0,
      "step": 1020
    },
    {
      "entropy": 6.400601530075074,
      "epoch": 0.8032915360501567,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999953676909322,
      "loss": 6.3543,
      "mean_token_accuracy": 0.11396604254841805,
      "num_tokens": 1798849.0,
      "step": 1025
    },
    {
      "entropy": 6.416995716094971,
      "epoch": 0.8072100313479624,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004999932365177502,
      "loss": 6.2295,
      "mean_token_accuracy": 0.12402529492974282,
      "num_tokens": 1808019.0,
      "step": 1030
    },
    {
      "entropy": 6.492223310470581,
      "epoch": 0.8111285266457681,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999907032451735,
      "loss": 6.3318,
      "mean_token_accuracy": 0.11823122128844261,
      "num_tokens": 1816550.0,
      "step": 1035
    },
    {
      "entropy": 6.270820474624633,
      "epoch": 0.8150470219435737,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999877678777296,
      "loss": 6.2441,
      "mean_token_accuracy": 0.127071038633585,
      "num_tokens": 1825238.0,
      "step": 1040
    },
    {
      "entropy": 6.365108633041382,
      "epoch": 0.8189655172413793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999844304206645,
      "loss": 6.2098,
      "mean_token_accuracy": 0.12079643234610557,
      "num_tokens": 1833645.0,
      "step": 1045
    },
    {
      "entropy": 6.404587888717652,
      "epoch": 0.822884012539185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999806908799428,
      "loss": 6.2862,
      "mean_token_accuracy": 0.11846316829323769,
      "num_tokens": 1842892.0,
      "step": 1050
    },
    {
      "entropy": 6.343483209609985,
      "epoch": 0.8268025078369906,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999765492622474,
      "loss": 6.3311,
      "mean_token_accuracy": 0.12146014720201492,
      "num_tokens": 1851400.0,
      "step": 1055
    },
    {
      "entropy": 6.481566047668457,
      "epoch": 0.8307210031347962,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999720055749804,
      "loss": 6.2614,
      "mean_token_accuracy": 0.1218932181596756,
      "num_tokens": 1860088.0,
      "step": 1060
    },
    {
      "entropy": 6.418999338150025,
      "epoch": 0.8346394984326019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999670598262619,
      "loss": 6.3271,
      "mean_token_accuracy": 0.12037511318922042,
      "num_tokens": 1868558.0,
      "step": 1065
    },
    {
      "entropy": 6.369568681716919,
      "epoch": 0.8385579937304075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999617120249308,
      "loss": 6.3198,
      "mean_token_accuracy": 0.11883664280176162,
      "num_tokens": 1877624.0,
      "step": 1070
    },
    {
      "entropy": 6.448858261108398,
      "epoch": 0.8424764890282131,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004999559621805445,
      "loss": 6.4296,
      "mean_token_accuracy": 0.11337628290057182,
      "num_tokens": 1886505.0,
      "step": 1075
    },
    {
      "entropy": 6.279574394226074,
      "epoch": 0.8463949843260188,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999498103033788,
      "loss": 6.1911,
      "mean_token_accuracy": 0.1266665816307068,
      "num_tokens": 1895701.0,
      "step": 1080
    },
    {
      "entropy": 6.333665657043457,
      "epoch": 0.8503134796238244,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999432564044284,
      "loss": 6.1551,
      "mean_token_accuracy": 0.12536471113562583,
      "num_tokens": 1904640.0,
      "step": 1085
    },
    {
      "entropy": 6.3966066360473635,
      "epoch": 0.85423197492163,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999363004954058,
      "loss": 6.3837,
      "mean_token_accuracy": 0.11640885770320893,
      "num_tokens": 1914216.0,
      "step": 1090
    },
    {
      "entropy": 6.347075700759888,
      "epoch": 0.8581504702194357,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999289425887425,
      "loss": 6.2932,
      "mean_token_accuracy": 0.12313793152570725,
      "num_tokens": 1923280.0,
      "step": 1095
    },
    {
      "entropy": 6.454287147521972,
      "epoch": 0.8620689655172413,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999211826975884,
      "loss": 6.3132,
      "mean_token_accuracy": 0.126905470341444,
      "num_tokens": 1931403.0,
      "step": 1100
    },
    {
      "entropy": 6.4443706512451175,
      "epoch": 0.8659874608150471,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999130208358114,
      "loss": 6.3174,
      "mean_token_accuracy": 0.12164701670408248,
      "num_tokens": 1940774.0,
      "step": 1105
    },
    {
      "entropy": 6.394332456588745,
      "epoch": 0.8699059561128527,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999044570179983,
      "loss": 6.4178,
      "mean_token_accuracy": 0.11814009100198745,
      "num_tokens": 1950045.0,
      "step": 1110
    },
    {
      "entropy": 6.467902517318725,
      "epoch": 0.8738244514106583,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998954912594541,
      "loss": 6.2796,
      "mean_token_accuracy": 0.12150320187211036,
      "num_tokens": 1959303.0,
      "step": 1115
    },
    {
      "entropy": 6.338294458389282,
      "epoch": 0.877742946708464,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998861235762018,
      "loss": 6.266,
      "mean_token_accuracy": 0.12415505647659301,
      "num_tokens": 1968583.0,
      "step": 1120
    },
    {
      "entropy": 6.4064007759094235,
      "epoch": 0.8816614420062696,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998763539849832,
      "loss": 6.3821,
      "mean_token_accuracy": 0.11770115494728088,
      "num_tokens": 1977489.0,
      "step": 1125
    },
    {
      "entropy": 6.2882476329803465,
      "epoch": 0.8855799373040752,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998661825032579,
      "loss": 6.1443,
      "mean_token_accuracy": 0.1252014622092247,
      "num_tokens": 1985308.0,
      "step": 1130
    },
    {
      "entropy": 6.315972471237183,
      "epoch": 0.8894984326018809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998556091492043,
      "loss": 6.3124,
      "mean_token_accuracy": 0.11539808958768845,
      "num_tokens": 1993930.0,
      "step": 1135
    },
    {
      "entropy": 6.357517290115356,
      "epoch": 0.8934169278996865,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998446339417184,
      "loss": 6.302,
      "mean_token_accuracy": 0.11564947813749313,
      "num_tokens": 2003802.0,
      "step": 1140
    },
    {
      "entropy": 6.386947774887085,
      "epoch": 0.8973354231974922,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998332569004149,
      "loss": 6.2131,
      "mean_token_accuracy": 0.12580082193017006,
      "num_tokens": 2011957.0,
      "step": 1145
    },
    {
      "entropy": 6.292334413528442,
      "epoch": 0.9012539184952978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998214780456263,
      "loss": 6.2042,
      "mean_token_accuracy": 0.12626376748085022,
      "num_tokens": 2020741.0,
      "step": 1150
    },
    {
      "entropy": 6.330618238449096,
      "epoch": 0.9051724137931034,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998092973984033,
      "loss": 6.2945,
      "mean_token_accuracy": 0.12051898464560509,
      "num_tokens": 2029804.0,
      "step": 1155
    },
    {
      "entropy": 6.426625776290893,
      "epoch": 0.9090909090909091,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997967149805147,
      "loss": 6.3338,
      "mean_token_accuracy": 0.12462769895792007,
      "num_tokens": 2038732.0,
      "step": 1160
    },
    {
      "entropy": 6.363294315338135,
      "epoch": 0.9130094043887147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997837308144474,
      "loss": 6.2391,
      "mean_token_accuracy": 0.13117346838116645,
      "num_tokens": 2046362.0,
      "step": 1165
    },
    {
      "entropy": 6.337823486328125,
      "epoch": 0.9169278996865203,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997703449234062,
      "loss": 6.2264,
      "mean_token_accuracy": 0.12225590869784356,
      "num_tokens": 2056018.0,
      "step": 1170
    },
    {
      "entropy": 6.2902881622314455,
      "epoch": 0.920846394984326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997565573313139,
      "loss": 6.209,
      "mean_token_accuracy": 0.11975079849362373,
      "num_tokens": 2065077.0,
      "step": 1175
    },
    {
      "entropy": 6.288452243804931,
      "epoch": 0.9247648902821317,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997423680628109,
      "loss": 6.1988,
      "mean_token_accuracy": 0.12091932371258736,
      "num_tokens": 2073405.0,
      "step": 1180
    },
    {
      "entropy": 6.159915637969971,
      "epoch": 0.9286833855799373,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499727777143256,
      "loss": 6.159,
      "mean_token_accuracy": 0.12859885394573212,
      "num_tokens": 2081891.0,
      "step": 1185
    },
    {
      "entropy": 6.459240245819092,
      "epoch": 0.932601880877743,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997127845987255,
      "loss": 6.2215,
      "mean_token_accuracy": 0.13154146820306778,
      "num_tokens": 2090732.0,
      "step": 1190
    },
    {
      "entropy": 6.181884670257569,
      "epoch": 0.9365203761755486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996973904560135,
      "loss": 6.165,
      "mean_token_accuracy": 0.13058446869254112,
      "num_tokens": 2099352.0,
      "step": 1195
    },
    {
      "entropy": 6.359483480453491,
      "epoch": 0.9404388714733543,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996815947426317,
      "loss": 6.1959,
      "mean_token_accuracy": 0.12631767764687538,
      "num_tokens": 2107365.0,
      "step": 1200
    },
    {
      "entropy": 6.259723424911499,
      "epoch": 0.9443573667711599,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996653974868097,
      "loss": 6.0728,
      "mean_token_accuracy": 0.12826675027608872,
      "num_tokens": 2116192.0,
      "step": 1205
    },
    {
      "entropy": 6.181558895111084,
      "epoch": 0.9482758620689655,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996487987174946,
      "loss": 6.1668,
      "mean_token_accuracy": 0.1246504932641983,
      "num_tokens": 2124410.0,
      "step": 1210
    },
    {
      "entropy": 6.323344802856445,
      "epoch": 0.9521943573667712,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996317984643512,
      "loss": 6.1916,
      "mean_token_accuracy": 0.12328546494245529,
      "num_tokens": 2132799.0,
      "step": 1215
    },
    {
      "entropy": 6.306202745437622,
      "epoch": 0.9561128526645768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996143967577615,
      "loss": 6.3533,
      "mean_token_accuracy": 0.11768218651413917,
      "num_tokens": 2141589.0,
      "step": 1220
    },
    {
      "entropy": 6.337114143371582,
      "epoch": 0.9600313479623824,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004995965936288254,
      "loss": 6.2465,
      "mean_token_accuracy": 0.1237958587706089,
      "num_tokens": 2150704.0,
      "step": 1225
    },
    {
      "entropy": 6.185662794113159,
      "epoch": 0.9639498432601881,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995783891093597,
      "loss": 6.1759,
      "mean_token_accuracy": 0.12887666672468184,
      "num_tokens": 2159587.0,
      "step": 1230
    },
    {
      "entropy": 6.192087316513062,
      "epoch": 0.9678683385579937,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499559783231899,
      "loss": 6.1353,
      "mean_token_accuracy": 0.12168272957205772,
      "num_tokens": 2169071.0,
      "step": 1235
    },
    {
      "entropy": 6.3143566131591795,
      "epoch": 0.9717868338557993,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995407760296952,
      "loss": 6.2276,
      "mean_token_accuracy": 0.12075230255722999,
      "num_tokens": 2178248.0,
      "step": 1240
    },
    {
      "entropy": 6.30960898399353,
      "epoch": 0.975705329153605,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995213675367169,
      "loss": 6.1378,
      "mean_token_accuracy": 0.11877818033099174,
      "num_tokens": 2187520.0,
      "step": 1245
    },
    {
      "entropy": 6.156886005401612,
      "epoch": 0.9796238244514106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995015577876504,
      "loss": 6.2089,
      "mean_token_accuracy": 0.12378308922052383,
      "num_tokens": 2196861.0,
      "step": 1250
    },
    {
      "entropy": 6.270166254043579,
      "epoch": 0.9835423197492164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499481346817899,
      "loss": 6.13,
      "mean_token_accuracy": 0.1273690603673458,
      "num_tokens": 2205523.0,
      "step": 1255
    },
    {
      "entropy": 6.210124588012695,
      "epoch": 0.987460815047022,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994607346635829,
      "loss": 6.0823,
      "mean_token_accuracy": 0.1346528261899948,
      "num_tokens": 2214291.0,
      "step": 1260
    },
    {
      "entropy": 6.258036661148071,
      "epoch": 0.9913793103448276,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994397213615394,
      "loss": 6.1947,
      "mean_token_accuracy": 0.1263251446187496,
      "num_tokens": 2223093.0,
      "step": 1265
    },
    {
      "entropy": 6.241307115554809,
      "epoch": 0.9952978056426333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994183069493225,
      "loss": 6.1524,
      "mean_token_accuracy": 0.12528934627771376,
      "num_tokens": 2231551.0,
      "step": 1270
    },
    {
      "entropy": 6.143906736373902,
      "epoch": 0.9992163009404389,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993964914652034,
      "loss": 6.1585,
      "mean_token_accuracy": 0.13556302934885026,
      "num_tokens": 2240650.0,
      "step": 1275
    },
    {
      "entropy": 6.344608116149902,
      "epoch": 1.0031347962382444,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993742749481699,
      "loss": 6.0429,
      "mean_token_accuracy": 0.12464264035224915,
      "num_tokens": 2249392.0,
      "step": 1280
    },
    {
      "entropy": 6.091021728515625,
      "epoch": 1.0070532915360502,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993516574379265,
      "loss": 5.8549,
      "mean_token_accuracy": 0.13539742380380632,
      "num_tokens": 2258100.0,
      "step": 1285
    },
    {
      "entropy": 6.187678098678589,
      "epoch": 1.0109717868338557,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993286389748943,
      "loss": 5.9221,
      "mean_token_accuracy": 0.13306491151452066,
      "num_tokens": 2266661.0,
      "step": 1290
    },
    {
      "entropy": 6.146251630783081,
      "epoch": 1.0148902821316614,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993052196002112,
      "loss": 6.0288,
      "mean_token_accuracy": 0.12982963621616364,
      "num_tokens": 2275475.0,
      "step": 1295
    },
    {
      "entropy": 6.161657190322876,
      "epoch": 1.0188087774294672,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992813993557312,
      "loss": 5.9267,
      "mean_token_accuracy": 0.13516829237341882,
      "num_tokens": 2284077.0,
      "step": 1300
    },
    {
      "entropy": 6.11398606300354,
      "epoch": 1.0227272727272727,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499257178284025,
      "loss": 5.9547,
      "mean_token_accuracy": 0.12690627500414847,
      "num_tokens": 2293029.0,
      "step": 1305
    },
    {
      "entropy": 6.120842313766479,
      "epoch": 1.0266457680250785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992325564283797,
      "loss": 5.8673,
      "mean_token_accuracy": 0.14652334675192832,
      "num_tokens": 2301574.0,
      "step": 1310
    },
    {
      "entropy": 6.083501148223877,
      "epoch": 1.030564263322884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992075338327984,
      "loss": 5.8479,
      "mean_token_accuracy": 0.13996412083506585,
      "num_tokens": 2310366.0,
      "step": 1315
    },
    {
      "entropy": 6.188275384902954,
      "epoch": 1.0344827586206897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991821105420006,
      "loss": 5.9953,
      "mean_token_accuracy": 0.12889379039406776,
      "num_tokens": 2318943.0,
      "step": 1320
    },
    {
      "entropy": 6.178983068466186,
      "epoch": 1.0384012539184952,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991562866014219,
      "loss": 5.9852,
      "mean_token_accuracy": 0.13627749904990197,
      "num_tokens": 2327173.0,
      "step": 1325
    },
    {
      "entropy": 6.165381383895874,
      "epoch": 1.042319749216301,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991300620572138,
      "loss": 5.9485,
      "mean_token_accuracy": 0.13148076757788657,
      "num_tokens": 2336194.0,
      "step": 1330
    },
    {
      "entropy": 6.115203237533569,
      "epoch": 1.0462382445141065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991034369562438,
      "loss": 6.0124,
      "mean_token_accuracy": 0.13209957554936408,
      "num_tokens": 2344721.0,
      "step": 1335
    },
    {
      "entropy": 6.131257820129394,
      "epoch": 1.0501567398119123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990764113460953,
      "loss": 5.9498,
      "mean_token_accuracy": 0.13055929765105248,
      "num_tokens": 2353406.0,
      "step": 1340
    },
    {
      "entropy": 6.2284763813018795,
      "epoch": 1.0540752351097178,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990489852750675,
      "loss": 6.11,
      "mean_token_accuracy": 0.1256877563893795,
      "num_tokens": 2361896.0,
      "step": 1345
    },
    {
      "entropy": 6.11944842338562,
      "epoch": 1.0579937304075235,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990211587921751,
      "loss": 5.9414,
      "mean_token_accuracy": 0.13127534314990044,
      "num_tokens": 2371861.0,
      "step": 1350
    },
    {
      "entropy": 6.063726902008057,
      "epoch": 1.061912225705329,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989929319471487,
      "loss": 5.8775,
      "mean_token_accuracy": 0.13494107648730277,
      "num_tokens": 2381555.0,
      "step": 1355
    },
    {
      "entropy": 6.090635299682617,
      "epoch": 1.0658307210031348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989643047904341,
      "loss": 5.8576,
      "mean_token_accuracy": 0.14259101897478105,
      "num_tokens": 2389851.0,
      "step": 1360
    },
    {
      "entropy": 6.012026739120484,
      "epoch": 1.0697492163009406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989352773731928,
      "loss": 5.8025,
      "mean_token_accuracy": 0.1421963430941105,
      "num_tokens": 2397764.0,
      "step": 1365
    },
    {
      "entropy": 6.188090085983276,
      "epoch": 1.073667711598746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989058497473014,
      "loss": 5.9497,
      "mean_token_accuracy": 0.13977306112647056,
      "num_tokens": 2406688.0,
      "step": 1370
    },
    {
      "entropy": 6.121643972396851,
      "epoch": 1.0775862068965518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988760219653518,
      "loss": 6.0132,
      "mean_token_accuracy": 0.13169477209448816,
      "num_tokens": 2415475.0,
      "step": 1375
    },
    {
      "entropy": 6.123086738586426,
      "epoch": 1.0815047021943573,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988457940806513,
      "loss": 5.9703,
      "mean_token_accuracy": 0.12858632132411002,
      "num_tokens": 2425719.0,
      "step": 1380
    },
    {
      "entropy": 6.025482082366944,
      "epoch": 1.085423197492163,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988151661472218,
      "loss": 5.8792,
      "mean_token_accuracy": 0.13557613492012024,
      "num_tokens": 2435138.0,
      "step": 1385
    },
    {
      "entropy": 6.151245784759522,
      "epoch": 1.0893416927899686,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987841382198006,
      "loss": 6.0136,
      "mean_token_accuracy": 0.13455865979194642,
      "num_tokens": 2443976.0,
      "step": 1390
    },
    {
      "entropy": 5.99044942855835,
      "epoch": 1.0932601880877744,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987527103538394,
      "loss": 5.9018,
      "mean_token_accuracy": 0.13744083419442177,
      "num_tokens": 2453287.0,
      "step": 1395
    },
    {
      "entropy": 6.198356246948242,
      "epoch": 1.09717868338558,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498720882605505,
      "loss": 6.0023,
      "mean_token_accuracy": 0.12969460785388948,
      "num_tokens": 2462054.0,
      "step": 1400
    },
    {
      "entropy": 6.005903244018555,
      "epoch": 1.1010971786833856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004986886550316788,
      "loss": 5.8813,
      "mean_token_accuracy": 0.14015732035040857,
      "num_tokens": 2470333.0,
      "step": 1405
    },
    {
      "entropy": 6.0531473636627195,
      "epoch": 1.1050156739811912,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986560276899565,
      "loss": 5.9526,
      "mean_token_accuracy": 0.13291476517915726,
      "num_tokens": 2479621.0,
      "step": 1410
    },
    {
      "entropy": 6.125274991989135,
      "epoch": 1.108934169278997,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986230006386488,
      "loss": 5.9006,
      "mean_token_accuracy": 0.13800743818283082,
      "num_tokens": 2488020.0,
      "step": 1415
    },
    {
      "entropy": 6.111095190048218,
      "epoch": 1.1128526645768024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985895739367802,
      "loss": 5.9408,
      "mean_token_accuracy": 0.13480502888560295,
      "num_tokens": 2497779.0,
      "step": 1420
    },
    {
      "entropy": 6.066828060150146,
      "epoch": 1.1167711598746082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004985557476440895,
      "loss": 5.8817,
      "mean_token_accuracy": 0.1367568626999855,
      "num_tokens": 2506791.0,
      "step": 1425
    },
    {
      "entropy": 6.096922731399536,
      "epoch": 1.1206896551724137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00049852152182103,
      "loss": 5.9866,
      "mean_token_accuracy": 0.13348396867513657,
      "num_tokens": 2515576.0,
      "step": 1430
    },
    {
      "entropy": 6.147091960906982,
      "epoch": 1.1246081504702194,
      "grad_norm": 1.25,
      "learning_rate": 0.0004984868965287686,
      "loss": 5.9147,
      "mean_token_accuracy": 0.136567921936512,
      "num_tokens": 2523933.0,
      "step": 1435
    },
    {
      "entropy": 6.00162410736084,
      "epoch": 1.1285266457680252,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984518718291864,
      "loss": 5.8647,
      "mean_token_accuracy": 0.1318059407174587,
      "num_tokens": 2532453.0,
      "step": 1440
    },
    {
      "entropy": 6.0632120132446286,
      "epoch": 1.1324451410658307,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984164477848783,
      "loss": 5.9105,
      "mean_token_accuracy": 0.13608234971761704,
      "num_tokens": 2541149.0,
      "step": 1445
    },
    {
      "entropy": 6.078012609481812,
      "epoch": 1.1363636363636362,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983806244591528,
      "loss": 5.8876,
      "mean_token_accuracy": 0.14004134014248848,
      "num_tokens": 2550302.0,
      "step": 1450
    },
    {
      "entropy": 5.941012239456176,
      "epoch": 1.140282131661442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983444019160318,
      "loss": 5.7526,
      "mean_token_accuracy": 0.14576120525598527,
      "num_tokens": 2558507.0,
      "step": 1455
    },
    {
      "entropy": 6.106155681610107,
      "epoch": 1.1442006269592477,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004983077802202511,
      "loss": 6.0189,
      "mean_token_accuracy": 0.13469743952155114,
      "num_tokens": 2567487.0,
      "step": 1460
    },
    {
      "entropy": 6.060013628005981,
      "epoch": 1.1481191222570533,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982707594372595,
      "loss": 5.8779,
      "mean_token_accuracy": 0.1318593814969063,
      "num_tokens": 2576232.0,
      "step": 1465
    },
    {
      "entropy": 6.010817146301269,
      "epoch": 1.152037617554859,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498233339633219,
      "loss": 5.8846,
      "mean_token_accuracy": 0.1372211568057537,
      "num_tokens": 2585081.0,
      "step": 1470
    },
    {
      "entropy": 6.013401889801026,
      "epoch": 1.1559561128526645,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498195520875005,
      "loss": 5.858,
      "mean_token_accuracy": 0.1376308247447014,
      "num_tokens": 2593890.0,
      "step": 1475
    },
    {
      "entropy": 6.076926422119141,
      "epoch": 1.1598746081504703,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981573032302059,
      "loss": 5.9343,
      "mean_token_accuracy": 0.13154210969805719,
      "num_tokens": 2602011.0,
      "step": 1480
    },
    {
      "entropy": 6.001340532302857,
      "epoch": 1.1637931034482758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981186867671225,
      "loss": 5.8382,
      "mean_token_accuracy": 0.1394493967294693,
      "num_tokens": 2610366.0,
      "step": 1485
    },
    {
      "entropy": 5.969131422042847,
      "epoch": 1.1677115987460815,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980796715547691,
      "loss": 5.8465,
      "mean_token_accuracy": 0.14186024740338327,
      "num_tokens": 2619737.0,
      "step": 1490
    },
    {
      "entropy": 6.0836954593658445,
      "epoch": 1.171630094043887,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980402576628717,
      "loss": 5.8896,
      "mean_token_accuracy": 0.13914738819003106,
      "num_tokens": 2628176.0,
      "step": 1495
    },
    {
      "entropy": 5.938668918609619,
      "epoch": 1.1755485893416928,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980004451618697,
      "loss": 5.8927,
      "mean_token_accuracy": 0.13917603492736816,
      "num_tokens": 2636195.0,
      "step": 1500
    },
    {
      "epoch": 1.1755485893416928,
      "eval_entropy": 5.986282704412475,
      "eval_loss": 6.191744327545166,
      "eval_mean_token_accuracy": 0.13095851831657943,
      "eval_num_tokens": 2636195.0,
      "eval_runtime": 2.853,
      "eval_samples_per_second": 1444.817,
      "eval_steps_per_second": 180.865,
      "step": 1500
    },
    {
      "entropy": 6.155323219299317,
      "epoch": 1.1794670846394983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979602341229144,
      "loss": 5.9544,
      "mean_token_accuracy": 0.13536008521914483,
      "num_tokens": 2645035.0,
      "step": 1505
    },
    {
      "entropy": 6.0477063179016115,
      "epoch": 1.183385579937304,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979196246178694,
      "loss": 5.9375,
      "mean_token_accuracy": 0.13434076979756354,
      "num_tokens": 2653239.0,
      "step": 1510
    },
    {
      "entropy": 5.962043952941895,
      "epoch": 1.1873040752351098,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004978786167193105,
      "loss": 5.9298,
      "mean_token_accuracy": 0.13612919598817824,
      "num_tokens": 2662532.0,
      "step": 1515
    },
    {
      "entropy": 6.097779035568237,
      "epoch": 1.1912225705329154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004978372105005254,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1357118271291256,
      "num_tokens": 2671867.0,
      "step": 1520
    },
    {
      "entropy": 5.931209087371826,
      "epoch": 1.1951410658307209,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977954060355139,
      "loss": 5.8616,
      "mean_token_accuracy": 0.1355850227177143,
      "num_tokens": 2680693.0,
      "step": 1525
    },
    {
      "entropy": 6.132346057891846,
      "epoch": 1.1990595611285266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977532033989871,
      "loss": 5.9434,
      "mean_token_accuracy": 0.13678196668624878,
      "num_tokens": 2689228.0,
      "step": 1530
    },
    {
      "entropy": 5.966204833984375,
      "epoch": 1.2029780564263324,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977106026663681,
      "loss": 5.8242,
      "mean_token_accuracy": 0.13174520581960678,
      "num_tokens": 2698483.0,
      "step": 1535
    },
    {
      "entropy": 6.072604417800903,
      "epoch": 1.206896551724138,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976676039137914,
      "loss": 5.9645,
      "mean_token_accuracy": 0.1317337729036808,
      "num_tokens": 2706920.0,
      "step": 1540
    },
    {
      "entropy": 6.017421102523803,
      "epoch": 1.2108150470219436,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004976242072181026,
      "loss": 5.8541,
      "mean_token_accuracy": 0.14151965230703353,
      "num_tokens": 2715954.0,
      "step": 1545
    },
    {
      "entropy": 5.9726134777069095,
      "epoch": 1.2147335423197492,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975804126568587,
      "loss": 5.8649,
      "mean_token_accuracy": 0.14412466660141945,
      "num_tokens": 2724890.0,
      "step": 1550
    },
    {
      "entropy": 6.04607458114624,
      "epoch": 1.218652037617555,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004975362203083277,
      "loss": 5.8895,
      "mean_token_accuracy": 0.13369582816958428,
      "num_tokens": 2733688.0,
      "step": 1555
    },
    {
      "entropy": 6.079061126708984,
      "epoch": 1.2225705329153604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004974916302514886,
      "loss": 5.9016,
      "mean_token_accuracy": 0.13256118297576905,
      "num_tokens": 2743364.0,
      "step": 1560
    },
    {
      "entropy": 5.943713235855102,
      "epoch": 1.2264890282131662,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004974466425660307,
      "loss": 5.9392,
      "mean_token_accuracy": 0.12959515154361725,
      "num_tokens": 2754276.0,
      "step": 1565
    },
    {
      "entropy": 6.054774904251099,
      "epoch": 1.2304075235109717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004974012573323545,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1476442575454712,
      "num_tokens": 2762680.0,
      "step": 1570
    },
    {
      "entropy": 5.864394569396973,
      "epoch": 1.2343260188087775,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973554746315709,
      "loss": 5.8482,
      "mean_token_accuracy": 0.14662062674760817,
      "num_tokens": 2771161.0,
      "step": 1575
    },
    {
      "entropy": 6.048675584793091,
      "epoch": 1.238244514106583,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004973092945455009,
      "loss": 5.8051,
      "mean_token_accuracy": 0.14707713723182678,
      "num_tokens": 2779321.0,
      "step": 1580
    },
    {
      "entropy": 5.943394899368286,
      "epoch": 1.2421630094043887,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004972627171566757,
      "loss": 5.9796,
      "mean_token_accuracy": 0.13489320799708365,
      "num_tokens": 2788425.0,
      "step": 1585
    },
    {
      "entropy": 5.981634616851807,
      "epoch": 1.2460815047021945,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004972157425483368,
      "loss": 5.8248,
      "mean_token_accuracy": 0.14254749417304993,
      "num_tokens": 2796747.0,
      "step": 1590
    },
    {
      "entropy": 5.935612487792969,
      "epoch": 1.25,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004971683708044353,
      "loss": 5.8731,
      "mean_token_accuracy": 0.14065025225281716,
      "num_tokens": 2805335.0,
      "step": 1595
    },
    {
      "entropy": 6.01358585357666,
      "epoch": 1.2539184952978055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004971206020096323,
      "loss": 5.7607,
      "mean_token_accuracy": 0.14524763226509094,
      "num_tokens": 2813350.0,
      "step": 1600
    },
    {
      "entropy": 6.021692228317261,
      "epoch": 1.2578369905956113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004970724362492984,
      "loss": 5.9608,
      "mean_token_accuracy": 0.13947487324476243,
      "num_tokens": 2821922.0,
      "step": 1605
    },
    {
      "entropy": 5.959397029876709,
      "epoch": 1.261755485893417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970238736095135,
      "loss": 5.8589,
      "mean_token_accuracy": 0.1373360872268677,
      "num_tokens": 2830832.0,
      "step": 1610
    },
    {
      "entropy": 6.025397825241089,
      "epoch": 1.2656739811912225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969749141770671,
      "loss": 5.8598,
      "mean_token_accuracy": 0.1413205847144127,
      "num_tokens": 2839699.0,
      "step": 1615
    },
    {
      "entropy": 5.997832584381103,
      "epoch": 1.2695924764890283,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004969255580394575,
      "loss": 5.9218,
      "mean_token_accuracy": 0.14140960872173308,
      "num_tokens": 2848293.0,
      "step": 1620
    },
    {
      "entropy": 6.035713386535645,
      "epoch": 1.2735109717868338,
      "grad_norm": 1.171875,
      "learning_rate": 0.000496875805284892,
      "loss": 5.8727,
      "mean_token_accuracy": 0.14097950309515,
      "num_tokens": 2856468.0,
      "step": 1625
    },
    {
      "entropy": 5.976678323745728,
      "epoch": 1.2774294670846396,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968256560022871,
      "loss": 5.8147,
      "mean_token_accuracy": 0.14333799257874488,
      "num_tokens": 2865366.0,
      "step": 1630
    },
    {
      "entropy": 5.939387321472168,
      "epoch": 1.281347962382445,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004967751102812676,
      "loss": 5.8761,
      "mean_token_accuracy": 0.14351602345705033,
      "num_tokens": 2873520.0,
      "step": 1635
    },
    {
      "entropy": 6.00247712135315,
      "epoch": 1.2852664576802508,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004967241682121669,
      "loss": 5.9385,
      "mean_token_accuracy": 0.13742373287677764,
      "num_tokens": 2883324.0,
      "step": 1640
    },
    {
      "entropy": 6.022455883026123,
      "epoch": 1.2891849529780564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004966728298860267,
      "loss": 5.8665,
      "mean_token_accuracy": 0.1368059866130352,
      "num_tokens": 2892200.0,
      "step": 1645
    },
    {
      "entropy": 5.961077690124512,
      "epoch": 1.293103448275862,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004966210953945969,
      "loss": 5.8382,
      "mean_token_accuracy": 0.1419058546423912,
      "num_tokens": 2900597.0,
      "step": 1650
    },
    {
      "entropy": 6.017272663116455,
      "epoch": 1.2970219435736676,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965689648303355,
      "loss": 5.8791,
      "mean_token_accuracy": 0.1380735568702221,
      "num_tokens": 2909869.0,
      "step": 1655
    },
    {
      "entropy": 5.8662271976470945,
      "epoch": 1.3009404388714734,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004965164382864083,
      "loss": 5.8916,
      "mean_token_accuracy": 0.14124979376792907,
      "num_tokens": 2919166.0,
      "step": 1660
    },
    {
      "entropy": 6.08414101600647,
      "epoch": 1.3048589341692791,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004964635158566886,
      "loss": 5.8954,
      "mean_token_accuracy": 0.14140134826302528,
      "num_tokens": 2927884.0,
      "step": 1665
    },
    {
      "entropy": 5.943005084991455,
      "epoch": 1.3087774294670846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004964101976357574,
      "loss": 5.8419,
      "mean_token_accuracy": 0.14933276697993278,
      "num_tokens": 2936737.0,
      "step": 1670
    },
    {
      "entropy": 5.990343284606934,
      "epoch": 1.3126959247648902,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000496356483718903,
      "loss": 5.9346,
      "mean_token_accuracy": 0.1422124423086643,
      "num_tokens": 2944203.0,
      "step": 1675
    },
    {
      "entropy": 6.030170917510986,
      "epoch": 1.316614420062696,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004963023742021211,
      "loss": 5.9368,
      "mean_token_accuracy": 0.13771088048815727,
      "num_tokens": 2952639.0,
      "step": 1680
    },
    {
      "entropy": 5.980809211730957,
      "epoch": 1.3205329153605017,
      "grad_norm": 1.125,
      "learning_rate": 0.000496247869182114,
      "loss": 5.8751,
      "mean_token_accuracy": 0.13748376145958902,
      "num_tokens": 2961482.0,
      "step": 1685
    },
    {
      "entropy": 5.976028776168823,
      "epoch": 1.3244514106583072,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004961929687562909,
      "loss": 5.8906,
      "mean_token_accuracy": 0.13681933358311654,
      "num_tokens": 2970927.0,
      "step": 1690
    },
    {
      "entropy": 5.919792747497558,
      "epoch": 1.328369905956113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004961376730227681,
      "loss": 5.8038,
      "mean_token_accuracy": 0.14354753047227858,
      "num_tokens": 2979329.0,
      "step": 1695
    },
    {
      "entropy": 5.945391416549683,
      "epoch": 1.3322884012539185,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960819820803675,
      "loss": 5.7922,
      "mean_token_accuracy": 0.14268116950988768,
      "num_tokens": 2986894.0,
      "step": 1700
    },
    {
      "entropy": 5.93856897354126,
      "epoch": 1.3362068965517242,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960258960286185,
      "loss": 5.7954,
      "mean_token_accuracy": 0.14203962236642836,
      "num_tokens": 2994721.0,
      "step": 1705
    },
    {
      "entropy": 5.9834705829620365,
      "epoch": 1.3401253918495297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004959694149677553,
      "loss": 5.9459,
      "mean_token_accuracy": 0.13363023549318315,
      "num_tokens": 3003165.0,
      "step": 1710
    },
    {
      "entropy": 6.01661696434021,
      "epoch": 1.3440438871473355,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959125389987193,
      "loss": 5.9257,
      "mean_token_accuracy": 0.13172226995229722,
      "num_tokens": 3012178.0,
      "step": 1715
    },
    {
      "entropy": 5.932300090789795,
      "epoch": 1.347962382445141,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004958552682231567,
      "loss": 5.8234,
      "mean_token_accuracy": 0.14171281978487968,
      "num_tokens": 3021685.0,
      "step": 1720
    },
    {
      "entropy": 5.928375720977783,
      "epoch": 1.3518808777429467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004957976027434199,
      "loss": 5.8741,
      "mean_token_accuracy": 0.13452255502343177,
      "num_tokens": 3030667.0,
      "step": 1725
    },
    {
      "entropy": 6.019004249572754,
      "epoch": 1.3557993730407523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004957395426625663,
      "loss": 5.9333,
      "mean_token_accuracy": 0.13655466735363006,
      "num_tokens": 3039501.0,
      "step": 1730
    },
    {
      "entropy": 5.933786916732788,
      "epoch": 1.359717868338558,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004956810880843587,
      "loss": 5.8624,
      "mean_token_accuracy": 0.13548908457159997,
      "num_tokens": 3048547.0,
      "step": 1735
    },
    {
      "entropy": 6.0302653312683105,
      "epoch": 1.3636363636363638,
      "grad_norm": 1.21875,
      "learning_rate": 0.000495622239113265,
      "loss": 5.8121,
      "mean_token_accuracy": 0.14338299483060837,
      "num_tokens": 3056521.0,
      "step": 1740
    },
    {
      "entropy": 5.849083042144775,
      "epoch": 1.3675548589341693,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955629958544577,
      "loss": 5.8126,
      "mean_token_accuracy": 0.14086603075265886,
      "num_tokens": 3066007.0,
      "step": 1745
    },
    {
      "entropy": 6.053999328613282,
      "epoch": 1.3714733542319748,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004955033584138143,
      "loss": 6.0003,
      "mean_token_accuracy": 0.12978003472089766,
      "num_tokens": 3075111.0,
      "step": 1750
    },
    {
      "entropy": 5.875246000289917,
      "epoch": 1.3753918495297806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004954433268979164,
      "loss": 5.7881,
      "mean_token_accuracy": 0.152919040620327,
      "num_tokens": 3083766.0,
      "step": 1755
    },
    {
      "entropy": 5.886160612106323,
      "epoch": 1.3793103448275863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953829014140502,
      "loss": 5.8577,
      "mean_token_accuracy": 0.13894471079111098,
      "num_tokens": 3092103.0,
      "step": 1760
    },
    {
      "entropy": 5.972563314437866,
      "epoch": 1.3832288401253918,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004953220820702057,
      "loss": 5.8714,
      "mean_token_accuracy": 0.13644422814249993,
      "num_tokens": 3101286.0,
      "step": 1765
    },
    {
      "entropy": 6.092113876342774,
      "epoch": 1.3871473354231976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952608689750771,
      "loss": 5.8627,
      "mean_token_accuracy": 0.1330120176076889,
      "num_tokens": 3110554.0,
      "step": 1770
    },
    {
      "entropy": 5.918315887451172,
      "epoch": 1.391065830721003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004951992622380619,
      "loss": 5.7475,
      "mean_token_accuracy": 0.1461511582136154,
      "num_tokens": 3119265.0,
      "step": 1775
    },
    {
      "entropy": 5.870513153076172,
      "epoch": 1.3949843260188088,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004951372619692615,
      "loss": 5.7627,
      "mean_token_accuracy": 0.14474907889962196,
      "num_tokens": 3127962.0,
      "step": 1780
    },
    {
      "entropy": 5.979340076446533,
      "epoch": 1.3989028213166144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004950748682794804,
      "loss": 5.9175,
      "mean_token_accuracy": 0.13927276879549028,
      "num_tokens": 3137265.0,
      "step": 1785
    },
    {
      "entropy": 5.829528379440307,
      "epoch": 1.40282131661442,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004950120812802262,
      "loss": 5.7332,
      "mean_token_accuracy": 0.14469338655471803,
      "num_tokens": 3146458.0,
      "step": 1790
    },
    {
      "entropy": 6.006521940231323,
      "epoch": 1.4067398119122256,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004949489010837095,
      "loss": 5.8905,
      "mean_token_accuracy": 0.13321248143911363,
      "num_tokens": 3155557.0,
      "step": 1795
    },
    {
      "entropy": 5.917591524124146,
      "epoch": 1.4106583072100314,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004948853278028436,
      "loss": 5.7598,
      "mean_token_accuracy": 0.15007541924715043,
      "num_tokens": 3163478.0,
      "step": 1800
    },
    {
      "entropy": 5.903986883163452,
      "epoch": 1.414576802507837,
      "grad_norm": 1.171875,
      "learning_rate": 0.000494821361551244,
      "loss": 5.8502,
      "mean_token_accuracy": 0.13634541109204293,
      "num_tokens": 3172632.0,
      "step": 1805
    },
    {
      "entropy": 5.879254722595215,
      "epoch": 1.4184952978056427,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004947570024432291,
      "loss": 5.897,
      "mean_token_accuracy": 0.14529131725430489,
      "num_tokens": 3181325.0,
      "step": 1810
    },
    {
      "entropy": 5.980233764648437,
      "epoch": 1.4224137931034484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004946922505938189,
      "loss": 5.9481,
      "mean_token_accuracy": 0.14158461317420007,
      "num_tokens": 3189783.0,
      "step": 1815
    },
    {
      "entropy": 5.982886266708374,
      "epoch": 1.426332288401254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004946271061187354,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1370651327073574,
      "num_tokens": 3198605.0,
      "step": 1820
    },
    {
      "entropy": 5.875323295593262,
      "epoch": 1.4302507836990594,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945615691344025,
      "loss": 5.8894,
      "mean_token_accuracy": 0.13310048058629037,
      "num_tokens": 3208061.0,
      "step": 1825
    },
    {
      "entropy": 5.941305637359619,
      "epoch": 1.4341692789968652,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004944956397579453,
      "loss": 5.7912,
      "mean_token_accuracy": 0.14802727848291397,
      "num_tokens": 3216093.0,
      "step": 1830
    },
    {
      "entropy": 5.901020240783692,
      "epoch": 1.438087774294671,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004944293181071902,
      "loss": 5.7855,
      "mean_token_accuracy": 0.14875897616147996,
      "num_tokens": 3224446.0,
      "step": 1835
    },
    {
      "entropy": 5.942932271957398,
      "epoch": 1.4420062695924765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004943626043006649,
      "loss": 5.8867,
      "mean_token_accuracy": 0.13621754497289656,
      "num_tokens": 3233398.0,
      "step": 1840
    },
    {
      "entropy": 5.9624693393707275,
      "epoch": 1.4459247648902822,
      "grad_norm": 1.109375,
      "learning_rate": 0.000494295498457598,
      "loss": 5.9052,
      "mean_token_accuracy": 0.13407745957374573,
      "num_tokens": 3243391.0,
      "step": 1845
    },
    {
      "entropy": 5.926957750320435,
      "epoch": 1.4498432601880877,
      "grad_norm": 1.125,
      "learning_rate": 0.0004942280006979179,
      "loss": 5.84,
      "mean_token_accuracy": 0.14388877302408218,
      "num_tokens": 3251954.0,
      "step": 1850
    },
    {
      "entropy": 5.921461391448974,
      "epoch": 1.4537617554858935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004941601111422546,
      "loss": 5.8504,
      "mean_token_accuracy": 0.14268624037504196,
      "num_tokens": 3261121.0,
      "step": 1855
    },
    {
      "entropy": 5.826611518859863,
      "epoch": 1.457680250783699,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004940918299119375,
      "loss": 5.7798,
      "mean_token_accuracy": 0.14540517330169678,
      "num_tokens": 3270179.0,
      "step": 1860
    },
    {
      "entropy": 5.817315769195557,
      "epoch": 1.4615987460815048,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004940231571289962,
      "loss": 5.7913,
      "mean_token_accuracy": 0.13905104100704194,
      "num_tokens": 3279749.0,
      "step": 1865
    },
    {
      "entropy": 5.950955724716186,
      "epoch": 1.4655172413793103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004939540929161603,
      "loss": 5.7957,
      "mean_token_accuracy": 0.1480885237455368,
      "num_tokens": 3288458.0,
      "step": 1870
    },
    {
      "entropy": 5.870092535018921,
      "epoch": 1.469435736677116,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004938846373968586,
      "loss": 5.7682,
      "mean_token_accuracy": 0.14406080767512322,
      "num_tokens": 3297231.0,
      "step": 1875
    },
    {
      "entropy": 5.94631724357605,
      "epoch": 1.4733542319749215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004938147906952194,
      "loss": 5.9334,
      "mean_token_accuracy": 0.14153699949383736,
      "num_tokens": 3306915.0,
      "step": 1880
    },
    {
      "entropy": 5.926647710800171,
      "epoch": 1.4772727272727273,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049374455293607,
      "loss": 5.886,
      "mean_token_accuracy": 0.13828219845891,
      "num_tokens": 3315985.0,
      "step": 1885
    },
    {
      "entropy": 5.947432613372802,
      "epoch": 1.481191222570533,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004936739242449369,
      "loss": 5.8505,
      "mean_token_accuracy": 0.13950431123375892,
      "num_tokens": 3324924.0,
      "step": 1890
    },
    {
      "entropy": 5.935120105743408,
      "epoch": 1.4851097178683386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004936029047480447,
      "loss": 5.8462,
      "mean_token_accuracy": 0.1395495243370533,
      "num_tokens": 3333464.0,
      "step": 1895
    },
    {
      "entropy": 5.889332723617554,
      "epoch": 1.489028213166144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004935314945723171,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1458078570663929,
      "num_tokens": 3342441.0,
      "step": 1900
    },
    {
      "entropy": 5.865824174880982,
      "epoch": 1.4929467084639498,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004934596938453754,
      "loss": 5.801,
      "mean_token_accuracy": 0.13902750983834267,
      "num_tokens": 3351338.0,
      "step": 1905
    },
    {
      "entropy": 5.9495453357696535,
      "epoch": 1.4968652037617556,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004933875026955391,
      "loss": 5.7673,
      "mean_token_accuracy": 0.14163718521595,
      "num_tokens": 3360356.0,
      "step": 1910
    },
    {
      "entropy": 5.913040637969971,
      "epoch": 1.500783699059561,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004933149212518258,
      "loss": 5.869,
      "mean_token_accuracy": 0.1361314684152603,
      "num_tokens": 3369485.0,
      "step": 1915
    },
    {
      "entropy": 5.876099491119385,
      "epoch": 1.5047021943573666,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004932419496439501,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1474532313644886,
      "num_tokens": 3378355.0,
      "step": 1920
    },
    {
      "entropy": 5.891150426864624,
      "epoch": 1.5086206896551724,
      "grad_norm": 1.125,
      "learning_rate": 0.000493168588002324,
      "loss": 5.7252,
      "mean_token_accuracy": 0.14671744257211686,
      "num_tokens": 3386729.0,
      "step": 1925
    },
    {
      "entropy": 6.020740795135498,
      "epoch": 1.5125391849529781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004930948364580569,
      "loss": 5.9521,
      "mean_token_accuracy": 0.13383505940437318,
      "num_tokens": 3395822.0,
      "step": 1930
    },
    {
      "entropy": 5.8799830913543705,
      "epoch": 1.5164576802507836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004930206951429546,
      "loss": 5.8552,
      "mean_token_accuracy": 0.14399714767932892,
      "num_tokens": 3404671.0,
      "step": 1935
    },
    {
      "entropy": 5.891398096084595,
      "epoch": 1.5203761755485894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004929461641895197,
      "loss": 5.7791,
      "mean_token_accuracy": 0.14337211176753045,
      "num_tokens": 3413289.0,
      "step": 1940
    },
    {
      "entropy": 5.854188442230225,
      "epoch": 1.5242946708463951,
      "grad_norm": 1.203125,
      "learning_rate": 0.000492871243730951,
      "loss": 5.8445,
      "mean_token_accuracy": 0.14026380628347396,
      "num_tokens": 3421960.0,
      "step": 1945
    },
    {
      "entropy": 5.937834072113037,
      "epoch": 1.5282131661442007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004927959339011437,
      "loss": 5.839,
      "mean_token_accuracy": 0.13955261707305908,
      "num_tokens": 3431958.0,
      "step": 1950
    },
    {
      "entropy": 5.740097761154175,
      "epoch": 1.5321316614420062,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004927202348346885,
      "loss": 5.6604,
      "mean_token_accuracy": 0.1493469387292862,
      "num_tokens": 3441438.0,
      "step": 1955
    },
    {
      "entropy": 5.925070333480835,
      "epoch": 1.536050156739812,
      "grad_norm": 1.109375,
      "learning_rate": 0.000492644146666872,
      "loss": 5.7737,
      "mean_token_accuracy": 0.15302247405052186,
      "num_tokens": 3450176.0,
      "step": 1960
    },
    {
      "entropy": 5.784520673751831,
      "epoch": 1.5399686520376177,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925676695336761,
      "loss": 5.7783,
      "mean_token_accuracy": 0.1460975795984268,
      "num_tokens": 3458685.0,
      "step": 1965
    },
    {
      "entropy": 5.939446878433228,
      "epoch": 1.5438871473354232,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004924908035717777,
      "loss": 5.9218,
      "mean_token_accuracy": 0.13721313923597336,
      "num_tokens": 3468362.0,
      "step": 1970
    },
    {
      "entropy": 5.996397972106934,
      "epoch": 1.5478056426332287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000492413548918549,
      "loss": 5.8444,
      "mean_token_accuracy": 0.14055879041552544,
      "num_tokens": 3476791.0,
      "step": 1975
    },
    {
      "entropy": 5.791240644454956,
      "epoch": 1.5517241379310345,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004923359057120563,
      "loss": 5.7567,
      "mean_token_accuracy": 0.14039518237113952,
      "num_tokens": 3486486.0,
      "step": 1980
    },
    {
      "entropy": 5.824171447753907,
      "epoch": 1.5556426332288402,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004922578740910608,
      "loss": 5.7176,
      "mean_token_accuracy": 0.15190425664186477,
      "num_tokens": 3496145.0,
      "step": 1985
    },
    {
      "entropy": 5.760051012039185,
      "epoch": 1.5595611285266457,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004921794541950177,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1465641237795353,
      "num_tokens": 3504644.0,
      "step": 1990
    },
    {
      "entropy": 5.8565061569213865,
      "epoch": 1.5634796238244513,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004921006461640758,
      "loss": 5.7313,
      "mean_token_accuracy": 0.14198062419891358,
      "num_tokens": 3513126.0,
      "step": 1995
    },
    {
      "entropy": 5.699295711517334,
      "epoch": 1.567398119122257,
      "grad_norm": 1.34375,
      "learning_rate": 0.000492021450139078,
      "loss": 5.6865,
      "mean_token_accuracy": 0.1480669930577278,
      "num_tokens": 3521505.0,
      "step": 2000
    },
    {
      "epoch": 1.567398119122257,
      "eval_entropy": 5.672700917073922,
      "eval_loss": 6.029964923858643,
      "eval_mean_token_accuracy": 0.13995708463092646,
      "eval_num_tokens": 3521505.0,
      "eval_runtime": 2.855,
      "eval_samples_per_second": 1443.769,
      "eval_steps_per_second": 180.734,
      "step": 2000
    },
    {
      "entropy": 5.9574120998382565,
      "epoch": 1.5713166144200628,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004919418662615605,
      "loss": 5.9376,
      "mean_token_accuracy": 0.13867377862334251,
      "num_tokens": 3530008.0,
      "step": 2005
    },
    {
      "entropy": 5.9297913074493405,
      "epoch": 1.5752351097178683,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004918618946737525,
      "loss": 5.8213,
      "mean_token_accuracy": 0.14330511540174484,
      "num_tokens": 3538778.0,
      "step": 2010
    },
    {
      "entropy": 5.865578556060791,
      "epoch": 1.579153605015674,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917815355185762,
      "loss": 5.7886,
      "mean_token_accuracy": 0.1410187527537346,
      "num_tokens": 3547108.0,
      "step": 2015
    },
    {
      "entropy": 5.835587406158448,
      "epoch": 1.5830721003134798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004917007889396464,
      "loss": 5.837,
      "mean_token_accuracy": 0.1368314415216446,
      "num_tokens": 3556094.0,
      "step": 2020
    },
    {
      "entropy": 5.858216381072998,
      "epoch": 1.5869905956112853,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004916196550812706,
      "loss": 5.6885,
      "mean_token_accuracy": 0.1482911989092827,
      "num_tokens": 3564222.0,
      "step": 2025
    },
    {
      "entropy": 5.711031723022461,
      "epoch": 1.5909090909090908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004915381340884477,
      "loss": 5.7125,
      "mean_token_accuracy": 0.15178643986582757,
      "num_tokens": 3573123.0,
      "step": 2030
    },
    {
      "entropy": 5.912963628768921,
      "epoch": 1.5948275862068966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004914562261068693,
      "loss": 5.7347,
      "mean_token_accuracy": 0.14733487069606782,
      "num_tokens": 3581462.0,
      "step": 2035
    },
    {
      "entropy": 5.7520280361175535,
      "epoch": 1.5987460815047023,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004913739312829181,
      "loss": 5.7551,
      "mean_token_accuracy": 0.1488973803818226,
      "num_tokens": 3590718.0,
      "step": 2040
    },
    {
      "entropy": 5.812071180343628,
      "epoch": 1.6026645768025078,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004912912497636683,
      "loss": 5.7788,
      "mean_token_accuracy": 0.14240214824676514,
      "num_tokens": 3599890.0,
      "step": 2045
    },
    {
      "entropy": 5.9684983253479,
      "epoch": 1.6065830721003134,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004912081816968853,
      "loss": 5.8616,
      "mean_token_accuracy": 0.14224711507558824,
      "num_tokens": 3607927.0,
      "step": 2050
    },
    {
      "entropy": 5.866142892837525,
      "epoch": 1.6105015673981191,
      "grad_norm": 1.15625,
      "learning_rate": 0.000491124727231025,
      "loss": 5.7005,
      "mean_token_accuracy": 0.15074294060468674,
      "num_tokens": 3615935.0,
      "step": 2055
    },
    {
      "entropy": 5.735476112365722,
      "epoch": 1.6144200626959249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004910408865152343,
      "loss": 5.7498,
      "mean_token_accuracy": 0.14708520323038102,
      "num_tokens": 3624475.0,
      "step": 2060
    },
    {
      "entropy": 6.000697422027588,
      "epoch": 1.6183385579937304,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004909566596993498,
      "loss": 5.9374,
      "mean_token_accuracy": 0.1381327621638775,
      "num_tokens": 3633995.0,
      "step": 2065
    },
    {
      "entropy": 5.861016702651978,
      "epoch": 1.622257053291536,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004908720469338988,
      "loss": 5.8094,
      "mean_token_accuracy": 0.14452051296830176,
      "num_tokens": 3643121.0,
      "step": 2070
    },
    {
      "entropy": 5.835006523132324,
      "epoch": 1.6261755485893417,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004907870483700979,
      "loss": 5.7401,
      "mean_token_accuracy": 0.1410243555903435,
      "num_tokens": 3652285.0,
      "step": 2075
    },
    {
      "entropy": 5.813826656341552,
      "epoch": 1.6300940438871474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004907016641598534,
      "loss": 5.8232,
      "mean_token_accuracy": 0.14193576872348784,
      "num_tokens": 3661441.0,
      "step": 2080
    },
    {
      "entropy": 5.849268865585327,
      "epoch": 1.634012539184953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004906158944557607,
      "loss": 5.6933,
      "mean_token_accuracy": 0.15349719524383545,
      "num_tokens": 3669364.0,
      "step": 2085
    },
    {
      "entropy": 5.878481912612915,
      "epoch": 1.6379310344827587,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000490529739411104,
      "loss": 5.8009,
      "mean_token_accuracy": 0.14380127266049386,
      "num_tokens": 3677848.0,
      "step": 2090
    },
    {
      "entropy": 5.790279960632324,
      "epoch": 1.6418495297805644,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004904431991798565,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15105993747711183,
      "num_tokens": 3686617.0,
      "step": 2095
    },
    {
      "entropy": 5.82244086265564,
      "epoch": 1.64576802507837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004903562739166797,
      "loss": 5.6836,
      "mean_token_accuracy": 0.14716468155384063,
      "num_tokens": 3695106.0,
      "step": 2100
    },
    {
      "entropy": 5.731297779083252,
      "epoch": 1.6496865203761755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004902689637769229,
      "loss": 5.6316,
      "mean_token_accuracy": 0.1544244959950447,
      "num_tokens": 3703167.0,
      "step": 2105
    },
    {
      "entropy": 5.850043249130249,
      "epoch": 1.6536050156739812,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004901812689166237,
      "loss": 5.7411,
      "mean_token_accuracy": 0.14633739218115807,
      "num_tokens": 3711597.0,
      "step": 2110
    },
    {
      "entropy": 5.808914756774902,
      "epoch": 1.657523510971787,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004900931894925069,
      "loss": 5.8445,
      "mean_token_accuracy": 0.14314616173505784,
      "num_tokens": 3720277.0,
      "step": 2115
    },
    {
      "entropy": 5.803114509582519,
      "epoch": 1.6614420062695925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004900047256619849,
      "loss": 5.8091,
      "mean_token_accuracy": 0.1463458240032196,
      "num_tokens": 3729832.0,
      "step": 2120
    },
    {
      "entropy": 5.832270526885987,
      "epoch": 1.665360501567398,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004899158775831566,
      "loss": 5.7648,
      "mean_token_accuracy": 0.15041996985673906,
      "num_tokens": 3738390.0,
      "step": 2125
    },
    {
      "entropy": 5.838447761535645,
      "epoch": 1.6692789968652038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004898266454148081,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1450522780418396,
      "num_tokens": 3747046.0,
      "step": 2130
    },
    {
      "entropy": 5.7696085453033445,
      "epoch": 1.6731974921630095,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004897370293164119,
      "loss": 5.7837,
      "mean_token_accuracy": 0.13788653165102005,
      "num_tokens": 3755818.0,
      "step": 2135
    },
    {
      "entropy": 6.014819002151489,
      "epoch": 1.677115987460815,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004896470294481262,
      "loss": 5.873,
      "mean_token_accuracy": 0.1447507619857788,
      "num_tokens": 3764123.0,
      "step": 2140
    },
    {
      "entropy": 5.868058919906616,
      "epoch": 1.6810344827586206,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004895566459707954,
      "loss": 5.928,
      "mean_token_accuracy": 0.1357198879122734,
      "num_tokens": 3773105.0,
      "step": 2145
    },
    {
      "entropy": 5.8731566905975345,
      "epoch": 1.6849529780564263,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004894658790459498,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1499609887599945,
      "num_tokens": 3781417.0,
      "step": 2150
    },
    {
      "entropy": 5.944635963439941,
      "epoch": 1.688871473354232,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004893747288358041,
      "loss": 5.8672,
      "mean_token_accuracy": 0.14666299670934677,
      "num_tokens": 3789657.0,
      "step": 2155
    },
    {
      "entropy": 5.897546195983887,
      "epoch": 1.6927899686520376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000489283195503259,
      "loss": 5.8762,
      "mean_token_accuracy": 0.14236082658171653,
      "num_tokens": 3798700.0,
      "step": 2160
    },
    {
      "entropy": 5.820228528976441,
      "epoch": 1.6967084639498433,
      "grad_norm": 1.171875,
      "learning_rate": 0.000489191279211899,
      "loss": 5.7223,
      "mean_token_accuracy": 0.14710770919919014,
      "num_tokens": 3807299.0,
      "step": 2165
    },
    {
      "entropy": 5.885312223434449,
      "epoch": 1.700626959247649,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004890989801259935,
      "loss": 5.8276,
      "mean_token_accuracy": 0.1482617050409317,
      "num_tokens": 3816292.0,
      "step": 2170
    },
    {
      "entropy": 5.931986904144287,
      "epoch": 1.7045454545454546,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000489006298410496,
      "loss": 5.7392,
      "mean_token_accuracy": 0.14349312484264373,
      "num_tokens": 3825070.0,
      "step": 2175
    },
    {
      "entropy": 5.804163599014283,
      "epoch": 1.70846394984326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004889132342310438,
      "loss": 5.7293,
      "mean_token_accuracy": 0.14909339994192122,
      "num_tokens": 3834016.0,
      "step": 2180
    },
    {
      "entropy": 5.880694341659546,
      "epoch": 1.7123824451410659,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004888197877539577,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1458732567727566,
      "num_tokens": 3842172.0,
      "step": 2185
    },
    {
      "entropy": 5.79048523902893,
      "epoch": 1.7163009404388716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004887259591462417,
      "loss": 5.683,
      "mean_token_accuracy": 0.14992706328630448,
      "num_tokens": 3851109.0,
      "step": 2190
    },
    {
      "entropy": 5.782387828826904,
      "epoch": 1.7202194357366771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004886317485755825,
      "loss": 5.796,
      "mean_token_accuracy": 0.14876937195658685,
      "num_tokens": 3859656.0,
      "step": 2195
    },
    {
      "entropy": 5.809633064270019,
      "epoch": 1.7241379310344827,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004885371562103498,
      "loss": 5.6571,
      "mean_token_accuracy": 0.15718313902616501,
      "num_tokens": 3868698.0,
      "step": 2200
    },
    {
      "entropy": 5.76657247543335,
      "epoch": 1.7280564263322884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004884421822195957,
      "loss": 5.7444,
      "mean_token_accuracy": 0.14893446266651153,
      "num_tokens": 3877474.0,
      "step": 2205
    },
    {
      "entropy": 5.85271372795105,
      "epoch": 1.7319749216300941,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883468267730538,
      "loss": 5.7336,
      "mean_token_accuracy": 0.14580635279417037,
      "num_tokens": 3886328.0,
      "step": 2210
    },
    {
      "entropy": 5.817337465286255,
      "epoch": 1.7358934169278997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048825109004114006,
      "loss": 5.6926,
      "mean_token_accuracy": 0.15190036296844484,
      "num_tokens": 3894424.0,
      "step": 2215
    },
    {
      "entropy": 5.8421002388000485,
      "epoch": 1.7398119122257052,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004881549721949513,
      "loss": 5.8065,
      "mean_token_accuracy": 0.1421722039580345,
      "num_tokens": 3903746.0,
      "step": 2220
    },
    {
      "entropy": 5.735415267944336,
      "epoch": 1.743730407523511,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004880584734062658,
      "loss": 5.7647,
      "mean_token_accuracy": 0.1507815882563591,
      "num_tokens": 3912033.0,
      "step": 2225
    },
    {
      "entropy": 5.788500356674194,
      "epoch": 1.7476489028213167,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004879615938475425,
      "loss": 5.6733,
      "mean_token_accuracy": 0.14568700939416884,
      "num_tokens": 3920388.0,
      "step": 2230
    },
    {
      "entropy": 5.820400190353394,
      "epoch": 1.7515673981191222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004878643336919209,
      "loss": 5.8767,
      "mean_token_accuracy": 0.14171729907393454,
      "num_tokens": 3929926.0,
      "step": 2235
    },
    {
      "entropy": 5.996213912963867,
      "epoch": 1.7554858934169277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004877666931132206,
      "loss": 5.8413,
      "mean_token_accuracy": 0.13652574941515921,
      "num_tokens": 3938696.0,
      "step": 2240
    },
    {
      "entropy": 5.85072169303894,
      "epoch": 1.7594043887147337,
      "grad_norm": 1.125,
      "learning_rate": 0.0004876686722859413,
      "loss": 5.6819,
      "mean_token_accuracy": 0.1550109453499317,
      "num_tokens": 3947537.0,
      "step": 2245
    },
    {
      "entropy": 5.824322128295899,
      "epoch": 1.7633228840125392,
      "grad_norm": 1.15625,
      "learning_rate": 0.000487570271385262,
      "loss": 5.6797,
      "mean_token_accuracy": 0.14713258892297745,
      "num_tokens": 3956539.0,
      "step": 2250
    },
    {
      "entropy": 5.770116233825684,
      "epoch": 1.7672413793103448,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004874714905870411,
      "loss": 5.6556,
      "mean_token_accuracy": 0.14795414805412294,
      "num_tokens": 3964927.0,
      "step": 2255
    },
    {
      "entropy": 5.781585693359375,
      "epoch": 1.7711598746081505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004873723300678159,
      "loss": 5.7993,
      "mean_token_accuracy": 0.14609729051589965,
      "num_tokens": 3973565.0,
      "step": 2260
    },
    {
      "entropy": 5.8436461925506595,
      "epoch": 1.7750783699059562,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048727279000480226,
      "loss": 5.8134,
      "mean_token_accuracy": 0.1419944316148758,
      "num_tokens": 3982683.0,
      "step": 2265
    },
    {
      "entropy": 5.901409387588501,
      "epoch": 1.7789968652037618,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048717287057589454,
      "loss": 5.7444,
      "mean_token_accuracy": 0.14614344835281373,
      "num_tokens": 3991462.0,
      "step": 2270
    },
    {
      "entropy": 5.5885340690612795,
      "epoch": 1.7829153605015673,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004870725719596648,
      "loss": 5.4992,
      "mean_token_accuracy": 0.15931711345911026,
      "num_tokens": 3999821.0,
      "step": 2275
    },
    {
      "entropy": 5.817122077941894,
      "epoch": 1.786833855799373,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004869718943353631,
      "loss": 5.7526,
      "mean_token_accuracy": 0.14650990962982177,
      "num_tokens": 4008655.0,
      "step": 2280
    },
    {
      "entropy": 5.852270984649659,
      "epoch": 1.7907523510971788,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048687083788291656,
      "loss": 5.7852,
      "mean_token_accuracy": 0.15168624818325044,
      "num_tokens": 4017317.0,
      "step": 2285
    },
    {
      "entropy": 5.706546688079834,
      "epoch": 1.7946708463949843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048676940278292953,
      "loss": 5.6183,
      "mean_token_accuracy": 0.15399519652128218,
      "num_tokens": 4025407.0,
      "step": 2290
    },
    {
      "entropy": 5.778053712844849,
      "epoch": 1.7985893416927898,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048666758921668286,
      "loss": 5.6273,
      "mean_token_accuracy": 0.15056194961071015,
      "num_tokens": 4034310.0,
      "step": 2295
    },
    {
      "entropy": 5.775419092178344,
      "epoch": 1.8025078369905956,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048656539736613403,
      "loss": 5.7907,
      "mean_token_accuracy": 0.14853218123316764,
      "num_tokens": 4042326.0,
      "step": 2300
    },
    {
      "entropy": 5.706247758865357,
      "epoch": 1.8064263322884013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004864628274139164,
      "loss": 5.6481,
      "mean_token_accuracy": 0.15559574365615844,
      "num_tokens": 4051248.0,
      "step": 2305
    },
    {
      "entropy": 5.8236981391906735,
      "epoch": 1.8103448275862069,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863598795433391,
      "loss": 5.7786,
      "mean_token_accuracy": 0.14639326483011245,
      "num_tokens": 4060227.0,
      "step": 2310
    },
    {
      "entropy": 5.91852650642395,
      "epoch": 1.8142633228840124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048625655393838666,
      "loss": 5.8551,
      "mean_token_accuracy": 0.1396035686135292,
      "num_tokens": 4069700.0,
      "step": 2315
    },
    {
      "entropy": 5.817505264282227,
      "epoch": 1.8181818181818183,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004861528507837186,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14971303641796113,
      "num_tokens": 4079363.0,
      "step": 2320
    },
    {
      "entropy": 5.909413576126099,
      "epoch": 1.8221003134796239,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004860487702646695,
      "loss": 5.8089,
      "mean_token_accuracy": 0.14138850718736648,
      "num_tokens": 4087944.0,
      "step": 2325
    },
    {
      "entropy": 5.663079452514649,
      "epoch": 1.8260188087774294,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004859443125672479,
      "loss": 5.5141,
      "mean_token_accuracy": 0.16887633204460145,
      "num_tokens": 4096431.0,
      "step": 2330
    },
    {
      "entropy": 5.716325759887695,
      "epoch": 1.8299373040752351,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004858394778781368,
      "loss": 5.6716,
      "mean_token_accuracy": 0.14433413445949556,
      "num_tokens": 4104894.0,
      "step": 2335
    },
    {
      "entropy": 5.810530185699463,
      "epoch": 1.8338557993730409,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004857342663846927,
      "loss": 5.7169,
      "mean_token_accuracy": 0.14703929871320726,
      "num_tokens": 4112878.0,
      "step": 2340
    },
    {
      "entropy": 5.7495544910430905,
      "epoch": 1.8377742946708464,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004856286782749456,
      "loss": 5.7282,
      "mean_token_accuracy": 0.15359484106302262,
      "num_tokens": 4122501.0,
      "step": 2345
    },
    {
      "entropy": 5.722039604187012,
      "epoch": 1.841692789968652,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004855227137375986,
      "loss": 5.6971,
      "mean_token_accuracy": 0.14616180807352067,
      "num_tokens": 4131301.0,
      "step": 2350
    },
    {
      "entropy": 5.7759270668029785,
      "epoch": 1.8456112852664577,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004854163729620275,
      "loss": 5.7346,
      "mean_token_accuracy": 0.15014693140983582,
      "num_tokens": 4140076.0,
      "step": 2355
    },
    {
      "entropy": 5.80101170539856,
      "epoch": 1.8495297805642634,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004853096561382805,
      "loss": 5.6482,
      "mean_token_accuracy": 0.1573817789554596,
      "num_tokens": 4148728.0,
      "step": 2360
    },
    {
      "entropy": 5.722895526885987,
      "epoch": 1.853448275862069,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004852025634570779,
      "loss": 5.817,
      "mean_token_accuracy": 0.14882018342614173,
      "num_tokens": 4157520.0,
      "step": 2365
    },
    {
      "entropy": 5.815022802352905,
      "epoch": 1.8573667711598745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004850950951098116,
      "loss": 5.6208,
      "mean_token_accuracy": 0.15457617491483688,
      "num_tokens": 4165547.0,
      "step": 2370
    },
    {
      "entropy": 5.7170045375823975,
      "epoch": 1.8612852664576802,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004849872512885451,
      "loss": 5.7669,
      "mean_token_accuracy": 0.14189089238643646,
      "num_tokens": 4174656.0,
      "step": 2375
    },
    {
      "entropy": 5.836828517913818,
      "epoch": 1.865203761755486,
      "grad_norm": 1.125,
      "learning_rate": 0.0004848790321860127,
      "loss": 5.7994,
      "mean_token_accuracy": 0.1457727000117302,
      "num_tokens": 4183452.0,
      "step": 2380
    },
    {
      "entropy": 5.839574909210205,
      "epoch": 1.8691222570532915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048477043799561946,
      "loss": 5.7025,
      "mean_token_accuracy": 0.14705116599798201,
      "num_tokens": 4192306.0,
      "step": 2385
    },
    {
      "entropy": 5.765735483169555,
      "epoch": 1.873040752351097,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004846614689114409,
      "loss": 5.7357,
      "mean_token_accuracy": 0.14695977419614792,
      "num_tokens": 4201201.0,
      "step": 2390
    },
    {
      "entropy": 5.742426204681396,
      "epoch": 1.876959247648903,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004845521251282223,
      "loss": 5.6466,
      "mean_token_accuracy": 0.15739916563034057,
      "num_tokens": 4209374.0,
      "step": 2395
    },
    {
      "entropy": 5.652604055404663,
      "epoch": 1.8808777429467085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004844424068413789,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15140311568975448,
      "num_tokens": 4217258.0,
      "step": 2400
    },
    {
      "entropy": 5.726423931121826,
      "epoch": 1.884796238244514,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048433231424699504,
      "loss": 5.6578,
      "mean_token_accuracy": 0.15339543521404267,
      "num_tokens": 4226484.0,
      "step": 2405
    },
    {
      "entropy": 5.76325569152832,
      "epoch": 1.8887147335423198,
      "grad_norm": 1.125,
      "learning_rate": 0.00048422184754182384,
      "loss": 5.6794,
      "mean_token_accuracy": 0.1510704293847084,
      "num_tokens": 4235546.0,
      "step": 2410
    },
    {
      "entropy": 5.839502143859863,
      "epoch": 1.8926332288401255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004841110069232875,
      "loss": 5.8354,
      "mean_token_accuracy": 0.14865082055330275,
      "num_tokens": 4244549.0,
      "step": 2415
    },
    {
      "entropy": 5.769964933395386,
      "epoch": 1.896551724137931,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048399979258947597,
      "loss": 5.7221,
      "mean_token_accuracy": 0.15206404626369477,
      "num_tokens": 4252918.0,
      "step": 2420
    },
    {
      "entropy": 5.813162088394165,
      "epoch": 1.9004702194357366,
      "grad_norm": 1.125,
      "learning_rate": 0.0004838882047391474,
      "loss": 5.7137,
      "mean_token_accuracy": 0.14775785952806472,
      "num_tokens": 4261674.0,
      "step": 2425
    },
    {
      "entropy": 5.754944515228272,
      "epoch": 1.9043887147335423,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048377624357172724,
      "loss": 5.6275,
      "mean_token_accuracy": 0.15521396249532698,
      "num_tokens": 4269986.0,
      "step": 2430
    },
    {
      "entropy": 5.772288084030151,
      "epoch": 1.908307210031348,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048366390928730843,
      "loss": 5.7277,
      "mean_token_accuracy": 0.15137290358543395,
      "num_tokens": 4278131.0,
      "step": 2435
    },
    {
      "entropy": 5.772933292388916,
      "epoch": 1.9122257053291536,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004835512020866504,
      "loss": 5.6901,
      "mean_token_accuracy": 0.14994694739580156,
      "num_tokens": 4287214.0,
      "step": 2440
    },
    {
      "entropy": 5.747813320159912,
      "epoch": 1.9161442006269591,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048343812217117925,
      "loss": 5.6528,
      "mean_token_accuracy": 0.15408559143543243,
      "num_tokens": 4295949.0,
      "step": 2445
    },
    {
      "entropy": 5.741509056091308,
      "epoch": 1.9200626959247649,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048332466974298723,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15061129480600358,
      "num_tokens": 4305014.0,
      "step": 2450
    },
    {
      "entropy": 5.716789960861206,
      "epoch": 1.9239811912225706,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048321084500483203,
      "loss": 5.6184,
      "mean_token_accuracy": 0.15444232225418092,
      "num_tokens": 4313560.0,
      "step": 2455
    },
    {
      "entropy": 5.753894281387329,
      "epoch": 1.9278996865203761,
      "grad_norm": 1.171875,
      "learning_rate": 0.000483096648160137,
      "loss": 5.7105,
      "mean_token_accuracy": 0.15123845040798187,
      "num_tokens": 4323280.0,
      "step": 2460
    },
    {
      "entropy": 5.806096601486206,
      "epoch": 1.9318181818181817,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048298207941299047,
      "loss": 5.747,
      "mean_token_accuracy": 0.1527293801307678,
      "num_tokens": 4331757.0,
      "step": 2465
    },
    {
      "entropy": 5.656486320495605,
      "epoch": 1.9357366771159876,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048286713896814536,
      "loss": 5.6845,
      "mean_token_accuracy": 0.15062677338719369,
      "num_tokens": 4341453.0,
      "step": 2470
    },
    {
      "entropy": 5.872777032852173,
      "epoch": 1.9396551724137931,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048275182703101877,
      "loss": 5.6272,
      "mean_token_accuracy": 0.15510055273771287,
      "num_tokens": 4349095.0,
      "step": 2475
    },
    {
      "entropy": 5.647492599487305,
      "epoch": 1.9435736677115987,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048263614380769193,
      "loss": 5.734,
      "mean_token_accuracy": 0.14693398848176004,
      "num_tokens": 4357913.0,
      "step": 2480
    },
    {
      "entropy": 5.7816098690032955,
      "epoch": 1.9474921630094044,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048252008950490957,
      "loss": 5.7042,
      "mean_token_accuracy": 0.1510899305343628,
      "num_tokens": 4366586.0,
      "step": 2485
    },
    {
      "entropy": 5.780308771133423,
      "epoch": 1.9514106583072102,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048240366433007935,
      "loss": 5.6439,
      "mean_token_accuracy": 0.1548413008451462,
      "num_tokens": 4375393.0,
      "step": 2490
    },
    {
      "entropy": 5.719468688964843,
      "epoch": 1.9553291536050157,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048228686849127213,
      "loss": 5.707,
      "mean_token_accuracy": 0.15866673439741136,
      "num_tokens": 4383747.0,
      "step": 2495
    },
    {
      "entropy": 5.827725028991699,
      "epoch": 1.9592476489028212,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004821697021972209,
      "loss": 5.69,
      "mean_token_accuracy": 0.14475997686386108,
      "num_tokens": 4392085.0,
      "step": 2500
    },
    {
      "epoch": 1.9592476489028212,
      "eval_entropy": 5.705850722253785,
      "eval_loss": 5.8995537757873535,
      "eval_mean_token_accuracy": 0.14770855615276468,
      "eval_num_tokens": 4392085.0,
      "eval_runtime": 3.0359,
      "eval_samples_per_second": 1357.773,
      "eval_steps_per_second": 169.969,
      "step": 2500
    },
    {
      "entropy": 5.789532089233399,
      "epoch": 1.963166144200627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004820521656573208,
      "loss": 5.7037,
      "mean_token_accuracy": 0.1507359117269516,
      "num_tokens": 4400843.0,
      "step": 2505
    },
    {
      "entropy": 5.641516876220703,
      "epoch": 1.9670846394984327,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004819342590816288,
      "loss": 5.664,
      "mean_token_accuracy": 0.15229431241750718,
      "num_tokens": 4409126.0,
      "step": 2510
    },
    {
      "entropy": 5.780962181091309,
      "epoch": 1.9710031347962382,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004818159826808631,
      "loss": 5.7947,
      "mean_token_accuracy": 0.14660751819610596,
      "num_tokens": 4418401.0,
      "step": 2515
    },
    {
      "entropy": 5.7990799903869625,
      "epoch": 1.9749216300940438,
      "grad_norm": 1.125,
      "learning_rate": 0.0004816973366664026,
      "loss": 5.7212,
      "mean_token_accuracy": 0.1470862217247486,
      "num_tokens": 4428141.0,
      "step": 2520
    },
    {
      "entropy": 5.763505554199218,
      "epoch": 1.9788401253918495,
      "grad_norm": 1.125,
      "learning_rate": 0.0004815783212502871,
      "loss": 5.7216,
      "mean_token_accuracy": 0.1449996516108513,
      "num_tokens": 4437158.0,
      "step": 2525
    },
    {
      "entropy": 5.798718881607056,
      "epoch": 1.9827586206896552,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048145893664521645,
      "loss": 5.7848,
      "mean_token_accuracy": 0.1408971928060055,
      "num_tokens": 4445901.0,
      "step": 2530
    },
    {
      "entropy": 5.741735363006592,
      "epoch": 1.9866771159874608,
      "grad_norm": 1.25,
      "learning_rate": 0.00048133918306455023,
      "loss": 5.6081,
      "mean_token_accuracy": 0.16412827521562576,
      "num_tokens": 4454200.0,
      "step": 2535
    },
    {
      "entropy": 5.753818655014038,
      "epoch": 1.9905956112852663,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004812190607223075,
      "loss": 5.7198,
      "mean_token_accuracy": 0.1448679745197296,
      "num_tokens": 4463484.0,
      "step": 2540
    },
    {
      "entropy": 5.637834119796753,
      "epoch": 1.9945141065830723,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048109856983316655,
      "loss": 5.5579,
      "mean_token_accuracy": 0.15664460957050325,
      "num_tokens": 4472659.0,
      "step": 2545
    },
    {
      "entropy": 5.7259618759155275,
      "epoch": 1.9984326018808778,
      "grad_norm": 1.140625,
      "learning_rate": 0.000480977710612464,
      "loss": 5.6156,
      "mean_token_accuracy": 0.14887530580163003,
      "num_tokens": 4481355.0,
      "step": 2550
    },
    {
      "entropy": 5.752556180953979,
      "epoch": 2.0023510971786833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004808564832761948,
      "loss": 5.4781,
      "mean_token_accuracy": 0.15195177793502807,
      "num_tokens": 4490005.0,
      "step": 2555
    },
    {
      "entropy": 5.674878787994385,
      "epoch": 2.006269592476489,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004807348880410119,
      "loss": 5.2969,
      "mean_token_accuracy": 0.163840214908123,
      "num_tokens": 4498645.0,
      "step": 2560
    },
    {
      "entropy": 5.641942977905273,
      "epoch": 2.010188087774295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004806129251242258,
      "loss": 5.3845,
      "mean_token_accuracy": 0.15954576134681703,
      "num_tokens": 4509675.0,
      "step": 2565
    },
    {
      "entropy": 5.6735498905181885,
      "epoch": 2.0141065830721003,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048049059474380393,
      "loss": 5.2179,
      "mean_token_accuracy": 0.166520294547081,
      "num_tokens": 4518463.0,
      "step": 2570
    },
    {
      "entropy": 5.706894254684448,
      "epoch": 2.018025078369906,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048036789711837047,
      "loss": 5.3649,
      "mean_token_accuracy": 0.15959171056747437,
      "num_tokens": 4527120.0,
      "step": 2575
    },
    {
      "entropy": 5.680517148971558,
      "epoch": 2.0219435736677114,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048024483246720607,
      "loss": 5.397,
      "mean_token_accuracy": 0.15788237601518632,
      "num_tokens": 4535962.0,
      "step": 2580
    },
    {
      "entropy": 5.706279850006103,
      "epoch": 2.0258620689655173,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004801214010102472,
      "loss": 5.3536,
      "mean_token_accuracy": 0.15642207115888596,
      "num_tokens": 4545233.0,
      "step": 2585
    },
    {
      "entropy": 5.604188776016235,
      "epoch": 2.029780564263323,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004799976029680858,
      "loss": 5.227,
      "mean_token_accuracy": 0.16073910295963287,
      "num_tokens": 4554658.0,
      "step": 2590
    },
    {
      "entropy": 5.717430782318115,
      "epoch": 2.0336990595611284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004798734385619691,
      "loss": 5.4145,
      "mean_token_accuracy": 0.16037535965442656,
      "num_tokens": 4563964.0,
      "step": 2595
    },
    {
      "entropy": 5.666239976882935,
      "epoch": 2.0376175548589344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479748908013799,
      "loss": 5.2896,
      "mean_token_accuracy": 0.1706826001405716,
      "num_tokens": 4572865.0,
      "step": 2600
    },
    {
      "entropy": 5.570950031280518,
      "epoch": 2.04153605015674,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004796240115461319,
      "loss": 5.2642,
      "mean_token_accuracy": 0.17284168303012848,
      "num_tokens": 4581021.0,
      "step": 2605
    },
    {
      "entropy": 5.529971885681152,
      "epoch": 2.0454545454545454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004794987493821779,
      "loss": 5.3565,
      "mean_token_accuracy": 0.16215605586767196,
      "num_tokens": 4590779.0,
      "step": 2610
    },
    {
      "entropy": 5.651610660552978,
      "epoch": 2.049373040752351,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047937312174580084,
      "loss": 5.3414,
      "mean_token_accuracy": 0.16156901866197587,
      "num_tokens": 4599831.0,
      "step": 2615
    },
    {
      "entropy": 5.721800804138184,
      "epoch": 2.053291536050157,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004792471288615177,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1500769391655922,
      "num_tokens": 4609264.0,
      "step": 2620
    },
    {
      "entropy": 5.70226821899414,
      "epoch": 2.0572100313479624,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004791207709544981,
      "loss": 5.374,
      "mean_token_accuracy": 0.1676947593688965,
      "num_tokens": 4617885.0,
      "step": 2625
    },
    {
      "entropy": 5.5007805824279785,
      "epoch": 2.061128526645768,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047899404825056424,
      "loss": 5.3022,
      "mean_token_accuracy": 0.16264875382184982,
      "num_tokens": 4626536.0,
      "step": 2630
    },
    {
      "entropy": 5.664446115493774,
      "epoch": 2.0650470219435735,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004788669609761901,
      "loss": 5.4064,
      "mean_token_accuracy": 0.1559653714299202,
      "num_tokens": 4635563.0,
      "step": 2635
    },
    {
      "entropy": 5.576727724075317,
      "epoch": 2.0689655172413794,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047873950935850107,
      "loss": 5.3001,
      "mean_token_accuracy": 0.15983396619558335,
      "num_tokens": 4644822.0,
      "step": 2640
    },
    {
      "entropy": 5.616795778274536,
      "epoch": 2.072884012539185,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004786116936252742,
      "loss": 5.2834,
      "mean_token_accuracy": 0.16793755143880845,
      "num_tokens": 4652709.0,
      "step": 2645
    },
    {
      "entropy": 5.555015659332275,
      "epoch": 2.0768025078369905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004784835140049367,
      "loss": 5.3014,
      "mean_token_accuracy": 0.16771620512008667,
      "num_tokens": 4662014.0,
      "step": 2650
    },
    {
      "entropy": 5.66242470741272,
      "epoch": 2.080721003134796,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004783549707265663,
      "loss": 5.2894,
      "mean_token_accuracy": 0.1594408079981804,
      "num_tokens": 4670618.0,
      "step": 2655
    },
    {
      "entropy": 5.5888073444366455,
      "epoch": 2.084639498432602,
      "grad_norm": 1.125,
      "learning_rate": 0.00047822606401989084,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1533523418009281,
      "num_tokens": 4679937.0,
      "step": 2660
    },
    {
      "entropy": 5.6302934169769285,
      "epoch": 2.0885579937304075,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004780967941152873,
      "loss": 5.238,
      "mean_token_accuracy": 0.17676321119070054,
      "num_tokens": 4688006.0,
      "step": 2665
    },
    {
      "entropy": 5.542519521713257,
      "epoch": 2.092476489028213,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004779671612437822,
      "loss": 5.1899,
      "mean_token_accuracy": 0.173385126888752,
      "num_tokens": 4697106.0,
      "step": 2670
    },
    {
      "entropy": 5.543998193740845,
      "epoch": 2.096394984326019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047783716563705035,
      "loss": 5.3423,
      "mean_token_accuracy": 0.16200381517410278,
      "num_tokens": 4705582.0,
      "step": 2675
    },
    {
      "entropy": 5.640423583984375,
      "epoch": 2.1003134796238245,
      "grad_norm": 1.234375,
      "learning_rate": 0.000477706807527415,
      "loss": 5.2754,
      "mean_token_accuracy": 0.1639168754220009,
      "num_tokens": 4713640.0,
      "step": 2680
    },
    {
      "entropy": 5.603535509109497,
      "epoch": 2.10423197492163,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004775760871478472,
      "loss": 5.4069,
      "mean_token_accuracy": 0.160996450483799,
      "num_tokens": 4722729.0,
      "step": 2685
    },
    {
      "entropy": 5.627258968353272,
      "epoch": 2.1081504702194356,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047744500473196564,
      "loss": 5.2855,
      "mean_token_accuracy": 0.1675833374261856,
      "num_tokens": 4732099.0,
      "step": 2690
    },
    {
      "entropy": 5.648689317703247,
      "epoch": 2.1120689655172415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047731356051403556,
      "loss": 5.3168,
      "mean_token_accuracy": 0.15454934388399125,
      "num_tokens": 4741766.0,
      "step": 2695
    },
    {
      "entropy": 5.469576168060303,
      "epoch": 2.115987460815047,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004771817547289693,
      "loss": 5.2196,
      "mean_token_accuracy": 0.1669618755578995,
      "num_tokens": 4750890.0,
      "step": 2700
    },
    {
      "entropy": 5.580454540252686,
      "epoch": 2.1199059561128526,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004770495876123251,
      "loss": 5.3545,
      "mean_token_accuracy": 0.16359022557735442,
      "num_tokens": 4760063.0,
      "step": 2705
    },
    {
      "entropy": 5.643666553497314,
      "epoch": 2.123824451410658,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004769170594003071,
      "loss": 5.3426,
      "mean_token_accuracy": 0.1646828219294548,
      "num_tokens": 4768928.0,
      "step": 2710
    },
    {
      "entropy": 5.535561656951904,
      "epoch": 2.127742946708464,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047678417032976457,
      "loss": 5.3436,
      "mean_token_accuracy": 0.1597463920712471,
      "num_tokens": 4778129.0,
      "step": 2715
    },
    {
      "entropy": 5.582833862304687,
      "epoch": 2.1316614420062696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004766509206381919,
      "loss": 5.3256,
      "mean_token_accuracy": 0.16664166301488875,
      "num_tokens": 4786759.0,
      "step": 2720
    },
    {
      "entropy": 5.543987512588501,
      "epoch": 2.135579937304075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004765173105637279,
      "loss": 5.3209,
      "mean_token_accuracy": 0.16914047598838805,
      "num_tokens": 4795355.0,
      "step": 2725
    },
    {
      "entropy": 5.563296318054199,
      "epoch": 2.139498432601881,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047638334034515547,
      "loss": 5.2819,
      "mean_token_accuracy": 0.16699936389923095,
      "num_tokens": 4804030.0,
      "step": 2730
    },
    {
      "entropy": 5.451429939270019,
      "epoch": 2.1434169278996866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047624901022190106,
      "loss": 5.2285,
      "mean_token_accuracy": 0.1854039654135704,
      "num_tokens": 4812779.0,
      "step": 2735
    },
    {
      "entropy": 5.612224864959717,
      "epoch": 2.147335423197492,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047611432043403437,
      "loss": 5.349,
      "mean_token_accuracy": 0.16258929520845414,
      "num_tokens": 4822292.0,
      "step": 2740
    },
    {
      "entropy": 5.713485240936279,
      "epoch": 2.1512539184952977,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004759792712222679,
      "loss": 5.408,
      "mean_token_accuracy": 0.16127049475908278,
      "num_tokens": 4830524.0,
      "step": 2745
    },
    {
      "entropy": 5.600223112106323,
      "epoch": 2.1551724137931036,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004758438628279565,
      "loss": 5.4051,
      "mean_token_accuracy": 0.16541811376810073,
      "num_tokens": 4839328.0,
      "step": 2750
    },
    {
      "entropy": 5.586366415023804,
      "epoch": 2.159090909090909,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047570809549309697,
      "loss": 5.2408,
      "mean_token_accuracy": 0.17708596885204314,
      "num_tokens": 4847711.0,
      "step": 2755
    },
    {
      "entropy": 5.651467561721802,
      "epoch": 2.1630094043887147,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004755719694603275,
      "loss": 5.4092,
      "mean_token_accuracy": 0.16118254661560058,
      "num_tokens": 4856706.0,
      "step": 2760
    },
    {
      "entropy": 5.636726808547974,
      "epoch": 2.16692789968652,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004754354849729274,
      "loss": 5.4163,
      "mean_token_accuracy": 0.16455628871917724,
      "num_tokens": 4865250.0,
      "step": 2765
    },
    {
      "entropy": 5.60169677734375,
      "epoch": 2.170846394984326,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047529864227481653,
      "loss": 5.3433,
      "mean_token_accuracy": 0.16435989141464233,
      "num_tokens": 4873826.0,
      "step": 2770
    },
    {
      "entropy": 5.669528484344482,
      "epoch": 2.1747648902821317,
      "grad_norm": 1.171875,
      "learning_rate": 0.000475161441610555,
      "loss": 5.3172,
      "mean_token_accuracy": 0.16513787358999252,
      "num_tokens": 4882631.0,
      "step": 2775
    },
    {
      "entropy": 5.560564279556274,
      "epoch": 2.1786833855799372,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004750238832253427,
      "loss": 5.3731,
      "mean_token_accuracy": 0.1643209844827652,
      "num_tokens": 4891007.0,
      "step": 2780
    },
    {
      "entropy": 5.567132472991943,
      "epoch": 2.1826018808777428,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004748859673650187,
      "loss": 5.2642,
      "mean_token_accuracy": 0.1682204231619835,
      "num_tokens": 4899692.0,
      "step": 2785
    },
    {
      "entropy": 5.564448595046997,
      "epoch": 2.1865203761755487,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047474769427606115,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1644543319940567,
      "num_tokens": 4909354.0,
      "step": 2790
    },
    {
      "entropy": 5.57516770362854,
      "epoch": 2.1904388714733543,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004746090642055863,
      "loss": 5.3296,
      "mean_token_accuracy": 0.16534655094146727,
      "num_tokens": 4918400.0,
      "step": 2795
    },
    {
      "entropy": 5.628228569030762,
      "epoch": 2.19435736677116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047447007740134857,
      "loss": 5.4266,
      "mean_token_accuracy": 0.15290777534246444,
      "num_tokens": 4926852.0,
      "step": 2800
    },
    {
      "entropy": 5.629791641235352,
      "epoch": 2.1982758620689653,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047433073411174,
      "loss": 5.3439,
      "mean_token_accuracy": 0.16628239601850509,
      "num_tokens": 4935717.0,
      "step": 2805
    },
    {
      "entropy": 5.503053808212281,
      "epoch": 2.2021943573667713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004741910345857896,
      "loss": 5.2523,
      "mean_token_accuracy": 0.16592369377613067,
      "num_tokens": 4945389.0,
      "step": 2810
    },
    {
      "entropy": 5.681710863113404,
      "epoch": 2.206112852664577,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047405097907316315,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1633175492286682,
      "num_tokens": 4953670.0,
      "step": 2815
    },
    {
      "entropy": 5.543265533447266,
      "epoch": 2.2100313479623823,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004739105678241625,
      "loss": 5.4136,
      "mean_token_accuracy": 0.16170129925012589,
      "num_tokens": 4962383.0,
      "step": 2820
    },
    {
      "entropy": 5.524320983886719,
      "epoch": 2.2139498432601883,
      "grad_norm": 1.25,
      "learning_rate": 0.0004737698010897253,
      "loss": 5.2117,
      "mean_token_accuracy": 0.16661869436502458,
      "num_tokens": 4970672.0,
      "step": 2825
    },
    {
      "entropy": 5.6738903522491455,
      "epoch": 2.217868338557994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004736286791214245,
      "loss": 5.3747,
      "mean_token_accuracy": 0.15929104834795,
      "num_tokens": 4979303.0,
      "step": 2830
    },
    {
      "entropy": 5.573844003677368,
      "epoch": 2.2217868338557993,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047348720217146807,
      "loss": 5.2594,
      "mean_token_accuracy": 0.17054332792758942,
      "num_tokens": 4988322.0,
      "step": 2835
    },
    {
      "entropy": 5.466978693008423,
      "epoch": 2.225705329153605,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047334537049269806,
      "loss": 5.3355,
      "mean_token_accuracy": 0.16777932047843933,
      "num_tokens": 4997250.0,
      "step": 2840
    },
    {
      "entropy": 5.600286865234375,
      "epoch": 2.229623824451411,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004732031843385909,
      "loss": 5.4267,
      "mean_token_accuracy": 0.1585068553686142,
      "num_tokens": 5005459.0,
      "step": 2845
    },
    {
      "entropy": 5.495544385910034,
      "epoch": 2.2335423197492164,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004730606439632562,
      "loss": 5.1551,
      "mean_token_accuracy": 0.1816363513469696,
      "num_tokens": 5013704.0,
      "step": 2850
    },
    {
      "entropy": 5.431286334991455,
      "epoch": 2.237460815047022,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004729177496214367,
      "loss": 5.288,
      "mean_token_accuracy": 0.1695888951420784,
      "num_tokens": 5022545.0,
      "step": 2855
    },
    {
      "entropy": 5.527831506729126,
      "epoch": 2.2413793103448274,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047277450156850767,
      "loss": 5.3118,
      "mean_token_accuracy": 0.17252337634563447,
      "num_tokens": 5031133.0,
      "step": 2860
    },
    {
      "entropy": 5.604853677749634,
      "epoch": 2.2452978056426334,
      "grad_norm": 1.25,
      "learning_rate": 0.0004726309000604768,
      "loss": 5.3161,
      "mean_token_accuracy": 0.1649239718914032,
      "num_tokens": 5039424.0,
      "step": 2865
    },
    {
      "entropy": 5.648041248321533,
      "epoch": 2.249216300940439,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004724869453539832,
      "loss": 5.5172,
      "mean_token_accuracy": 0.1547361545264721,
      "num_tokens": 5048868.0,
      "step": 2870
    },
    {
      "entropy": 5.618503427505493,
      "epoch": 2.2531347962382444,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004723426377062972,
      "loss": 5.351,
      "mean_token_accuracy": 0.16053935885429382,
      "num_tokens": 5057532.0,
      "step": 2875
    },
    {
      "entropy": 5.597631216049194,
      "epoch": 2.2570532915360504,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047219797737532,
      "loss": 5.3526,
      "mean_token_accuracy": 0.16098810732364655,
      "num_tokens": 5066088.0,
      "step": 2880
    },
    {
      "entropy": 5.480603504180908,
      "epoch": 2.260971786833856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047205296461958314,
      "loss": 5.2604,
      "mean_token_accuracy": 0.16616066098213195,
      "num_tokens": 5074968.0,
      "step": 2885
    },
    {
      "entropy": 5.47146577835083,
      "epoch": 2.2648902821316614,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047190759969824785,
      "loss": 5.2782,
      "mean_token_accuracy": 0.16563865467905997,
      "num_tokens": 5083605.0,
      "step": 2890
    },
    {
      "entropy": 5.539209842681885,
      "epoch": 2.268808777429467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047176188287110485,
      "loss": 5.2884,
      "mean_token_accuracy": 0.16735636740922927,
      "num_tokens": 5092182.0,
      "step": 2895
    },
    {
      "entropy": 5.703541231155396,
      "epoch": 2.2727272727272725,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004716158143985737,
      "loss": 5.4828,
      "mean_token_accuracy": 0.1571749359369278,
      "num_tokens": 5100910.0,
      "step": 2900
    },
    {
      "entropy": 5.598054981231689,
      "epoch": 2.2766457680250785,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047146939454170245,
      "loss": 5.3554,
      "mean_token_accuracy": 0.16782473027706146,
      "num_tokens": 5109760.0,
      "step": 2905
    },
    {
      "entropy": 5.595303249359131,
      "epoch": 2.280564263322884,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004713226235621672,
      "loss": 5.3653,
      "mean_token_accuracy": 0.16245563328266144,
      "num_tokens": 5119193.0,
      "step": 2910
    },
    {
      "entropy": 5.53184404373169,
      "epoch": 2.2844827586206895,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004711755017222714,
      "loss": 5.2761,
      "mean_token_accuracy": 0.16629687696695328,
      "num_tokens": 5127480.0,
      "step": 2915
    },
    {
      "entropy": 5.648559045791626,
      "epoch": 2.2884012539184955,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047102802928494563,
      "loss": 5.4764,
      "mean_token_accuracy": 0.15637332051992417,
      "num_tokens": 5136648.0,
      "step": 2920
    },
    {
      "entropy": 5.578768920898438,
      "epoch": 2.292319749216301,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004708802065137471,
      "loss": 5.2737,
      "mean_token_accuracy": 0.16840785294771193,
      "num_tokens": 5144708.0,
      "step": 2925
    },
    {
      "entropy": 5.509908008575439,
      "epoch": 2.2962382445141065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004707320336728591,
      "loss": 5.3874,
      "mean_token_accuracy": 0.15815174877643584,
      "num_tokens": 5153388.0,
      "step": 2930
    },
    {
      "entropy": 5.623534393310547,
      "epoch": 2.300156739811912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004705835110270904,
      "loss": 5.3533,
      "mean_token_accuracy": 0.16146910190582275,
      "num_tokens": 5161874.0,
      "step": 2935
    },
    {
      "entropy": 5.560824823379517,
      "epoch": 2.304075235109718,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047043463884187517,
      "loss": 5.3557,
      "mean_token_accuracy": 0.1618855342268944,
      "num_tokens": 5170284.0,
      "step": 2940
    },
    {
      "entropy": 5.497574949264527,
      "epoch": 2.3079937304075235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047028541738327207,
      "loss": 5.326,
      "mean_token_accuracy": 0.1598228007555008,
      "num_tokens": 5179547.0,
      "step": 2945
    },
    {
      "entropy": 5.571848964691162,
      "epoch": 2.311912225705329,
      "grad_norm": 1.25,
      "learning_rate": 0.0004701358469179641,
      "loss": 5.3287,
      "mean_token_accuracy": 0.1698288217186928,
      "num_tokens": 5188256.0,
      "step": 2950
    },
    {
      "entropy": 5.594274997711182,
      "epoch": 2.3158307210031346,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004699859277132578,
      "loss": 5.4826,
      "mean_token_accuracy": 0.16385053992271423,
      "num_tokens": 5197464.0,
      "step": 2955
    },
    {
      "entropy": 5.6157807350158695,
      "epoch": 2.3197492163009406,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046983566003708336,
      "loss": 5.2634,
      "mean_token_accuracy": 0.17211963832378388,
      "num_tokens": 5205989.0,
      "step": 2960
    },
    {
      "entropy": 5.490232324600219,
      "epoch": 2.323667711598746,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046968504415799325,
      "loss": 5.3259,
      "mean_token_accuracy": 0.16186022460460664,
      "num_tokens": 5214490.0,
      "step": 2965
    },
    {
      "entropy": 5.529596757888794,
      "epoch": 2.3275862068965516,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004695340803451625,
      "loss": 5.3584,
      "mean_token_accuracy": 0.16376458406448363,
      "num_tokens": 5223335.0,
      "step": 2970
    },
    {
      "entropy": 5.487400341033935,
      "epoch": 2.3315047021943576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004693827688683879,
      "loss": 5.2898,
      "mean_token_accuracy": 0.16245129853487014,
      "num_tokens": 5231665.0,
      "step": 2975
    },
    {
      "entropy": 5.5195292949676515,
      "epoch": 2.335423197492163,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004692311099980878,
      "loss": 5.2879,
      "mean_token_accuracy": 0.16708355247974396,
      "num_tokens": 5239810.0,
      "step": 2980
    },
    {
      "entropy": 5.51229076385498,
      "epoch": 2.3393416927899686,
      "grad_norm": 1.25,
      "learning_rate": 0.00046907910400530097,
      "loss": 5.2546,
      "mean_token_accuracy": 0.17722575962543488,
      "num_tokens": 5247525.0,
      "step": 2985
    },
    {
      "entropy": 5.4491297721862795,
      "epoch": 2.343260188087774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004689267511616868,
      "loss": 5.3011,
      "mean_token_accuracy": 0.16801622956991197,
      "num_tokens": 5255528.0,
      "step": 2990
    },
    {
      "entropy": 5.519492959976196,
      "epoch": 2.34717868338558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046877405173952465,
      "loss": 5.2627,
      "mean_token_accuracy": 0.16878023892641067,
      "num_tokens": 5263944.0,
      "step": 2995
    },
    {
      "entropy": 5.568019199371338,
      "epoch": 2.3510971786833856,
      "grad_norm": 1.296875,
      "learning_rate": 0.000468621006011713,
      "loss": 5.3662,
      "mean_token_accuracy": 0.16010621339082717,
      "num_tokens": 5271724.0,
      "step": 3000
    },
    {
      "epoch": 2.3510971786833856,
      "eval_entropy": 5.434771962868151,
      "eval_loss": 5.8656816482543945,
      "eval_mean_token_accuracy": 0.15230537186441725,
      "eval_num_tokens": 5271724.0,
      "eval_runtime": 2.8543,
      "eval_samples_per_second": 1444.131,
      "eval_steps_per_second": 180.779,
      "step": 3000
    },
    {
      "entropy": 5.572009038925171,
      "epoch": 2.355015673981191,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046846761425176943,
      "loss": 5.3921,
      "mean_token_accuracy": 0.16172949075698853,
      "num_tokens": 5281199.0,
      "step": 3005
    },
    {
      "entropy": 5.65570912361145,
      "epoch": 2.3589341692789967,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004683138767338299,
      "loss": 5.4075,
      "mean_token_accuracy": 0.16199006289243698,
      "num_tokens": 5289782.0,
      "step": 3010
    },
    {
      "entropy": 5.402611207962036,
      "epoch": 2.3628526645768027,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004681597937326483,
      "loss": 5.2169,
      "mean_token_accuracy": 0.17421238571405412,
      "num_tokens": 5297922.0,
      "step": 3015
    },
    {
      "entropy": 5.52378478050232,
      "epoch": 2.366771159874608,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004680053655235959,
      "loss": 5.3576,
      "mean_token_accuracy": 0.16891258656978608,
      "num_tokens": 5306178.0,
      "step": 3020
    },
    {
      "entropy": 5.5647275924682615,
      "epoch": 2.3706896551724137,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004678505923826609,
      "loss": 5.3381,
      "mean_token_accuracy": 0.16613359600305558,
      "num_tokens": 5314936.0,
      "step": 3025
    },
    {
      "entropy": 5.547653722763061,
      "epoch": 2.3746081504702197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046769547458644817,
      "loss": 5.3539,
      "mean_token_accuracy": 0.16610560715198516,
      "num_tokens": 5323549.0,
      "step": 3030
    },
    {
      "entropy": 5.552595472335815,
      "epoch": 2.378526645768025,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004675400124121782,
      "loss": 5.3373,
      "mean_token_accuracy": 0.16588755548000336,
      "num_tokens": 5332310.0,
      "step": 3035
    },
    {
      "entropy": 5.6628941059112545,
      "epoch": 2.3824451410658307,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046738420613768716,
      "loss": 5.4674,
      "mean_token_accuracy": 0.1585412800312042,
      "num_tokens": 5340757.0,
      "step": 3040
    },
    {
      "entropy": 5.654271364212036,
      "epoch": 2.3863636363636362,
      "grad_norm": 1.25,
      "learning_rate": 0.00046722805604142614,
      "loss": 5.4677,
      "mean_token_accuracy": 0.1585947036743164,
      "num_tokens": 5349433.0,
      "step": 3045
    },
    {
      "entropy": 5.529625320434571,
      "epoch": 2.3902821316614418,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046707156240246076,
      "loss": 5.4046,
      "mean_token_accuracy": 0.15594652146100998,
      "num_tokens": 5358943.0,
      "step": 3050
    },
    {
      "entropy": 5.524866342544556,
      "epoch": 2.3942006269592477,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046691472550047027,
      "loss": 5.3264,
      "mean_token_accuracy": 0.16086422502994538,
      "num_tokens": 5367782.0,
      "step": 3055
    },
    {
      "entropy": 5.5129622459411625,
      "epoch": 2.3981191222570533,
      "grad_norm": 1.25,
      "learning_rate": 0.00046675754561574783,
      "loss": 5.1884,
      "mean_token_accuracy": 0.16960814148187636,
      "num_tokens": 5376186.0,
      "step": 3060
    },
    {
      "entropy": 5.5476171493530275,
      "epoch": 2.402037617554859,
      "grad_norm": 1.25,
      "learning_rate": 0.00046660002302919933,
      "loss": 5.4609,
      "mean_token_accuracy": 0.15653598308563232,
      "num_tokens": 5384888.0,
      "step": 3065
    },
    {
      "entropy": 5.588283061981201,
      "epoch": 2.4059561128526648,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004664421580223433,
      "loss": 5.3452,
      "mean_token_accuracy": 0.16461212933063507,
      "num_tokens": 5393394.0,
      "step": 3070
    },
    {
      "entropy": 5.627720546722412,
      "epoch": 2.4098746081504703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046628395087730995,
      "loss": 5.405,
      "mean_token_accuracy": 0.161533859372139,
      "num_tokens": 5402901.0,
      "step": 3075
    },
    {
      "entropy": 5.5674063205719,
      "epoch": 2.413793103448276,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004661254018768411,
      "loss": 5.3989,
      "mean_token_accuracy": 0.1623615577816963,
      "num_tokens": 5412704.0,
      "step": 3080
    },
    {
      "entropy": 5.542704439163208,
      "epoch": 2.4177115987460813,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004659665113042897,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17225506007671357,
      "num_tokens": 5420863.0,
      "step": 3085
    },
    {
      "entropy": 5.480785846710205,
      "epoch": 2.4216300940438873,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004658072794436187,
      "loss": 5.3052,
      "mean_token_accuracy": 0.17093472927808762,
      "num_tokens": 5430002.0,
      "step": 3090
    },
    {
      "entropy": 5.482878065109253,
      "epoch": 2.425548589341693,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046564770657940146,
      "loss": 5.2993,
      "mean_token_accuracy": 0.17024858742952348,
      "num_tokens": 5438424.0,
      "step": 3095
    },
    {
      "entropy": 5.4856819152832035,
      "epoch": 2.4294670846394983,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004654877929968205,
      "loss": 5.3217,
      "mean_token_accuracy": 0.1706370621919632,
      "num_tokens": 5446970.0,
      "step": 3100
    },
    {
      "entropy": 5.5803690433502195,
      "epoch": 2.433385579937304,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004653275389816673,
      "loss": 5.3947,
      "mean_token_accuracy": 0.15609584301710128,
      "num_tokens": 5456689.0,
      "step": 3105
    },
    {
      "entropy": 5.523205041885376,
      "epoch": 2.43730407523511,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046516694482034174,
      "loss": 5.2773,
      "mean_token_accuracy": 0.1692593887448311,
      "num_tokens": 5465269.0,
      "step": 3110
    },
    {
      "entropy": 5.53827395439148,
      "epoch": 2.4412225705329154,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046500601079985164,
      "loss": 5.3572,
      "mean_token_accuracy": 0.16684028506278992,
      "num_tokens": 5473930.0,
      "step": 3115
    },
    {
      "entropy": 5.512274646759034,
      "epoch": 2.445141065830721,
      "grad_norm": 1.25,
      "learning_rate": 0.0004648447372078123,
      "loss": 5.3749,
      "mean_token_accuracy": 0.17275755405426024,
      "num_tokens": 5482885.0,
      "step": 3120
    },
    {
      "entropy": 5.504965353012085,
      "epoch": 2.449059561128527,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004646831243324457,
      "loss": 5.3177,
      "mean_token_accuracy": 0.16810073852539062,
      "num_tokens": 5491489.0,
      "step": 3125
    },
    {
      "entropy": 5.583372259140015,
      "epoch": 2.4529780564263324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004645211724625802,
      "loss": 5.3758,
      "mean_token_accuracy": 0.16375170946121215,
      "num_tokens": 5499694.0,
      "step": 3130
    },
    {
      "entropy": 5.44045352935791,
      "epoch": 2.456896551724138,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046435888188765015,
      "loss": 5.2726,
      "mean_token_accuracy": 0.1700295850634575,
      "num_tokens": 5508037.0,
      "step": 3135
    },
    {
      "entropy": 5.52275652885437,
      "epoch": 2.4608150470219434,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004641962528976951,
      "loss": 5.2497,
      "mean_token_accuracy": 0.17125702649354935,
      "num_tokens": 5516644.0,
      "step": 3140
    },
    {
      "entropy": 5.4940837860107425,
      "epoch": 2.4647335423197494,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004640332857833593,
      "loss": 5.3131,
      "mean_token_accuracy": 0.1654887914657593,
      "num_tokens": 5524789.0,
      "step": 3145
    },
    {
      "entropy": 5.524436664581299,
      "epoch": 2.468652037617555,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046386998083589156,
      "loss": 5.35,
      "mean_token_accuracy": 0.16829695850610732,
      "num_tokens": 5533739.0,
      "step": 3150
    },
    {
      "entropy": 5.635737133026123,
      "epoch": 2.4725705329153604,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004637063383471442,
      "loss": 5.38,
      "mean_token_accuracy": 0.16446826010942459,
      "num_tokens": 5542735.0,
      "step": 3155
    },
    {
      "entropy": 5.472494125366211,
      "epoch": 2.476489028213166,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046354235860957287,
      "loss": 5.2174,
      "mean_token_accuracy": 0.1728504106402397,
      "num_tokens": 5551272.0,
      "step": 3160
    },
    {
      "entropy": 5.540900659561157,
      "epoch": 2.480407523510972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004633780419162361,
      "loss": 5.4231,
      "mean_token_accuracy": 0.15702152997255325,
      "num_tokens": 5559868.0,
      "step": 3165
    },
    {
      "entropy": 5.465464353561401,
      "epoch": 2.4843260188087775,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046321338856079435,
      "loss": 5.2077,
      "mean_token_accuracy": 0.1737042561173439,
      "num_tokens": 5568276.0,
      "step": 3170
    },
    {
      "entropy": 5.465699577331543,
      "epoch": 2.488244514106583,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046304839883750987,
      "loss": 5.2948,
      "mean_token_accuracy": 0.16228433400392533,
      "num_tokens": 5576462.0,
      "step": 3175
    },
    {
      "entropy": 5.542299032211304,
      "epoch": 2.492163009404389,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000462883073041246,
      "loss": 5.4591,
      "mean_token_accuracy": 0.1608154818415642,
      "num_tokens": 5586044.0,
      "step": 3180
    },
    {
      "entropy": 5.41082797050476,
      "epoch": 2.4960815047021945,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004627174114674669,
      "loss": 5.2255,
      "mean_token_accuracy": 0.17130375206470488,
      "num_tokens": 5594626.0,
      "step": 3185
    },
    {
      "entropy": 5.577104806900024,
      "epoch": 2.5,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004625514144122365,
      "loss": 5.2935,
      "mean_token_accuracy": 0.16783207952976226,
      "num_tokens": 5603268.0,
      "step": 3190
    },
    {
      "entropy": 5.510541343688965,
      "epoch": 2.5039184952978055,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004623850821722185,
      "loss": 5.2773,
      "mean_token_accuracy": 0.1686388924717903,
      "num_tokens": 5612053.0,
      "step": 3195
    },
    {
      "entropy": 5.514151287078858,
      "epoch": 2.507836990595611,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004622184150446756,
      "loss": 5.5407,
      "mean_token_accuracy": 0.15114220529794692,
      "num_tokens": 5621297.0,
      "step": 3200
    },
    {
      "entropy": 5.616940927505493,
      "epoch": 2.511755485893417,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046205141332746904,
      "loss": 5.3606,
      "mean_token_accuracy": 0.16217829287052155,
      "num_tokens": 5629683.0,
      "step": 3205
    },
    {
      "entropy": 5.43563985824585,
      "epoch": 2.5156739811912225,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046188407731905787,
      "loss": 5.2808,
      "mean_token_accuracy": 0.1688768208026886,
      "num_tokens": 5638707.0,
      "step": 3210
    },
    {
      "entropy": 5.549165916442871,
      "epoch": 2.519592476489028,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004617164073184987,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1647924467921257,
      "num_tokens": 5648241.0,
      "step": 3215
    },
    {
      "entropy": 5.498393154144287,
      "epoch": 2.523510971786834,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046154840362544496,
      "loss": 5.3688,
      "mean_token_accuracy": 0.1656247302889824,
      "num_tokens": 5656743.0,
      "step": 3220
    },
    {
      "entropy": 5.500120210647583,
      "epoch": 2.5274294670846396,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004613800665401466,
      "loss": 5.3275,
      "mean_token_accuracy": 0.1651181012392044,
      "num_tokens": 5666418.0,
      "step": 3225
    },
    {
      "entropy": 5.475159645080566,
      "epoch": 2.531347962382445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004612113963634493,
      "loss": 5.3016,
      "mean_token_accuracy": 0.16872531920671463,
      "num_tokens": 5675404.0,
      "step": 3230
    },
    {
      "entropy": 5.504612731933594,
      "epoch": 2.535266457680251,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004610423933967938,
      "loss": 5.3853,
      "mean_token_accuracy": 0.15886692255735396,
      "num_tokens": 5684699.0,
      "step": 3235
    },
    {
      "entropy": 5.6432898998260494,
      "epoch": 2.5391849529780566,
      "grad_norm": 1.171875,
      "learning_rate": 0.000460873057942216,
      "loss": 5.3855,
      "mean_token_accuracy": 0.1671795815229416,
      "num_tokens": 5693219.0,
      "step": 3240
    },
    {
      "entropy": 5.5588963508605955,
      "epoch": 2.543103448275862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004607033903023458,
      "loss": 5.3927,
      "mean_token_accuracy": 0.1640296459197998,
      "num_tokens": 5702005.0,
      "step": 3245
    },
    {
      "entropy": 5.606112480163574,
      "epoch": 2.5470219435736676,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046053339078040674,
      "loss": 5.3993,
      "mean_token_accuracy": 0.16274697929620743,
      "num_tokens": 5710715.0,
      "step": 3250
    },
    {
      "entropy": 5.494814682006836,
      "epoch": 2.550940438871473,
      "grad_norm": 1.25,
      "learning_rate": 0.0004603630596802155,
      "loss": 5.3251,
      "mean_token_accuracy": 0.1691939041018486,
      "num_tokens": 5719444.0,
      "step": 3255
    },
    {
      "entropy": 5.4016355037689205,
      "epoch": 2.554858934169279,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004601923973061814,
      "loss": 5.3096,
      "mean_token_accuracy": 0.16056416630744935,
      "num_tokens": 5727755.0,
      "step": 3260
    },
    {
      "entropy": 5.500792837142944,
      "epoch": 2.5587774294670846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046002140396330575,
      "loss": 5.2502,
      "mean_token_accuracy": 0.16782210320234298,
      "num_tokens": 5736933.0,
      "step": 3265
    },
    {
      "entropy": 5.553884267807007,
      "epoch": 2.56269592476489,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045985007995718154,
      "loss": 5.4092,
      "mean_token_accuracy": 0.16856626868247987,
      "num_tokens": 5745831.0,
      "step": 3270
    },
    {
      "entropy": 5.579045343399048,
      "epoch": 2.566614420062696,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004596784255939923,
      "loss": 5.4442,
      "mean_token_accuracy": 0.1629267528653145,
      "num_tokens": 5755419.0,
      "step": 3275
    },
    {
      "entropy": 5.525725603103638,
      "epoch": 2.5705329153605017,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004595064411805123,
      "loss": 5.3328,
      "mean_token_accuracy": 0.17061960101127624,
      "num_tokens": 5764008.0,
      "step": 3280
    },
    {
      "entropy": 5.477341508865356,
      "epoch": 2.574451410658307,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004593341270241057,
      "loss": 5.2416,
      "mean_token_accuracy": 0.16688467860221862,
      "num_tokens": 5772129.0,
      "step": 3285
    },
    {
      "entropy": 5.519706869125367,
      "epoch": 2.5783699059561127,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004591614834327257,
      "loss": 5.3595,
      "mean_token_accuracy": 0.16659962236881257,
      "num_tokens": 5781079.0,
      "step": 3290
    },
    {
      "entropy": 5.471915292739868,
      "epoch": 2.5822884012539182,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045898851071491444,
      "loss": 5.3291,
      "mean_token_accuracy": 0.16874849945306777,
      "num_tokens": 5790444.0,
      "step": 3295
    },
    {
      "entropy": 5.4016204357147215,
      "epoch": 2.586206896551724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004588152091798022,
      "loss": 5.2211,
      "mean_token_accuracy": 0.1672128587961197,
      "num_tokens": 5799828.0,
      "step": 3300
    },
    {
      "entropy": 5.603023099899292,
      "epoch": 2.5901253918495297,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004586415791371069,
      "loss": 5.4821,
      "mean_token_accuracy": 0.15468620955944062,
      "num_tokens": 5808765.0,
      "step": 3305
    },
    {
      "entropy": 5.571247291564942,
      "epoch": 2.5940438871473352,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004584676208971336,
      "loss": 5.2691,
      "mean_token_accuracy": 0.1765100583434105,
      "num_tokens": 5817281.0,
      "step": 3310
    },
    {
      "entropy": 5.448766040802002,
      "epoch": 2.597962382445141,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045829333477077384,
      "loss": 5.3016,
      "mean_token_accuracy": 0.18109697848558426,
      "num_tokens": 5825741.0,
      "step": 3315
    },
    {
      "entropy": 5.539952754974365,
      "epoch": 2.6018808777429467,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004581187210695053,
      "loss": 5.2821,
      "mean_token_accuracy": 0.17550428509712218,
      "num_tokens": 5834466.0,
      "step": 3320
    },
    {
      "entropy": 5.431042957305908,
      "epoch": 2.6057993730407523,
      "grad_norm": 1.234375,
      "learning_rate": 0.000457943780105391,
      "loss": 5.3121,
      "mean_token_accuracy": 0.17115687280893327,
      "num_tokens": 5843463.0,
      "step": 3325
    },
    {
      "entropy": 5.540793180465698,
      "epoch": 2.6097178683385582,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045776851219107856,
      "loss": 5.3427,
      "mean_token_accuracy": 0.167041015625,
      "num_tokens": 5851885.0,
      "step": 3330
    },
    {
      "entropy": 5.56136908531189,
      "epoch": 2.6136363636363638,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045759291763980035,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17018300741910936,
      "num_tokens": 5861537.0,
      "step": 3335
    },
    {
      "entropy": 5.4394289493560795,
      "epoch": 2.6175548589341693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045741699676537227,
      "loss": 5.2759,
      "mean_token_accuracy": 0.1729188710451126,
      "num_tokens": 5870332.0,
      "step": 3340
    },
    {
      "entropy": 5.512276315689087,
      "epoch": 2.621473354231975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045724074988219343,
      "loss": 5.4836,
      "mean_token_accuracy": 0.16204925179481505,
      "num_tokens": 5879240.0,
      "step": 3345
    },
    {
      "entropy": 5.439336442947388,
      "epoch": 2.6253918495297803,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045706417730524565,
      "loss": 5.239,
      "mean_token_accuracy": 0.1624177500605583,
      "num_tokens": 5887961.0,
      "step": 3350
    },
    {
      "entropy": 5.539043331146241,
      "epoch": 2.6293103448275863,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004568872793500927,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16323541700839997,
      "num_tokens": 5896551.0,
      "step": 3355
    },
    {
      "entropy": 5.431582450866699,
      "epoch": 2.633228840125392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045671005633287986,
      "loss": 5.2424,
      "mean_token_accuracy": 0.17058624625205993,
      "num_tokens": 5905485.0,
      "step": 3360
    },
    {
      "entropy": 5.482703018188476,
      "epoch": 2.6371473354231973,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004565325085703336,
      "loss": 5.2917,
      "mean_token_accuracy": 0.16687317788600922,
      "num_tokens": 5915166.0,
      "step": 3365
    },
    {
      "entropy": 5.519888830184937,
      "epoch": 2.6410658307210033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004563546363797602,
      "loss": 5.3259,
      "mean_token_accuracy": 0.17161272913217546,
      "num_tokens": 5923782.0,
      "step": 3370
    },
    {
      "entropy": 5.537396860122681,
      "epoch": 2.644984326018809,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004561764400790465,
      "loss": 5.4332,
      "mean_token_accuracy": 0.16172593981027603,
      "num_tokens": 5931904.0,
      "step": 3375
    },
    {
      "entropy": 5.522799825668335,
      "epoch": 2.6489028213166144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045599791998665796,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17676885724067687,
      "num_tokens": 5940067.0,
      "step": 3380
    },
    {
      "entropy": 5.4546473026275635,
      "epoch": 2.6528213166144203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004558190764216389,
      "loss": 5.3056,
      "mean_token_accuracy": 0.1724096104502678,
      "num_tokens": 5948663.0,
      "step": 3385
    },
    {
      "entropy": 5.579775714874268,
      "epoch": 2.656739811912226,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004556399097036119,
      "loss": 5.3241,
      "mean_token_accuracy": 0.17269116044044494,
      "num_tokens": 5957224.0,
      "step": 3390
    },
    {
      "entropy": 5.540528202056885,
      "epoch": 2.6606583072100314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004554604201527768,
      "loss": 5.4198,
      "mean_token_accuracy": 0.16985767632722854,
      "num_tokens": 5965604.0,
      "step": 3395
    },
    {
      "entropy": 5.461326837539673,
      "epoch": 2.664576802507837,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045528060808991075,
      "loss": 5.2393,
      "mean_token_accuracy": 0.1762665867805481,
      "num_tokens": 5974309.0,
      "step": 3400
    },
    {
      "entropy": 5.470024061203003,
      "epoch": 2.6684952978056424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004551004738363669,
      "loss": 5.251,
      "mean_token_accuracy": 0.1705390602350235,
      "num_tokens": 5982315.0,
      "step": 3405
    },
    {
      "entropy": 5.452718305587768,
      "epoch": 2.6724137931034484,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045492001771407434,
      "loss": 5.2918,
      "mean_token_accuracy": 0.17495300024747848,
      "num_tokens": 5990515.0,
      "step": 3410
    },
    {
      "entropy": 5.520227289199829,
      "epoch": 2.676332288401254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004547392400455374,
      "loss": 5.4512,
      "mean_token_accuracy": 0.1606644168496132,
      "num_tokens": 6000431.0,
      "step": 3415
    },
    {
      "entropy": 5.5755597114562985,
      "epoch": 2.6802507836990594,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004545581411538353,
      "loss": 5.3709,
      "mean_token_accuracy": 0.16724153012037277,
      "num_tokens": 6009022.0,
      "step": 3420
    },
    {
      "entropy": 5.514227437973022,
      "epoch": 2.6841692789968654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045437672136262083,
      "loss": 5.3027,
      "mean_token_accuracy": 0.16582043319940568,
      "num_tokens": 6019330.0,
      "step": 3425
    },
    {
      "entropy": 5.459601211547851,
      "epoch": 2.688087774294671,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004541949809961208,
      "loss": 5.3159,
      "mean_token_accuracy": 0.17988408952951432,
      "num_tokens": 6027492.0,
      "step": 3430
    },
    {
      "entropy": 5.524388551712036,
      "epoch": 2.6920062695924765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004540129203791346,
      "loss": 5.3378,
      "mean_token_accuracy": 0.1686471864581108,
      "num_tokens": 6036467.0,
      "step": 3435
    },
    {
      "entropy": 5.466223049163818,
      "epoch": 2.695924764890282,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045383053983703413,
      "loss": 5.2957,
      "mean_token_accuracy": 0.1695375457406044,
      "num_tokens": 6045711.0,
      "step": 3440
    },
    {
      "entropy": 5.500885486602783,
      "epoch": 2.6998432601880875,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045364783969576296,
      "loss": 5.3736,
      "mean_token_accuracy": 0.1609950840473175,
      "num_tokens": 6054626.0,
      "step": 3445
    },
    {
      "entropy": 5.559614562988282,
      "epoch": 2.7037617554858935,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045346482028183583,
      "loss": 5.4288,
      "mean_token_accuracy": 0.16276882886886596,
      "num_tokens": 6064118.0,
      "step": 3450
    },
    {
      "entropy": 5.523534679412842,
      "epoch": 2.707680250783699,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045328148192233823,
      "loss": 5.3338,
      "mean_token_accuracy": 0.16942061334848404,
      "num_tokens": 6072723.0,
      "step": 3455
    },
    {
      "entropy": 5.520214319229126,
      "epoch": 2.7115987460815045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004530978249449254,
      "loss": 5.3223,
      "mean_token_accuracy": 0.1724524199962616,
      "num_tokens": 6081535.0,
      "step": 3460
    },
    {
      "entropy": 5.474754762649536,
      "epoch": 2.7155172413793105,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004529138496778222,
      "loss": 5.2672,
      "mean_token_accuracy": 0.16802340596914292,
      "num_tokens": 6090524.0,
      "step": 3465
    },
    {
      "entropy": 5.498336887359619,
      "epoch": 2.719435736677116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004527295564498222,
      "loss": 5.3441,
      "mean_token_accuracy": 0.166372212767601,
      "num_tokens": 6099870.0,
      "step": 3470
    },
    {
      "entropy": 5.500373983383179,
      "epoch": 2.7233542319749215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004525449455902874,
      "loss": 5.3115,
      "mean_token_accuracy": 0.16563377976417543,
      "num_tokens": 6107760.0,
      "step": 3475
    },
    {
      "entropy": 5.528516864776611,
      "epoch": 2.7272727272727275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004523600174291473,
      "loss": 5.3948,
      "mean_token_accuracy": 0.16702790558338165,
      "num_tokens": 6117135.0,
      "step": 3480
    },
    {
      "entropy": 5.451120853424072,
      "epoch": 2.731191222570533,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004521747722968985,
      "loss": 5.3347,
      "mean_token_accuracy": 0.16837533414363862,
      "num_tokens": 6125711.0,
      "step": 3485
    },
    {
      "entropy": 5.553459644317627,
      "epoch": 2.7351097178683386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045198921052460396,
      "loss": 5.3881,
      "mean_token_accuracy": 0.16275455951690673,
      "num_tokens": 6134623.0,
      "step": 3490
    },
    {
      "entropy": 5.583516311645508,
      "epoch": 2.739028213166144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004518033324438928,
      "loss": 5.3859,
      "mean_token_accuracy": 0.16478580981492996,
      "num_tokens": 6143228.0,
      "step": 3495
    },
    {
      "entropy": 5.478849411010742,
      "epoch": 2.7429467084639496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004516171383869593,
      "loss": 5.3613,
      "mean_token_accuracy": 0.16525180339813234,
      "num_tokens": 6152021.0,
      "step": 3500
    },
    {
      "epoch": 2.7429467084639496,
      "eval_entropy": 5.374510490617086,
      "eval_loss": 5.8104634284973145,
      "eval_mean_token_accuracy": 0.15645403335549574,
      "eval_num_tokens": 6152021.0,
      "eval_runtime": 2.8448,
      "eval_samples_per_second": 1448.984,
      "eval_steps_per_second": 181.387,
      "step": 3500
    },
    {
      "entropy": 5.565016078948974,
      "epoch": 2.7468652037617556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045143062868656234,
      "loss": 5.422,
      "mean_token_accuracy": 0.1618006110191345,
      "num_tokens": 6161389.0,
      "step": 3505
    },
    {
      "entropy": 5.6011796474456785,
      "epoch": 2.750783699059561,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000451243803676025,
      "loss": 5.346,
      "mean_token_accuracy": 0.16280375123023988,
      "num_tokens": 6170381.0,
      "step": 3510
    },
    {
      "entropy": 5.447231483459473,
      "epoch": 2.7547021943573666,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045105666368923397,
      "loss": 5.2867,
      "mean_token_accuracy": 0.16635821759700775,
      "num_tokens": 6179071.0,
      "step": 3515
    },
    {
      "entropy": 5.380543994903564,
      "epoch": 2.7586206896551726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045086920906063866,
      "loss": 5.3138,
      "mean_token_accuracy": 0.16474552005529403,
      "num_tokens": 6187896.0,
      "step": 3520
    },
    {
      "entropy": 5.468551397323608,
      "epoch": 2.762539184952978,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045068144012525095,
      "loss": 5.2773,
      "mean_token_accuracy": 0.177284175157547,
      "num_tokens": 6196328.0,
      "step": 3525
    },
    {
      "entropy": 5.445127344131469,
      "epoch": 2.7664576802507836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045049335721864426,
      "loss": 5.3314,
      "mean_token_accuracy": 0.16377530992031097,
      "num_tokens": 6204703.0,
      "step": 3530
    },
    {
      "entropy": 5.461795330047607,
      "epoch": 2.7703761755485896,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045030496067695336,
      "loss": 5.3062,
      "mean_token_accuracy": 0.16890790015459062,
      "num_tokens": 6213496.0,
      "step": 3535
    },
    {
      "entropy": 5.478771924972534,
      "epoch": 2.774294670846395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004501162508368733,
      "loss": 5.267,
      "mean_token_accuracy": 0.17821568995714188,
      "num_tokens": 6222112.0,
      "step": 3540
    },
    {
      "entropy": 5.510504865646363,
      "epoch": 2.7782131661442007,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004499272280356594,
      "loss": 5.3775,
      "mean_token_accuracy": 0.15529398024082183,
      "num_tokens": 6230288.0,
      "step": 3545
    },
    {
      "entropy": 5.474312973022461,
      "epoch": 2.782131661442006,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004497378926111257,
      "loss": 5.2414,
      "mean_token_accuracy": 0.1762033596634865,
      "num_tokens": 6238608.0,
      "step": 3550
    },
    {
      "entropy": 5.5118636131286625,
      "epoch": 2.7860501567398117,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004495482449016456,
      "loss": 5.3604,
      "mean_token_accuracy": 0.16791626214981079,
      "num_tokens": 6247902.0,
      "step": 3555
    },
    {
      "entropy": 5.489766931533813,
      "epoch": 2.7899686520376177,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004493582852461501,
      "loss": 5.348,
      "mean_token_accuracy": 0.16909869015216827,
      "num_tokens": 6257433.0,
      "step": 3560
    },
    {
      "entropy": 5.421799612045288,
      "epoch": 2.793887147335423,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004491680139841281,
      "loss": 5.3253,
      "mean_token_accuracy": 0.16281854957342148,
      "num_tokens": 6266017.0,
      "step": 3565
    },
    {
      "entropy": 5.529870796203613,
      "epoch": 2.7978056426332287,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000448977431455625,
      "loss": 5.4052,
      "mean_token_accuracy": 0.15905255526304246,
      "num_tokens": 6274759.0,
      "step": 3570
    },
    {
      "entropy": 5.5512055397033695,
      "epoch": 2.8017241379310347,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044878653800124285,
      "loss": 5.2697,
      "mean_token_accuracy": 0.16887140423059463,
      "num_tokens": 6283699.0,
      "step": 3575
    },
    {
      "entropy": 5.444827842712402,
      "epoch": 2.80564263322884,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004485953339621391,
      "loss": 5.306,
      "mean_token_accuracy": 0.1720012679696083,
      "num_tokens": 6292562.0,
      "step": 3580
    },
    {
      "entropy": 5.465062952041626,
      "epoch": 2.8095611285266457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004484038196800265,
      "loss": 5.2776,
      "mean_token_accuracy": 0.17041739225387573,
      "num_tokens": 6300928.0,
      "step": 3585
    },
    {
      "entropy": 5.419831705093384,
      "epoch": 2.8134796238244513,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004482119954971719,
      "loss": 5.2028,
      "mean_token_accuracy": 0.17201321870088576,
      "num_tokens": 6309994.0,
      "step": 3590
    },
    {
      "entropy": 5.478939580917358,
      "epoch": 2.817398119122257,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044801986175639635,
      "loss": 5.3101,
      "mean_token_accuracy": 0.17052568197250367,
      "num_tokens": 6318856.0,
      "step": 3595
    },
    {
      "entropy": 5.457127952575684,
      "epoch": 2.8213166144200628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004478274188010741,
      "loss": 5.3152,
      "mean_token_accuracy": 0.16498181521892546,
      "num_tokens": 6327274.0,
      "step": 3600
    },
    {
      "entropy": 5.577486801147461,
      "epoch": 2.8252351097178683,
      "grad_norm": 1.25,
      "learning_rate": 0.00044763466697513173,
      "loss": 5.3782,
      "mean_token_accuracy": 0.17814266681671143,
      "num_tokens": 6335957.0,
      "step": 3605
    },
    {
      "entropy": 5.54471526145935,
      "epoch": 2.829153605015674,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044744160662304805,
      "loss": 5.392,
      "mean_token_accuracy": 0.17500135749578477,
      "num_tokens": 6344673.0,
      "step": 3610
    },
    {
      "entropy": 5.424971103668213,
      "epoch": 2.83307210031348,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044724823808985325,
      "loss": 5.2555,
      "mean_token_accuracy": 0.17783398926258087,
      "num_tokens": 6353085.0,
      "step": 3615
    },
    {
      "entropy": 5.4651954650878904,
      "epoch": 2.8369905956112853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004470545617211283,
      "loss": 5.2962,
      "mean_token_accuracy": 0.16942965239286423,
      "num_tokens": 6362107.0,
      "step": 3620
    },
    {
      "entropy": 5.4898402214050295,
      "epoch": 2.840909090909091,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044686057786300423,
      "loss": 5.2969,
      "mean_token_accuracy": 0.16850359737873077,
      "num_tokens": 6370381.0,
      "step": 3625
    },
    {
      "entropy": 5.497567844390869,
      "epoch": 2.844827586206897,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044666628686216154,
      "loss": 5.3333,
      "mean_token_accuracy": 0.16255157738924025,
      "num_tokens": 6378790.0,
      "step": 3630
    },
    {
      "entropy": 5.470732355117798,
      "epoch": 2.8487460815047023,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044647168906583,
      "loss": 5.3134,
      "mean_token_accuracy": 0.16725441068410873,
      "num_tokens": 6387997.0,
      "step": 3635
    },
    {
      "entropy": 5.531575345993042,
      "epoch": 2.852664576802508,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044627678482178716,
      "loss": 5.3213,
      "mean_token_accuracy": 0.17100094854831696,
      "num_tokens": 6397458.0,
      "step": 3640
    },
    {
      "entropy": 5.477775812149048,
      "epoch": 2.8565830721003134,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004460815744783587,
      "loss": 5.273,
      "mean_token_accuracy": 0.167954421043396,
      "num_tokens": 6405481.0,
      "step": 3645
    },
    {
      "entropy": 5.481853246688843,
      "epoch": 2.860501567398119,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000445886058384417,
      "loss": 5.3921,
      "mean_token_accuracy": 0.16211470365524291,
      "num_tokens": 6414087.0,
      "step": 3650
    },
    {
      "entropy": 5.508256721496582,
      "epoch": 2.864420062695925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004456902368893811,
      "loss": 5.197,
      "mean_token_accuracy": 0.1776396945118904,
      "num_tokens": 6421934.0,
      "step": 3655
    },
    {
      "entropy": 5.478158855438233,
      "epoch": 2.8683385579937304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004454941103432158,
      "loss": 5.2415,
      "mean_token_accuracy": 0.17561646401882172,
      "num_tokens": 6431028.0,
      "step": 3660
    },
    {
      "entropy": 5.337583780288696,
      "epoch": 2.872257053291536,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044529767909643093,
      "loss": 5.218,
      "mean_token_accuracy": 0.1768454745411873,
      "num_tokens": 6439913.0,
      "step": 3665
    },
    {
      "entropy": 5.524216508865356,
      "epoch": 2.876175548589342,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004451009435000811,
      "loss": 5.2758,
      "mean_token_accuracy": 0.16853948384523393,
      "num_tokens": 6449161.0,
      "step": 3670
    },
    {
      "entropy": 5.492571067810059,
      "epoch": 2.8800940438871474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004449039039057647,
      "loss": 5.3944,
      "mean_token_accuracy": 0.16869514882564546,
      "num_tokens": 6457834.0,
      "step": 3675
    },
    {
      "entropy": 5.412122774124145,
      "epoch": 2.884012539184953,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044470656066562336,
      "loss": 5.215,
      "mean_token_accuracy": 0.1781695917248726,
      "num_tokens": 6465879.0,
      "step": 3680
    },
    {
      "entropy": 5.4064820289611815,
      "epoch": 2.887931034482759,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004445089141323415,
      "loss": 5.2255,
      "mean_token_accuracy": 0.17291576862335206,
      "num_tokens": 6474801.0,
      "step": 3685
    },
    {
      "entropy": 5.454866647720337,
      "epoch": 2.8918495297805644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044431096465914554,
      "loss": 5.2871,
      "mean_token_accuracy": 0.17472967207431794,
      "num_tokens": 6483532.0,
      "step": 3690
    },
    {
      "entropy": 5.474235343933105,
      "epoch": 2.89576802507837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044411271259980315,
      "loss": 5.2485,
      "mean_token_accuracy": 0.17116280347108842,
      "num_tokens": 6491372.0,
      "step": 3695
    },
    {
      "entropy": 5.389340114593506,
      "epoch": 2.8996865203761755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004439141583086231,
      "loss": 5.2555,
      "mean_token_accuracy": 0.1679220274090767,
      "num_tokens": 6500001.0,
      "step": 3700
    },
    {
      "entropy": 5.51658992767334,
      "epoch": 2.903605015673981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044371530214045395,
      "loss": 5.352,
      "mean_token_accuracy": 0.16550127565860748,
      "num_tokens": 6507750.0,
      "step": 3705
    },
    {
      "entropy": 5.426118183135986,
      "epoch": 2.907523510971787,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044351614445068413,
      "loss": 5.2479,
      "mean_token_accuracy": 0.16929992139339448,
      "num_tokens": 6517888.0,
      "step": 3710
    },
    {
      "entropy": 5.504793167114258,
      "epoch": 2.9114420062695925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044331668559524043,
      "loss": 5.3676,
      "mean_token_accuracy": 0.16317006200551987,
      "num_tokens": 6526573.0,
      "step": 3715
    },
    {
      "entropy": 5.440666913986206,
      "epoch": 2.915360501567398,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004431169259305883,
      "loss": 5.309,
      "mean_token_accuracy": 0.1699930503964424,
      "num_tokens": 6536544.0,
      "step": 3720
    },
    {
      "entropy": 5.5378429889678955,
      "epoch": 2.919278996865204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004429168658137306,
      "loss": 5.3225,
      "mean_token_accuracy": 0.16724765300750732,
      "num_tokens": 6544982.0,
      "step": 3725
    },
    {
      "entropy": 5.444597244262695,
      "epoch": 2.9231974921630095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044271650560220746,
      "loss": 5.3524,
      "mean_token_accuracy": 0.1679960548877716,
      "num_tokens": 6553664.0,
      "step": 3730
    },
    {
      "entropy": 5.47899866104126,
      "epoch": 2.927115987460815,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044251584565409464,
      "loss": 5.3182,
      "mean_token_accuracy": 0.17241291105747222,
      "num_tokens": 6562234.0,
      "step": 3735
    },
    {
      "entropy": 5.441190624237061,
      "epoch": 2.9310344827586206,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004423148863280044,
      "loss": 5.2889,
      "mean_token_accuracy": 0.17095829248428346,
      "num_tokens": 6570707.0,
      "step": 3740
    },
    {
      "entropy": 5.483555459976197,
      "epoch": 2.934952978056426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044211362798308334,
      "loss": 5.3315,
      "mean_token_accuracy": 0.1714997708797455,
      "num_tokens": 6580054.0,
      "step": 3745
    },
    {
      "entropy": 5.471543598175049,
      "epoch": 2.938871473354232,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004419120709790129,
      "loss": 5.3877,
      "mean_token_accuracy": 0.16535573005676268,
      "num_tokens": 6588585.0,
      "step": 3750
    },
    {
      "entropy": 5.527312517166138,
      "epoch": 2.9427899686520376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044171021567600814,
      "loss": 5.3688,
      "mean_token_accuracy": 0.16477798372507096,
      "num_tokens": 6597290.0,
      "step": 3755
    },
    {
      "entropy": 5.513667488098145,
      "epoch": 2.946708463949843,
      "grad_norm": 1.25,
      "learning_rate": 0.00044150806243481715,
      "loss": 5.304,
      "mean_token_accuracy": 0.16890741586685182,
      "num_tokens": 6604966.0,
      "step": 3760
    },
    {
      "entropy": 5.532940292358399,
      "epoch": 2.950626959247649,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004413056116167206,
      "loss": 5.2708,
      "mean_token_accuracy": 0.17608776092529296,
      "num_tokens": 6613588.0,
      "step": 3765
    },
    {
      "entropy": 5.579522323608399,
      "epoch": 2.9545454545454546,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004411028635835309,
      "loss": 5.4958,
      "mean_token_accuracy": 0.15222854167222977,
      "num_tokens": 6622907.0,
      "step": 3770
    },
    {
      "entropy": 5.457161331176758,
      "epoch": 2.95846394984326,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004408998186975917,
      "loss": 5.2812,
      "mean_token_accuracy": 0.17085075974464417,
      "num_tokens": 6632074.0,
      "step": 3775
    },
    {
      "entropy": 5.524838972091675,
      "epoch": 2.962382445141066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044069647732177696,
      "loss": 5.3123,
      "mean_token_accuracy": 0.1721285179257393,
      "num_tokens": 6640595.0,
      "step": 3780
    },
    {
      "entropy": 5.529404592514038,
      "epoch": 2.9663009404388716,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044049283981949103,
      "loss": 5.3632,
      "mean_token_accuracy": 0.166115140914917,
      "num_tokens": 6649861.0,
      "step": 3785
    },
    {
      "entropy": 5.52064962387085,
      "epoch": 2.970219435736677,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004402889065546667,
      "loss": 5.3983,
      "mean_token_accuracy": 0.16248751878738404,
      "num_tokens": 6659157.0,
      "step": 3790
    },
    {
      "entropy": 5.47776460647583,
      "epoch": 2.9741379310344827,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044008467789176625,
      "loss": 5.3561,
      "mean_token_accuracy": 0.16655384451150895,
      "num_tokens": 6667979.0,
      "step": 3795
    },
    {
      "entropy": 5.520320987701416,
      "epoch": 2.978056426332288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004398801541957791,
      "loss": 5.3105,
      "mean_token_accuracy": 0.16469419598579407,
      "num_tokens": 6677239.0,
      "step": 3800
    },
    {
      "entropy": 5.5184122085571286,
      "epoch": 2.981974921630094,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004396753358322223,
      "loss": 5.4317,
      "mean_token_accuracy": 0.161855086684227,
      "num_tokens": 6687317.0,
      "step": 3805
    },
    {
      "entropy": 5.447925186157226,
      "epoch": 2.9858934169278997,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004394702231671396,
      "loss": 5.3179,
      "mean_token_accuracy": 0.16239897012710572,
      "num_tokens": 6695732.0,
      "step": 3810
    },
    {
      "entropy": 5.462766981124878,
      "epoch": 2.989811912225705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004392648165671004,
      "loss": 5.2298,
      "mean_token_accuracy": 0.17857529520988463,
      "num_tokens": 6704272.0,
      "step": 3815
    },
    {
      "entropy": 5.417524147033691,
      "epoch": 2.993730407523511,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004390591163991998,
      "loss": 5.2739,
      "mean_token_accuracy": 0.17327528148889543,
      "num_tokens": 6713355.0,
      "step": 3820
    },
    {
      "entropy": 5.523505973815918,
      "epoch": 2.9976489028213167,
      "grad_norm": 1.25,
      "learning_rate": 0.00043885312303105725,
      "loss": 5.3575,
      "mean_token_accuracy": 0.16509936153888702,
      "num_tokens": 6721952.0,
      "step": 3825
    },
    {
      "entropy": 5.448745679855347,
      "epoch": 3.001567398119122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004386468368308163,
      "loss": 5.1564,
      "mean_token_accuracy": 0.1759412795305252,
      "num_tokens": 6730582.0,
      "step": 3830
    },
    {
      "entropy": 5.504661750793457,
      "epoch": 3.0054858934169277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004384402581671438,
      "loss": 4.9509,
      "mean_token_accuracy": 0.18449245840311052,
      "num_tokens": 6739370.0,
      "step": 3835
    },
    {
      "entropy": 5.426512622833252,
      "epoch": 3.0094043887147337,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004382333874092295,
      "loss": 4.9251,
      "mean_token_accuracy": 0.1898578643798828,
      "num_tokens": 6748282.0,
      "step": 3840
    },
    {
      "entropy": 5.472595262527466,
      "epoch": 3.0133228840125392,
      "grad_norm": 1.125,
      "learning_rate": 0.00043802622492678466,
      "loss": 5.0423,
      "mean_token_accuracy": 0.1788592368364334,
      "num_tokens": 6757622.0,
      "step": 3845
    },
    {
      "entropy": 5.523442554473877,
      "epoch": 3.0172413793103448,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004378187710900426,
      "loss": 4.9788,
      "mean_token_accuracy": 0.1851514607667923,
      "num_tokens": 6765780.0,
      "step": 3850
    },
    {
      "entropy": 5.43020281791687,
      "epoch": 3.0211598746081503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043761102626975674,
      "loss": 4.9903,
      "mean_token_accuracy": 0.1907791569828987,
      "num_tokens": 6774343.0,
      "step": 3855
    },
    {
      "entropy": 5.420247173309326,
      "epoch": 3.0250783699059562,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004374029908372007,
      "loss": 5.0101,
      "mean_token_accuracy": 0.18541126996278762,
      "num_tokens": 6783765.0,
      "step": 3860
    },
    {
      "entropy": 5.531079530715942,
      "epoch": 3.0289968652037618,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043719466516416774,
      "loss": 5.0514,
      "mean_token_accuracy": 0.18391268104314804,
      "num_tokens": 6792102.0,
      "step": 3865
    },
    {
      "entropy": 5.500016784667968,
      "epoch": 3.0329153605015673,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043698604962296946,
      "loss": 4.9672,
      "mean_token_accuracy": 0.18078313618898392,
      "num_tokens": 6801435.0,
      "step": 3870
    },
    {
      "entropy": 5.379806327819824,
      "epoch": 3.0368338557993733,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043677714458643566,
      "loss": 5.009,
      "mean_token_accuracy": 0.18003130108118057,
      "num_tokens": 6810409.0,
      "step": 3875
    },
    {
      "entropy": 5.355454921722412,
      "epoch": 3.040752351097179,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043656795042791357,
      "loss": 4.8785,
      "mean_token_accuracy": 0.1964130222797394,
      "num_tokens": 6818752.0,
      "step": 3880
    },
    {
      "entropy": 5.422845077514649,
      "epoch": 3.0446708463949843,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004363584675212671,
      "loss": 4.9879,
      "mean_token_accuracy": 0.18382852673530578,
      "num_tokens": 6827880.0,
      "step": 3885
    },
    {
      "entropy": 5.386878442764282,
      "epoch": 3.04858934169279,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004361486962408761,
      "loss": 4.9777,
      "mean_token_accuracy": 0.18993838876485825,
      "num_tokens": 6835865.0,
      "step": 3890
    },
    {
      "entropy": 5.344284200668335,
      "epoch": 3.052507836990596,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004359386369616359,
      "loss": 4.9347,
      "mean_token_accuracy": 0.19007452875375747,
      "num_tokens": 6843851.0,
      "step": 3895
    },
    {
      "entropy": 5.391309690475464,
      "epoch": 3.0564263322884013,
      "grad_norm": 1.25,
      "learning_rate": 0.0004357282900589565,
      "loss": 4.9787,
      "mean_token_accuracy": 0.18415525555610657,
      "num_tokens": 6852649.0,
      "step": 3900
    },
    {
      "entropy": 5.384064531326294,
      "epoch": 3.060344827586207,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043551765590876183,
      "loss": 4.9373,
      "mean_token_accuracy": 0.18840955942869186,
      "num_tokens": 6861607.0,
      "step": 3905
    },
    {
      "entropy": 5.383556270599366,
      "epoch": 3.0642633228840124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004353067348874894,
      "loss": 4.9616,
      "mean_token_accuracy": 0.18058305531740187,
      "num_tokens": 6870650.0,
      "step": 3910
    },
    {
      "entropy": 5.421644115447998,
      "epoch": 3.0681818181818183,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043509552737208923,
      "loss": 5.0131,
      "mean_token_accuracy": 0.17322253733873366,
      "num_tokens": 6879820.0,
      "step": 3915
    },
    {
      "entropy": 5.332171058654785,
      "epoch": 3.072100313479624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004348840337400233,
      "loss": 4.9186,
      "mean_token_accuracy": 0.1910848930478096,
      "num_tokens": 6888329.0,
      "step": 3920
    },
    {
      "entropy": 5.355079317092896,
      "epoch": 3.0760188087774294,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043467225436926517,
      "loss": 4.9963,
      "mean_token_accuracy": 0.18212777376174927,
      "num_tokens": 6897635.0,
      "step": 3925
    },
    {
      "entropy": 5.435818099975586,
      "epoch": 3.079937304075235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004344601896382988,
      "loss": 5.0451,
      "mean_token_accuracy": 0.18707363903522492,
      "num_tokens": 6907275.0,
      "step": 3930
    },
    {
      "entropy": 5.403489303588867,
      "epoch": 3.083855799373041,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043424783992611837,
      "loss": 4.982,
      "mean_token_accuracy": 0.18048081845045089,
      "num_tokens": 6916377.0,
      "step": 3935
    },
    {
      "entropy": 5.4124561786651615,
      "epoch": 3.0877742946708464,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043403520561222705,
      "loss": 4.8966,
      "mean_token_accuracy": 0.19746431559324265,
      "num_tokens": 6924355.0,
      "step": 3940
    },
    {
      "entropy": 5.387140226364136,
      "epoch": 3.091692789968652,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004338222870766371,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18432460874319076,
      "num_tokens": 6933328.0,
      "step": 3945
    },
    {
      "entropy": 5.439013242721558,
      "epoch": 3.0956112852664575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043360908469986827,
      "loss": 4.9834,
      "mean_token_accuracy": 0.18698500990867614,
      "num_tokens": 6942189.0,
      "step": 3950
    },
    {
      "entropy": 5.369312524795532,
      "epoch": 3.0995297805642634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004333955988629478,
      "loss": 4.9657,
      "mean_token_accuracy": 0.19216072261333467,
      "num_tokens": 6951188.0,
      "step": 3955
    },
    {
      "entropy": 5.316864252090454,
      "epoch": 3.103448275862069,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043318182994740945,
      "loss": 4.9186,
      "mean_token_accuracy": 0.19871540516614913,
      "num_tokens": 6958718.0,
      "step": 3960
    },
    {
      "entropy": 5.357178735733032,
      "epoch": 3.1073667711598745,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004329677783352931,
      "loss": 4.9773,
      "mean_token_accuracy": 0.18394005298614502,
      "num_tokens": 6968040.0,
      "step": 3965
    },
    {
      "entropy": 5.40136833190918,
      "epoch": 3.1112852664576804,
      "grad_norm": 1.25,
      "learning_rate": 0.0004327534444091436,
      "loss": 4.9912,
      "mean_token_accuracy": 0.18011486530303955,
      "num_tokens": 6976977.0,
      "step": 3970
    },
    {
      "entropy": 5.367156744003296,
      "epoch": 3.115203761755486,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043253882855201037,
      "loss": 4.9798,
      "mean_token_accuracy": 0.1904297098517418,
      "num_tokens": 6985859.0,
      "step": 3975
    },
    {
      "entropy": 5.269969749450683,
      "epoch": 3.1191222570532915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043232393114744683,
      "loss": 4.8928,
      "mean_token_accuracy": 0.19323855340480806,
      "num_tokens": 6994031.0,
      "step": 3980
    },
    {
      "entropy": 5.301603364944458,
      "epoch": 3.123040752351097,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004321087525795095,
      "loss": 4.9653,
      "mean_token_accuracy": 0.17997159510850907,
      "num_tokens": 7002527.0,
      "step": 3985
    },
    {
      "entropy": 5.39128475189209,
      "epoch": 3.126959247648903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004318932932327573,
      "loss": 5.0551,
      "mean_token_accuracy": 0.18612158298492432,
      "num_tokens": 7011983.0,
      "step": 3990
    },
    {
      "entropy": 5.441968250274658,
      "epoch": 3.1308777429467085,
      "grad_norm": 1.265625,
      "learning_rate": 0.000431677553492251,
      "loss": 5.006,
      "mean_token_accuracy": 0.18446675539016724,
      "num_tokens": 7020819.0,
      "step": 3995
    },
    {
      "entropy": 5.314369440078735,
      "epoch": 3.134796238244514,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043146153374355256,
      "loss": 4.8806,
      "mean_token_accuracy": 0.19490227550268174,
      "num_tokens": 7029257.0,
      "step": 4000
    },
    {
      "epoch": 3.134796238244514,
      "eval_entropy": 5.243911954783654,
      "eval_loss": 5.779608726501465,
      "eval_mean_token_accuracy": 0.1602230450326158,
      "eval_num_tokens": 7029257.0,
      "eval_runtime": 3.0149,
      "eval_samples_per_second": 1367.187,
      "eval_steps_per_second": 171.147,
      "step": 4000
    },
    {
      "entropy": 5.3803033351898195,
      "epoch": 3.1387147335423196,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043124523437272427,
      "loss": 5.0639,
      "mean_token_accuracy": 0.1767251282930374,
      "num_tokens": 7038942.0,
      "step": 4005
    },
    {
      "entropy": 5.401314163208008,
      "epoch": 3.1426332288401255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004310286557663282,
      "loss": 5.0279,
      "mean_token_accuracy": 0.19026288092136384,
      "num_tokens": 7048074.0,
      "step": 4010
    },
    {
      "entropy": 5.389622688293457,
      "epoch": 3.146551724137931,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004308117983114254,
      "loss": 4.9332,
      "mean_token_accuracy": 0.19302188009023666,
      "num_tokens": 7056887.0,
      "step": 4015
    },
    {
      "entropy": 5.353729200363159,
      "epoch": 3.1504702194357366,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004305946623955754,
      "loss": 4.9188,
      "mean_token_accuracy": 0.19034417271614074,
      "num_tokens": 7065462.0,
      "step": 4020
    },
    {
      "entropy": 5.403734874725342,
      "epoch": 3.1543887147335425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043037724840683516,
      "loss": 5.0471,
      "mean_token_accuracy": 0.18127024173736572,
      "num_tokens": 7074188.0,
      "step": 4025
    },
    {
      "entropy": 5.369546175003052,
      "epoch": 3.158307210031348,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00043015955673375876,
      "loss": 4.9972,
      "mean_token_accuracy": 0.18307553231716156,
      "num_tokens": 7082282.0,
      "step": 4030
    },
    {
      "entropy": 5.415471887588501,
      "epoch": 3.1622257053291536,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004299415877653966,
      "loss": 5.0505,
      "mean_token_accuracy": 0.18183055967092515,
      "num_tokens": 7090631.0,
      "step": 4035
    },
    {
      "entropy": 5.39145679473877,
      "epoch": 3.166144200626959,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004297233418912945,
      "loss": 5.0008,
      "mean_token_accuracy": 0.1812889888882637,
      "num_tokens": 7099546.0,
      "step": 4040
    },
    {
      "entropy": 5.425076675415039,
      "epoch": 3.170062695924765,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004295048195014932,
      "loss": 5.0386,
      "mean_token_accuracy": 0.1807874083518982,
      "num_tokens": 7109423.0,
      "step": 4045
    },
    {
      "entropy": 5.337812852859497,
      "epoch": 3.1739811912225706,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004292860209865277,
      "loss": 4.9091,
      "mean_token_accuracy": 0.1873628690838814,
      "num_tokens": 7117717.0,
      "step": 4050
    },
    {
      "entropy": 5.357939147949219,
      "epoch": 3.177899686520376,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004290669467374263,
      "loss": 5.1342,
      "mean_token_accuracy": 0.1729935497045517,
      "num_tokens": 7126573.0,
      "step": 4055
    },
    {
      "entropy": 5.324204778671264,
      "epoch": 3.1818181818181817,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00042884759714571037,
      "loss": 5.0284,
      "mean_token_accuracy": 0.18501825779676437,
      "num_tokens": 7134471.0,
      "step": 4060
    },
    {
      "entropy": 5.307945966720581,
      "epoch": 3.1857366771159876,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004286279726033932,
      "loss": 4.9595,
      "mean_token_accuracy": 0.18553654551506044,
      "num_tokens": 7143092.0,
      "step": 4065
    },
    {
      "entropy": 5.415725898742676,
      "epoch": 3.189655172413793,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042840807350297933,
      "loss": 5.0168,
      "mean_token_accuracy": 0.18869081735610962,
      "num_tokens": 7152137.0,
      "step": 4070
    },
    {
      "entropy": 5.3736060619354244,
      "epoch": 3.1935736677115987,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042818790023746407,
      "loss": 4.9757,
      "mean_token_accuracy": 0.18085607439279555,
      "num_tokens": 7160037.0,
      "step": 4075
    },
    {
      "entropy": 5.327694940567016,
      "epoch": 3.197492163009404,
      "grad_norm": 1.25,
      "learning_rate": 0.00042796745320033296,
      "loss": 4.9875,
      "mean_token_accuracy": 0.17759600132703782,
      "num_tokens": 7168354.0,
      "step": 4080
    },
    {
      "entropy": 5.376115369796753,
      "epoch": 3.20141065830721,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042774673278556043,
      "loss": 5.0581,
      "mean_token_accuracy": 0.18148307502269745,
      "num_tokens": 7176757.0,
      "step": 4085
    },
    {
      "entropy": 5.403470993041992,
      "epoch": 3.2053291536050157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004275257393876097,
      "loss": 5.0398,
      "mean_token_accuracy": 0.17965521514415742,
      "num_tokens": 7185489.0,
      "step": 4090
    },
    {
      "entropy": 5.433120775222778,
      "epoch": 3.209247648902821,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004273044734014318,
      "loss": 4.9832,
      "mean_token_accuracy": 0.18617940098047256,
      "num_tokens": 7194431.0,
      "step": 4095
    },
    {
      "entropy": 5.285513019561767,
      "epoch": 3.2131661442006267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042708293522246486,
      "loss": 4.9425,
      "mean_token_accuracy": 0.1923668012022972,
      "num_tokens": 7203074.0,
      "step": 4100
    },
    {
      "entropy": 5.3730559825897215,
      "epoch": 3.2170846394984327,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004268611252466337,
      "loss": 4.9896,
      "mean_token_accuracy": 0.1884973794221878,
      "num_tokens": 7211692.0,
      "step": 4105
    },
    {
      "entropy": 5.327292585372925,
      "epoch": 3.2210031347962382,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004266390438703486,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19039558470249177,
      "num_tokens": 7220285.0,
      "step": 4110
    },
    {
      "entropy": 5.435970211029053,
      "epoch": 3.2249216300940438,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042641669149050493,
      "loss": 5.0963,
      "mean_token_accuracy": 0.17844377160072328,
      "num_tokens": 7230007.0,
      "step": 4115
    },
    {
      "entropy": 5.362475204467773,
      "epoch": 3.2288401253918497,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004261940685044825,
      "loss": 4.9804,
      "mean_token_accuracy": 0.18439241796731948,
      "num_tokens": 7238520.0,
      "step": 4120
    },
    {
      "entropy": 5.280787563323974,
      "epoch": 3.2327586206896552,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042597117531014474,
      "loss": 4.9326,
      "mean_token_accuracy": 0.18893828988075256,
      "num_tokens": 7247228.0,
      "step": 4125
    },
    {
      "entropy": 5.438844394683838,
      "epoch": 3.2366771159874608,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004257480123058378,
      "loss": 5.1501,
      "mean_token_accuracy": 0.1765511766076088,
      "num_tokens": 7255782.0,
      "step": 4130
    },
    {
      "entropy": 5.429732608795166,
      "epoch": 3.2405956112852663,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042552457989039036,
      "loss": 5.0194,
      "mean_token_accuracy": 0.17550422847270966,
      "num_tokens": 7264946.0,
      "step": 4135
    },
    {
      "entropy": 5.237804317474366,
      "epoch": 3.2445141065830723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042530087846311213,
      "loss": 4.8506,
      "mean_token_accuracy": 0.19684191644191742,
      "num_tokens": 7273647.0,
      "step": 4140
    },
    {
      "entropy": 5.364433526992798,
      "epoch": 3.248432601880878,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042507690842379396,
      "loss": 5.0037,
      "mean_token_accuracy": 0.1836499258875847,
      "num_tokens": 7283393.0,
      "step": 4145
    },
    {
      "entropy": 5.427819633483887,
      "epoch": 3.2523510971786833,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042485267017270664,
      "loss": 5.0876,
      "mean_token_accuracy": 0.18043252676725388,
      "num_tokens": 7292368.0,
      "step": 4150
    },
    {
      "entropy": 5.331791877746582,
      "epoch": 3.256269592476489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042462816411060025,
      "loss": 5.0242,
      "mean_token_accuracy": 0.18351685255765915,
      "num_tokens": 7301505.0,
      "step": 4155
    },
    {
      "entropy": 5.339120292663575,
      "epoch": 3.260188087774295,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004244033906387035,
      "loss": 4.9754,
      "mean_token_accuracy": 0.19235653877258302,
      "num_tokens": 7310011.0,
      "step": 4160
    },
    {
      "entropy": 5.392613363265991,
      "epoch": 3.2641065830721003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004241783501587231,
      "loss": 5.0127,
      "mean_token_accuracy": 0.18839660435914993,
      "num_tokens": 7319295.0,
      "step": 4165
    },
    {
      "entropy": 5.373349857330322,
      "epoch": 3.268025078369906,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042395304307284284,
      "loss": 5.044,
      "mean_token_accuracy": 0.1815004900097847,
      "num_tokens": 7328390.0,
      "step": 4170
    },
    {
      "entropy": 5.418195056915283,
      "epoch": 3.271943573667712,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004237274697837229,
      "loss": 5.0737,
      "mean_token_accuracy": 0.17965858727693557,
      "num_tokens": 7337062.0,
      "step": 4175
    },
    {
      "entropy": 5.312235975265503,
      "epoch": 3.2758620689655173,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004235016306944996,
      "loss": 5.0527,
      "mean_token_accuracy": 0.18473271280527115,
      "num_tokens": 7346364.0,
      "step": 4180
    },
    {
      "entropy": 5.361359930038452,
      "epoch": 3.279780564263323,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004232755262087837,
      "loss": 4.9942,
      "mean_token_accuracy": 0.18626722395420076,
      "num_tokens": 7354896.0,
      "step": 4185
    },
    {
      "entropy": 5.358744192123413,
      "epoch": 3.2836990595611284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042304915673066083,
      "loss": 5.0346,
      "mean_token_accuracy": 0.19114716351032257,
      "num_tokens": 7363591.0,
      "step": 4190
    },
    {
      "entropy": 5.378021144866944,
      "epoch": 3.287617554858934,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042282252266468985,
      "loss": 5.0451,
      "mean_token_accuracy": 0.18212140798568727,
      "num_tokens": 7372891.0,
      "step": 4195
    },
    {
      "entropy": 5.341123962402344,
      "epoch": 3.29153605015674,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004225956244159025,
      "loss": 4.9557,
      "mean_token_accuracy": 0.18621090948581695,
      "num_tokens": 7381945.0,
      "step": 4200
    },
    {
      "entropy": 5.454095888137817,
      "epoch": 3.2954545454545454,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004223684623898029,
      "loss": 5.1132,
      "mean_token_accuracy": 0.18217166513204575,
      "num_tokens": 7391449.0,
      "step": 4205
    },
    {
      "entropy": 5.349137783050537,
      "epoch": 3.299373040752351,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004221410369923662,
      "loss": 4.953,
      "mean_token_accuracy": 0.1893253967165947,
      "num_tokens": 7399762.0,
      "step": 4210
    },
    {
      "entropy": 5.230040836334228,
      "epoch": 3.303291536050157,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042191334863003873,
      "loss": 4.8802,
      "mean_token_accuracy": 0.18941539376974106,
      "num_tokens": 7408711.0,
      "step": 4215
    },
    {
      "entropy": 5.243869590759277,
      "epoch": 3.3072100313479624,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004216853977097363,
      "loss": 4.9636,
      "mean_token_accuracy": 0.19784684032201766,
      "num_tokens": 7416868.0,
      "step": 4220
    },
    {
      "entropy": 5.399055147171021,
      "epoch": 3.311128526645768,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004214571846388442,
      "loss": 5.0738,
      "mean_token_accuracy": 0.17823962718248368,
      "num_tokens": 7425687.0,
      "step": 4225
    },
    {
      "entropy": 5.401248264312744,
      "epoch": 3.3150470219435735,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004212287098252164,
      "loss": 4.9941,
      "mean_token_accuracy": 0.18598046004772187,
      "num_tokens": 7433714.0,
      "step": 4230
    },
    {
      "entropy": 5.307060861587525,
      "epoch": 3.3189655172413794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004209999736771742,
      "loss": 5.0344,
      "mean_token_accuracy": 0.1938866674900055,
      "num_tokens": 7442765.0,
      "step": 4235
    },
    {
      "entropy": 5.356434059143067,
      "epoch": 3.322884012539185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004207709766035063,
      "loss": 5.021,
      "mean_token_accuracy": 0.1861105814576149,
      "num_tokens": 7450672.0,
      "step": 4240
    },
    {
      "entropy": 5.383820819854736,
      "epoch": 3.3268025078369905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004205417190134674,
      "loss": 5.0842,
      "mean_token_accuracy": 0.1816158562898636,
      "num_tokens": 7458758.0,
      "step": 4245
    },
    {
      "entropy": 5.369161891937256,
      "epoch": 3.330721003134796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004203122013167783,
      "loss": 5.0317,
      "mean_token_accuracy": 0.18005152344703673,
      "num_tokens": 7467527.0,
      "step": 4250
    },
    {
      "entropy": 5.365553665161133,
      "epoch": 3.334639498432602,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042008242392362413,
      "loss": 5.0285,
      "mean_token_accuracy": 0.1906298503279686,
      "num_tokens": 7476659.0,
      "step": 4255
    },
    {
      "entropy": 5.285857915878296,
      "epoch": 3.3385579937304075,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041985238724465433,
      "loss": 4.947,
      "mean_token_accuracy": 0.19531439542770385,
      "num_tokens": 7485813.0,
      "step": 4260
    },
    {
      "entropy": 5.396247816085816,
      "epoch": 3.342476489028213,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041962209169098193,
      "loss": 5.0666,
      "mean_token_accuracy": 0.18144663274288178,
      "num_tokens": 7494842.0,
      "step": 4265
    },
    {
      "entropy": 5.392434787750244,
      "epoch": 3.346394984326019,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004193915376741823,
      "loss": 5.0793,
      "mean_token_accuracy": 0.18012839704751968,
      "num_tokens": 7503633.0,
      "step": 4270
    },
    {
      "entropy": 5.323088550567627,
      "epoch": 3.3503134796238245,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004191607256062928,
      "loss": 4.9943,
      "mean_token_accuracy": 0.18584160059690474,
      "num_tokens": 7512759.0,
      "step": 4275
    },
    {
      "entropy": 5.379161262512207,
      "epoch": 3.35423197492163,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004189296558998121,
      "loss": 5.0984,
      "mean_token_accuracy": 0.18285783976316453,
      "num_tokens": 7522284.0,
      "step": 4280
    },
    {
      "entropy": 5.388297986984253,
      "epoch": 3.3581504702194356,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004186983289676992,
      "loss": 5.0035,
      "mean_token_accuracy": 0.18422911912202836,
      "num_tokens": 7531286.0,
      "step": 4285
    },
    {
      "entropy": 5.345287609100342,
      "epoch": 3.3620689655172415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041846674522337296,
      "loss": 5.0024,
      "mean_token_accuracy": 0.17747625410556794,
      "num_tokens": 7540450.0,
      "step": 4290
    },
    {
      "entropy": 5.304649782180786,
      "epoch": 3.365987460815047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041823490508071076,
      "loss": 4.9346,
      "mean_token_accuracy": 0.18764492645859718,
      "num_tokens": 7549386.0,
      "step": 4295
    },
    {
      "entropy": 5.3941480159759525,
      "epoch": 3.3699059561128526,
      "grad_norm": 1.15625,
      "learning_rate": 0.000418002808954049,
      "loss": 5.0333,
      "mean_token_accuracy": 0.18422217518091202,
      "num_tokens": 7558236.0,
      "step": 4300
    },
    {
      "entropy": 5.255111885070801,
      "epoch": 3.373824451410658,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041777045725818057,
      "loss": 4.901,
      "mean_token_accuracy": 0.18648242950439453,
      "num_tokens": 7566984.0,
      "step": 4305
    },
    {
      "entropy": 5.323056840896607,
      "epoch": 3.377742946708464,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000417537850408356,
      "loss": 5.0462,
      "mean_token_accuracy": 0.17651378512382507,
      "num_tokens": 7576018.0,
      "step": 4310
    },
    {
      "entropy": 5.373207521438599,
      "epoch": 3.3816614420062696,
      "grad_norm": 1.375,
      "learning_rate": 0.0004173049888202814,
      "loss": 5.0481,
      "mean_token_accuracy": 0.17836394011974335,
      "num_tokens": 7585294.0,
      "step": 4315
    },
    {
      "entropy": 5.314003896713257,
      "epoch": 3.385579937304075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170718729101179,
      "loss": 4.9762,
      "mean_token_accuracy": 0.18725571185350418,
      "num_tokens": 7594634.0,
      "step": 4320
    },
    {
      "entropy": 5.306338214874268,
      "epoch": 3.389498432601881,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041683850309448187,
      "loss": 5.0498,
      "mean_token_accuracy": 0.1815230906009674,
      "num_tokens": 7603153.0,
      "step": 4325
    },
    {
      "entropy": 5.390663003921508,
      "epoch": 3.3934169278996866,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041660487979044264,
      "loss": 4.983,
      "mean_token_accuracy": 0.18932773023843766,
      "num_tokens": 7612027.0,
      "step": 4330
    },
    {
      "entropy": 5.333336305618286,
      "epoch": 3.397335423197492,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004163710034155231,
      "loss": 5.0135,
      "mean_token_accuracy": 0.1788959339261055,
      "num_tokens": 7620725.0,
      "step": 4335
    },
    {
      "entropy": 5.293660402297974,
      "epoch": 3.4012539184952977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004161368743876982,
      "loss": 4.9828,
      "mean_token_accuracy": 0.19251447767019272,
      "num_tokens": 7628847.0,
      "step": 4340
    },
    {
      "entropy": 5.354874849319458,
      "epoch": 3.405172413793103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004159024931253945,
      "loss": 5.0473,
      "mean_token_accuracy": 0.18395871073007583,
      "num_tokens": 7637668.0,
      "step": 4345
    },
    {
      "entropy": 5.368787860870361,
      "epoch": 3.409090909090909,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041566786004748943,
      "loss": 5.0897,
      "mean_token_accuracy": 0.185278782248497,
      "num_tokens": 7646545.0,
      "step": 4350
    },
    {
      "entropy": 5.416274642944336,
      "epoch": 3.4130094043887147,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041543297557331015,
      "loss": 5.0759,
      "mean_token_accuracy": 0.17744662314653398,
      "num_tokens": 7654967.0,
      "step": 4355
    },
    {
      "entropy": 5.31085205078125,
      "epoch": 3.41692789968652,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004151978401226335,
      "loss": 5.001,
      "mean_token_accuracy": 0.1886196941137314,
      "num_tokens": 7663718.0,
      "step": 4360
    },
    {
      "entropy": 5.368908262252807,
      "epoch": 3.420846394984326,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041496245411568435,
      "loss": 5.0427,
      "mean_token_accuracy": 0.18625704795122147,
      "num_tokens": 7672430.0,
      "step": 4365
    },
    {
      "entropy": 5.350130033493042,
      "epoch": 3.4247648902821317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004147268179731359,
      "loss": 5.0152,
      "mean_token_accuracy": 0.18233199119567872,
      "num_tokens": 7681270.0,
      "step": 4370
    },
    {
      "entropy": 5.366523170471192,
      "epoch": 3.4286833855799372,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041449093211610815,
      "loss": 5.0291,
      "mean_token_accuracy": 0.18019224107265472,
      "num_tokens": 7690124.0,
      "step": 4375
    },
    {
      "entropy": 5.392818784713745,
      "epoch": 3.4326018808777428,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041425479696616734,
      "loss": 5.0693,
      "mean_token_accuracy": 0.1819964364171028,
      "num_tokens": 7699511.0,
      "step": 4380
    },
    {
      "entropy": 5.308698511123657,
      "epoch": 3.4365203761755487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004140184129453253,
      "loss": 4.9904,
      "mean_token_accuracy": 0.18379651010036469,
      "num_tokens": 7708796.0,
      "step": 4385
    },
    {
      "entropy": 5.363918590545654,
      "epoch": 3.4404388714733543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041378178047603845,
      "loss": 5.0079,
      "mean_token_accuracy": 0.18806022256612778,
      "num_tokens": 7717446.0,
      "step": 4390
    },
    {
      "entropy": 5.390868282318115,
      "epoch": 3.44435736677116,
      "grad_norm": 1.25,
      "learning_rate": 0.0004135448999812074,
      "loss": 5.035,
      "mean_token_accuracy": 0.18293258994817735,
      "num_tokens": 7726814.0,
      "step": 4395
    },
    {
      "entropy": 5.309965038299561,
      "epoch": 3.4482758620689653,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004133077718841763,
      "loss": 4.946,
      "mean_token_accuracy": 0.19565905928611754,
      "num_tokens": 7735895.0,
      "step": 4400
    },
    {
      "entropy": 5.327420949935913,
      "epoch": 3.4521943573667713,
      "grad_norm": 1.328125,
      "learning_rate": 0.00041307039660873113,
      "loss": 5.0176,
      "mean_token_accuracy": 0.1832549661397934,
      "num_tokens": 7744160.0,
      "step": 4405
    },
    {
      "entropy": 5.372987508773804,
      "epoch": 3.456112852664577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004128327745791002,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17300991266965865,
      "num_tokens": 7752844.0,
      "step": 4410
    },
    {
      "entropy": 5.336950254440308,
      "epoch": 3.4600313479623823,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004125949062199526,
      "loss": 5.0657,
      "mean_token_accuracy": 0.17855542600154878,
      "num_tokens": 7762327.0,
      "step": 4415
    },
    {
      "entropy": 5.325194787979126,
      "epoch": 3.4639498432601883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041235679195639766,
      "loss": 4.9922,
      "mean_token_accuracy": 0.19132798463106154,
      "num_tokens": 7770813.0,
      "step": 4420
    },
    {
      "entropy": 5.269763326644897,
      "epoch": 3.467868338557994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041211843221398406,
      "loss": 5.017,
      "mean_token_accuracy": 0.18555662482976915,
      "num_tokens": 7779357.0,
      "step": 4425
    },
    {
      "entropy": 5.4215740203857425,
      "epoch": 3.4717868338557993,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004118798274186994,
      "loss": 5.0511,
      "mean_token_accuracy": 0.17997359484434128,
      "num_tokens": 7787302.0,
      "step": 4430
    },
    {
      "entropy": 5.3214476108551025,
      "epoch": 3.475705329153605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004116409779969691,
      "loss": 5.0547,
      "mean_token_accuracy": 0.17838198989629744,
      "num_tokens": 7795651.0,
      "step": 4435
    },
    {
      "entropy": 5.281446552276611,
      "epoch": 3.479623824451411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041140188437565586,
      "loss": 5.0332,
      "mean_token_accuracy": 0.18731289356946945,
      "num_tokens": 7804644.0,
      "step": 4440
    },
    {
      "entropy": 5.373918962478638,
      "epoch": 3.4835423197492164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041116254698205873,
      "loss": 4.9998,
      "mean_token_accuracy": 0.19277618676424027,
      "num_tokens": 7812479.0,
      "step": 4445
    },
    {
      "entropy": 5.336103343963623,
      "epoch": 3.487460815047022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041092296624391244,
      "loss": 4.9882,
      "mean_token_accuracy": 0.1867508962750435,
      "num_tokens": 7820678.0,
      "step": 4450
    },
    {
      "entropy": 5.3215635299682615,
      "epoch": 3.4913793103448274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004106831425893865,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19161393642425537,
      "num_tokens": 7829063.0,
      "step": 4455
    },
    {
      "entropy": 5.394068336486816,
      "epoch": 3.4952978056426334,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004104430764470849,
      "loss": 5.0594,
      "mean_token_accuracy": 0.18754285722970962,
      "num_tokens": 7837274.0,
      "step": 4460
    },
    {
      "entropy": 5.382764053344727,
      "epoch": 3.499216300940439,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004102027682460445,
      "loss": 4.9707,
      "mean_token_accuracy": 0.19755763411521912,
      "num_tokens": 7845379.0,
      "step": 4465
    },
    {
      "entropy": 5.338766384124756,
      "epoch": 3.5031347962382444,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004099622184157353,
      "loss": 5.0197,
      "mean_token_accuracy": 0.18330055475234985,
      "num_tokens": 7853760.0,
      "step": 4470
    },
    {
      "entropy": 5.437317848205566,
      "epoch": 3.5070532915360504,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004097214273860586,
      "loss": 5.0674,
      "mean_token_accuracy": 0.17608548402786256,
      "num_tokens": 7862026.0,
      "step": 4475
    },
    {
      "entropy": 5.313975524902344,
      "epoch": 3.510971786833856,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004094803955873471,
      "loss": 5.025,
      "mean_token_accuracy": 0.19090429097414016,
      "num_tokens": 7870369.0,
      "step": 4480
    },
    {
      "entropy": 5.32178430557251,
      "epoch": 3.5148902821316614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004092391234503638,
      "loss": 5.031,
      "mean_token_accuracy": 0.1843138486146927,
      "num_tokens": 7878524.0,
      "step": 4485
    },
    {
      "entropy": 5.361889028549195,
      "epoch": 3.518808777429467,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040899761140630094,
      "loss": 5.0904,
      "mean_token_accuracy": 0.17608782351017,
      "num_tokens": 7887369.0,
      "step": 4490
    },
    {
      "entropy": 5.402906084060669,
      "epoch": 3.5227272727272725,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040875585988677985,
      "loss": 5.0635,
      "mean_token_accuracy": 0.17864606231451036,
      "num_tokens": 7896112.0,
      "step": 4495
    },
    {
      "entropy": 5.283060026168823,
      "epoch": 3.5266457680250785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004085138693238497,
      "loss": 4.9006,
      "mean_token_accuracy": 0.18907321691513063,
      "num_tokens": 7904580.0,
      "step": 4500
    },
    {
      "epoch": 3.5266457680250785,
      "eval_entropy": 5.228835704714753,
      "eval_loss": 5.749582767486572,
      "eval_mean_token_accuracy": 0.16210612278547168,
      "eval_num_tokens": 7904580.0,
      "eval_runtime": 2.8539,
      "eval_samples_per_second": 1444.325,
      "eval_steps_per_second": 180.803,
      "step": 4500
    },
    {
      "entropy": 5.384114694595337,
      "epoch": 3.530564263322884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004082716401499867,
      "loss": 5.0627,
      "mean_token_accuracy": 0.18273453116416932,
      "num_tokens": 7913755.0,
      "step": 4505
    },
    {
      "entropy": 5.338723373413086,
      "epoch": 3.5344827586206895,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040802917279809383,
      "loss": 5.0223,
      "mean_token_accuracy": 0.18780091851949693,
      "num_tokens": 7922385.0,
      "step": 4510
    },
    {
      "entropy": 5.318574047088623,
      "epoch": 3.5384012539184955,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040778646770149953,
      "loss": 5.0692,
      "mean_token_accuracy": 0.17643754780292512,
      "num_tokens": 7931634.0,
      "step": 4515
    },
    {
      "entropy": 5.299038887023926,
      "epoch": 3.542319749216301,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040754352529395716,
      "loss": 4.9721,
      "mean_token_accuracy": 0.18572138696908952,
      "num_tokens": 7939866.0,
      "step": 4520
    },
    {
      "entropy": 5.349331998825074,
      "epoch": 3.5462382445141065,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040730034600964415,
      "loss": 5.1,
      "mean_token_accuracy": 0.17437055855989456,
      "num_tokens": 7948961.0,
      "step": 4525
    },
    {
      "entropy": 5.211510515213012,
      "epoch": 3.5501567398119125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004070569302831613,
      "loss": 4.9489,
      "mean_token_accuracy": 0.18208392411470414,
      "num_tokens": 7958526.0,
      "step": 4530
    },
    {
      "entropy": 5.314155340194702,
      "epoch": 3.554075235109718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000406813278549532,
      "loss": 5.0447,
      "mean_token_accuracy": 0.18393439054489136,
      "num_tokens": 7966887.0,
      "step": 4535
    },
    {
      "entropy": 5.382942867279053,
      "epoch": 3.5579937304075235,
      "grad_norm": 1.359375,
      "learning_rate": 0.00040656939124420144,
      "loss": 5.0106,
      "mean_token_accuracy": 0.18369665443897248,
      "num_tokens": 7974929.0,
      "step": 4540
    },
    {
      "entropy": 5.324088096618652,
      "epoch": 3.561912225705329,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004063252688030358,
      "loss": 4.9947,
      "mean_token_accuracy": 0.18499946296215058,
      "num_tokens": 7983821.0,
      "step": 4545
    },
    {
      "entropy": 5.300717782974243,
      "epoch": 3.5658307210031346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004060809116623213,
      "loss": 4.999,
      "mean_token_accuracy": 0.18629878014326096,
      "num_tokens": 7992109.0,
      "step": 4550
    },
    {
      "entropy": 5.261916303634644,
      "epoch": 3.5697492163009406,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000405836320258764,
      "loss": 5.0049,
      "mean_token_accuracy": 0.19590796679258346,
      "num_tokens": 8001098.0,
      "step": 4555
    },
    {
      "entropy": 5.399741744995117,
      "epoch": 3.573667711598746,
      "grad_norm": 1.25,
      "learning_rate": 0.0004055914950294882,
      "loss": 5.022,
      "mean_token_accuracy": 0.1856943428516388,
      "num_tokens": 8009559.0,
      "step": 4560
    },
    {
      "entropy": 5.281548404693604,
      "epoch": 3.5775862068965516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040534643641203645,
      "loss": 4.9478,
      "mean_token_accuracy": 0.18791809678077698,
      "num_tokens": 8018021.0,
      "step": 4565
    },
    {
      "entropy": 5.28187108039856,
      "epoch": 3.5815047021943576,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004051011448443681,
      "loss": 4.9791,
      "mean_token_accuracy": 0.18905122578144073,
      "num_tokens": 8026258.0,
      "step": 4570
    },
    {
      "entropy": 5.337254667282105,
      "epoch": 3.585423197492163,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000404855620764859,
      "loss": 5.0344,
      "mean_token_accuracy": 0.1859464392066002,
      "num_tokens": 8036761.0,
      "step": 4575
    },
    {
      "entropy": 5.424194192886352,
      "epoch": 3.5893416927899686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004046098646123006,
      "loss": 5.0592,
      "mean_token_accuracy": 0.18633756786584854,
      "num_tokens": 8045115.0,
      "step": 4580
    },
    {
      "entropy": 5.411231136322021,
      "epoch": 3.593260188087774,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040436387682589876,
      "loss": 5.0653,
      "mean_token_accuracy": 0.1834258556365967,
      "num_tokens": 8054334.0,
      "step": 4585
    },
    {
      "entropy": 5.238807964324951,
      "epoch": 3.5971786833855797,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004041176578452737,
      "loss": 5.0359,
      "mean_token_accuracy": 0.18548041582107544,
      "num_tokens": 8063165.0,
      "step": 4590
    },
    {
      "entropy": 5.356834506988525,
      "epoch": 3.6010971786833856,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004038712081104587,
      "loss": 5.0179,
      "mean_token_accuracy": 0.18496917933225632,
      "num_tokens": 8072498.0,
      "step": 4595
    },
    {
      "entropy": 5.358961725234986,
      "epoch": 3.605015673981191,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040362452806189927,
      "loss": 5.0252,
      "mean_token_accuracy": 0.18584552109241487,
      "num_tokens": 8081455.0,
      "step": 4600
    },
    {
      "entropy": 5.255314350128174,
      "epoch": 3.6089341692789967,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004033776181404527,
      "loss": 5.0401,
      "mean_token_accuracy": 0.18378792256116866,
      "num_tokens": 8090992.0,
      "step": 4605
    },
    {
      "entropy": 5.355190086364746,
      "epoch": 3.6128526645768027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040313047878738704,
      "loss": 5.0457,
      "mean_token_accuracy": 0.1795653834939003,
      "num_tokens": 8099976.0,
      "step": 4610
    },
    {
      "entropy": 5.35163426399231,
      "epoch": 3.616771159874608,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004028831104443805,
      "loss": 4.9429,
      "mean_token_accuracy": 0.19904136657714844,
      "num_tokens": 8108275.0,
      "step": 4615
    },
    {
      "entropy": 5.267815685272216,
      "epoch": 3.6206896551724137,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004026355135535202,
      "loss": 5.0516,
      "mean_token_accuracy": 0.1834827572107315,
      "num_tokens": 8117441.0,
      "step": 4620
    },
    {
      "entropy": 5.311365699768066,
      "epoch": 3.6246081504702197,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040238768855730214,
      "loss": 4.9963,
      "mean_token_accuracy": 0.18370137959718705,
      "num_tokens": 8125709.0,
      "step": 4625
    },
    {
      "entropy": 5.356410884857178,
      "epoch": 3.628526645768025,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040213963589862963,
      "loss": 5.02,
      "mean_token_accuracy": 0.18664186745882033,
      "num_tokens": 8134418.0,
      "step": 4630
    },
    {
      "entropy": 5.356766986846924,
      "epoch": 3.6324451410658307,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004018913560208131,
      "loss": 5.0851,
      "mean_token_accuracy": 0.17501234263181686,
      "num_tokens": 8143729.0,
      "step": 4635
    },
    {
      "entropy": 5.271541166305542,
      "epoch": 3.6363636363636362,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004016428493675689,
      "loss": 4.9712,
      "mean_token_accuracy": 0.1830235615372658,
      "num_tokens": 8152613.0,
      "step": 4640
    },
    {
      "entropy": 5.281400394439697,
      "epoch": 3.6402821316614418,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004013941163830187,
      "loss": 4.9608,
      "mean_token_accuracy": 0.1870640829205513,
      "num_tokens": 8160862.0,
      "step": 4645
    },
    {
      "entropy": 5.354451370239258,
      "epoch": 3.6442006269592477,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004011451575116887,
      "loss": 5.0376,
      "mean_token_accuracy": 0.1867995724081993,
      "num_tokens": 8169369.0,
      "step": 4650
    },
    {
      "entropy": 5.278015661239624,
      "epoch": 3.6481191222570533,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004008959731985087,
      "loss": 4.9523,
      "mean_token_accuracy": 0.1851906254887581,
      "num_tokens": 8178152.0,
      "step": 4655
    },
    {
      "entropy": 5.2923859596252445,
      "epoch": 3.652037617554859,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040064656388881157,
      "loss": 4.9303,
      "mean_token_accuracy": 0.18778676986694337,
      "num_tokens": 8186690.0,
      "step": 4660
    },
    {
      "entropy": 5.26329460144043,
      "epoch": 3.6559561128526648,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004003969300283321,
      "loss": 4.9206,
      "mean_token_accuracy": 0.19149983078241348,
      "num_tokens": 8194976.0,
      "step": 4665
    },
    {
      "entropy": 5.275133895874023,
      "epoch": 3.6598746081504703,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040014707206320653,
      "loss": 5.0075,
      "mean_token_accuracy": 0.18456661105155944,
      "num_tokens": 8202964.0,
      "step": 4670
    },
    {
      "entropy": 5.301375865936279,
      "epoch": 3.663793103448276,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039989699043997153,
      "loss": 5.0235,
      "mean_token_accuracy": 0.18406348526477814,
      "num_tokens": 8213084.0,
      "step": 4675
    },
    {
      "entropy": 5.3162261009216305,
      "epoch": 3.6677115987460818,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039964668560556356,
      "loss": 4.9407,
      "mean_token_accuracy": 0.19182607233524324,
      "num_tokens": 8221645.0,
      "step": 4680
    },
    {
      "entropy": 5.257916784286499,
      "epoch": 3.6716300940438873,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039939615800731784,
      "loss": 4.9322,
      "mean_token_accuracy": 0.19391874819993973,
      "num_tokens": 8230195.0,
      "step": 4685
    },
    {
      "entropy": 5.286824893951416,
      "epoch": 3.675548589341693,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039914540809296795,
      "loss": 5.0114,
      "mean_token_accuracy": 0.1874181106686592,
      "num_tokens": 8239608.0,
      "step": 4690
    },
    {
      "entropy": 5.3544902324676515,
      "epoch": 3.6794670846394983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003988944363106445,
      "loss": 5.0598,
      "mean_token_accuracy": 0.18949063420295714,
      "num_tokens": 8247993.0,
      "step": 4695
    },
    {
      "entropy": 5.293319463729858,
      "epoch": 3.683385579937304,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003986432431088749,
      "loss": 5.0272,
      "mean_token_accuracy": 0.18091020584106446,
      "num_tokens": 8257279.0,
      "step": 4700
    },
    {
      "entropy": 5.4265107154846195,
      "epoch": 3.68730407523511,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000398391828936582,
      "loss": 5.0764,
      "mean_token_accuracy": 0.18089805245399476,
      "num_tokens": 8265166.0,
      "step": 4705
    },
    {
      "entropy": 5.267129850387573,
      "epoch": 3.6912225705329154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003981401942430838,
      "loss": 4.8689,
      "mean_token_accuracy": 0.19139912277460097,
      "num_tokens": 8273281.0,
      "step": 4710
    },
    {
      "entropy": 5.258962774276734,
      "epoch": 3.695141065830721,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039788833947809217,
      "loss": 5.0562,
      "mean_token_accuracy": 0.17835760861635208,
      "num_tokens": 8282432.0,
      "step": 4715
    },
    {
      "entropy": 5.251905632019043,
      "epoch": 3.699059561128527,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003976362650917125,
      "loss": 4.8996,
      "mean_token_accuracy": 0.19927400946617127,
      "num_tokens": 8290909.0,
      "step": 4720
    },
    {
      "entropy": 5.2755279541015625,
      "epoch": 3.7029780564263324,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039738397153444264,
      "loss": 5.0121,
      "mean_token_accuracy": 0.1859713986515999,
      "num_tokens": 8299581.0,
      "step": 4725
    },
    {
      "entropy": 5.308114051818848,
      "epoch": 3.706896551724138,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003971314592571719,
      "loss": 4.9996,
      "mean_token_accuracy": 0.18658531606197357,
      "num_tokens": 8308184.0,
      "step": 4730
    },
    {
      "entropy": 5.305037021636963,
      "epoch": 3.7108150470219434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003968787287111809,
      "loss": 5.0161,
      "mean_token_accuracy": 0.18689171373844146,
      "num_tokens": 8317054.0,
      "step": 4735
    },
    {
      "entropy": 5.265456056594848,
      "epoch": 3.714733542319749,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039662578034814,
      "loss": 5.0258,
      "mean_token_accuracy": 0.19158447980880738,
      "num_tokens": 8325961.0,
      "step": 4740
    },
    {
      "entropy": 5.440699625015259,
      "epoch": 3.718652037617555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039637261462010886,
      "loss": 5.1138,
      "mean_token_accuracy": 0.1796484261751175,
      "num_tokens": 8334401.0,
      "step": 4745
    },
    {
      "entropy": 5.388299179077149,
      "epoch": 3.7225705329153604,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003961192319795358,
      "loss": 5.095,
      "mean_token_accuracy": 0.18220945447683334,
      "num_tokens": 8343786.0,
      "step": 4750
    },
    {
      "entropy": 5.391376781463623,
      "epoch": 3.726489028213166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003958656328792565,
      "loss": 5.0656,
      "mean_token_accuracy": 0.18659788370132446,
      "num_tokens": 8352286.0,
      "step": 4755
    },
    {
      "entropy": 5.265856456756592,
      "epoch": 3.730407523510972,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039561181777249396,
      "loss": 4.9897,
      "mean_token_accuracy": 0.18892880976200105,
      "num_tokens": 8361274.0,
      "step": 4760
    },
    {
      "entropy": 5.304648303985596,
      "epoch": 3.7343260188087775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039535778711285676,
      "loss": 5.0261,
      "mean_token_accuracy": 0.19323430955410004,
      "num_tokens": 8369881.0,
      "step": 4765
    },
    {
      "entropy": 5.3594770431518555,
      "epoch": 3.738244514106583,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003951035413543388,
      "loss": 5.0918,
      "mean_token_accuracy": 0.16902616918087005,
      "num_tokens": 8378234.0,
      "step": 4770
    },
    {
      "entropy": 5.307481479644776,
      "epoch": 3.742163009404389,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039484908095131874,
      "loss": 5.011,
      "mean_token_accuracy": 0.1887112945318222,
      "num_tokens": 8387023.0,
      "step": 4775
    },
    {
      "entropy": 5.289135408401489,
      "epoch": 3.7460815047021945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003945944063585582,
      "loss": 4.9664,
      "mean_token_accuracy": 0.19265512824058534,
      "num_tokens": 8396110.0,
      "step": 4780
    },
    {
      "entropy": 5.280309963226318,
      "epoch": 3.75,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039433951803120225,
      "loss": 4.9374,
      "mean_token_accuracy": 0.19575012922286988,
      "num_tokens": 8405219.0,
      "step": 4785
    },
    {
      "entropy": 5.378737735748291,
      "epoch": 3.7539184952978055,
      "grad_norm": 1.25,
      "learning_rate": 0.00039408441642477764,
      "loss": 5.1451,
      "mean_token_accuracy": 0.17499377131462096,
      "num_tokens": 8413996.0,
      "step": 4790
    },
    {
      "entropy": 5.305036115646362,
      "epoch": 3.757836990595611,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003938291019951922,
      "loss": 4.8992,
      "mean_token_accuracy": 0.19813646078109742,
      "num_tokens": 8421931.0,
      "step": 4795
    },
    {
      "entropy": 5.302833604812622,
      "epoch": 3.761755485893417,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003935735751987344,
      "loss": 4.9854,
      "mean_token_accuracy": 0.17969918251037598,
      "num_tokens": 8430090.0,
      "step": 4800
    },
    {
      "entropy": 5.185124731063842,
      "epoch": 3.7656739811912225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039331783649207175,
      "loss": 4.9556,
      "mean_token_accuracy": 0.18346347361803056,
      "num_tokens": 8439311.0,
      "step": 4805
    },
    {
      "entropy": 5.306969928741455,
      "epoch": 3.769592476489028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039306188633225097,
      "loss": 5.0245,
      "mean_token_accuracy": 0.1876319020986557,
      "num_tokens": 8448952.0,
      "step": 4810
    },
    {
      "entropy": 5.280072450637817,
      "epoch": 3.773510971786834,
      "grad_norm": 1.25,
      "learning_rate": 0.0003928057251766965,
      "loss": 4.9631,
      "mean_token_accuracy": 0.18865569829940795,
      "num_tokens": 8457177.0,
      "step": 4815
    },
    {
      "entropy": 5.24211893081665,
      "epoch": 3.7774294670846396,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039254935348320984,
      "loss": 5.0109,
      "mean_token_accuracy": 0.18468278497457505,
      "num_tokens": 8465844.0,
      "step": 4820
    },
    {
      "entropy": 5.388170576095581,
      "epoch": 3.781347962382445,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039229277170996885,
      "loss": 5.1206,
      "mean_token_accuracy": 0.17690870612859727,
      "num_tokens": 8475198.0,
      "step": 4825
    },
    {
      "entropy": 5.34723858833313,
      "epoch": 3.785266457680251,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003920359803155266,
      "loss": 4.9844,
      "mean_token_accuracy": 0.19041041880846024,
      "num_tokens": 8484050.0,
      "step": 4830
    },
    {
      "entropy": 5.243993282318115,
      "epoch": 3.7891849529780566,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039177897975881115,
      "loss": 4.9888,
      "mean_token_accuracy": 0.18923878520727158,
      "num_tokens": 8491825.0,
      "step": 4835
    },
    {
      "entropy": 5.263262557983398,
      "epoch": 3.793103448275862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003915217704991239,
      "loss": 4.9788,
      "mean_token_accuracy": 0.19069561809301377,
      "num_tokens": 8500651.0,
      "step": 4840
    },
    {
      "entropy": 5.335864734649658,
      "epoch": 3.7970219435736676,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003912643529961397,
      "loss": 4.9983,
      "mean_token_accuracy": 0.18499371260404587,
      "num_tokens": 8508464.0,
      "step": 4845
    },
    {
      "entropy": 5.328866481781006,
      "epoch": 3.800940438871473,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003910067277099053,
      "loss": 5.0142,
      "mean_token_accuracy": 0.18994322270154954,
      "num_tokens": 8517676.0,
      "step": 4850
    },
    {
      "entropy": 5.337847423553467,
      "epoch": 3.804858934169279,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039074889510083894,
      "loss": 5.0526,
      "mean_token_accuracy": 0.18101195693016053,
      "num_tokens": 8526926.0,
      "step": 4855
    },
    {
      "entropy": 5.266471672058105,
      "epoch": 3.8087774294670846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003904908556297293,
      "loss": 4.9654,
      "mean_token_accuracy": 0.1841298222541809,
      "num_tokens": 8535814.0,
      "step": 4860
    },
    {
      "entropy": 5.364834880828857,
      "epoch": 3.81269592476489,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003902326097577345,
      "loss": 5.0565,
      "mean_token_accuracy": 0.18521469533443452,
      "num_tokens": 8544814.0,
      "step": 4865
    },
    {
      "entropy": 5.3280677318573,
      "epoch": 3.816614420062696,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038997415794638206,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18494193702936174,
      "num_tokens": 8553455.0,
      "step": 4870
    },
    {
      "entropy": 5.228104209899902,
      "epoch": 3.8205329153605017,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003897155006575672,
      "loss": 4.9671,
      "mean_token_accuracy": 0.18487488180398942,
      "num_tokens": 8562048.0,
      "step": 4875
    },
    {
      "entropy": 5.3047059059143065,
      "epoch": 3.824451410658307,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038945663835355247,
      "loss": 5.0127,
      "mean_token_accuracy": 0.18352398425340652,
      "num_tokens": 8571855.0,
      "step": 4880
    },
    {
      "entropy": 5.287893390655517,
      "epoch": 3.8283699059561127,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038919757149696665,
      "loss": 5.0212,
      "mean_token_accuracy": 0.18140169978141785,
      "num_tokens": 8580425.0,
      "step": 4885
    },
    {
      "entropy": 5.302046537399292,
      "epoch": 3.8322884012539182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038893830055080437,
      "loss": 5.0448,
      "mean_token_accuracy": 0.19019385129213334,
      "num_tokens": 8589751.0,
      "step": 4890
    },
    {
      "entropy": 5.333144092559815,
      "epoch": 3.836206896551724,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003886788259784248,
      "loss": 5.0579,
      "mean_token_accuracy": 0.18122659623622894,
      "num_tokens": 8598077.0,
      "step": 4895
    },
    {
      "entropy": 5.408076477050781,
      "epoch": 3.8401253918495297,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038841914824355093,
      "loss": 5.0662,
      "mean_token_accuracy": 0.18553077727556228,
      "num_tokens": 8606691.0,
      "step": 4900
    },
    {
      "entropy": 5.293684339523315,
      "epoch": 3.8440438871473352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038815926781026914,
      "loss": 5.0269,
      "mean_token_accuracy": 0.18251772820949555,
      "num_tokens": 8615477.0,
      "step": 4905
    },
    {
      "entropy": 5.26659140586853,
      "epoch": 3.847962382445141,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003878991851430279,
      "loss": 5.0267,
      "mean_token_accuracy": 0.18476115316152572,
      "num_tokens": 8623982.0,
      "step": 4910
    },
    {
      "entropy": 5.2914632797241214,
      "epoch": 3.8518808777429467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003876389007066371,
      "loss": 5.0529,
      "mean_token_accuracy": 0.18396950215101243,
      "num_tokens": 8633017.0,
      "step": 4915
    },
    {
      "entropy": 5.290062284469604,
      "epoch": 3.8557993730407523,
      "grad_norm": 1.25,
      "learning_rate": 0.0003873784149662672,
      "loss": 5.0024,
      "mean_token_accuracy": 0.18337597399950029,
      "num_tokens": 8642299.0,
      "step": 4920
    },
    {
      "entropy": 5.270841360092163,
      "epoch": 3.8597178683385582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003871177283874484,
      "loss": 4.9368,
      "mean_token_accuracy": 0.18476674407720567,
      "num_tokens": 8651008.0,
      "step": 4925
    },
    {
      "entropy": 5.266819667816162,
      "epoch": 3.8636363636363638,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038685684143606995,
      "loss": 5.0049,
      "mean_token_accuracy": 0.19042036384344102,
      "num_tokens": 8660103.0,
      "step": 4930
    },
    {
      "entropy": 5.193347978591919,
      "epoch": 3.8675548589341693,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003865957545783791,
      "loss": 4.9791,
      "mean_token_accuracy": 0.19078926146030425,
      "num_tokens": 8668646.0,
      "step": 4935
    },
    {
      "entropy": 5.288641929626465,
      "epoch": 3.871473354231975,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038633446828098046,
      "loss": 5.1232,
      "mean_token_accuracy": 0.18016830682754517,
      "num_tokens": 8678190.0,
      "step": 4940
    },
    {
      "entropy": 5.433268404006958,
      "epoch": 3.8753918495297803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000386072983010835,
      "loss": 5.1177,
      "mean_token_accuracy": 0.1755234643816948,
      "num_tokens": 8687336.0,
      "step": 4945
    },
    {
      "entropy": 5.333017349243164,
      "epoch": 3.8793103448275863,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038581129923525914,
      "loss": 5.028,
      "mean_token_accuracy": 0.1871548280119896,
      "num_tokens": 8695939.0,
      "step": 4950
    },
    {
      "entropy": 5.287285900115966,
      "epoch": 3.883228840125392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038554941742192445,
      "loss": 5.0549,
      "mean_token_accuracy": 0.17755318135023118,
      "num_tokens": 8704828.0,
      "step": 4955
    },
    {
      "entropy": 5.339581251144409,
      "epoch": 3.8871473354231973,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003852873380388561,
      "loss": 4.9705,
      "mean_token_accuracy": 0.194117371737957,
      "num_tokens": 8714427.0,
      "step": 4960
    },
    {
      "entropy": 5.321541786193848,
      "epoch": 3.8910658307210033,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003850250615544323,
      "loss": 5.0647,
      "mean_token_accuracy": 0.18604655265808107,
      "num_tokens": 8723737.0,
      "step": 4965
    },
    {
      "entropy": 5.23556694984436,
      "epoch": 3.894984326018809,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038476258843738386,
      "loss": 5.0186,
      "mean_token_accuracy": 0.18681242316961288,
      "num_tokens": 8733149.0,
      "step": 4970
    },
    {
      "entropy": 5.269054985046386,
      "epoch": 3.8989028213166144,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038449991915679273,
      "loss": 4.9855,
      "mean_token_accuracy": 0.18774406164884566,
      "num_tokens": 8742000.0,
      "step": 4975
    },
    {
      "entropy": 5.308100605010987,
      "epoch": 3.9028213166144203,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003842370541820915,
      "loss": 5.0393,
      "mean_token_accuracy": 0.1800890624523163,
      "num_tokens": 8751069.0,
      "step": 4980
    },
    {
      "entropy": 5.319173431396484,
      "epoch": 3.906739811912226,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038397399398306243,
      "loss": 4.9738,
      "mean_token_accuracy": 0.18709785789251326,
      "num_tokens": 8760092.0,
      "step": 4985
    },
    {
      "entropy": 5.307820558547974,
      "epoch": 3.9106583072100314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038371073902983684,
      "loss": 4.9874,
      "mean_token_accuracy": 0.1862591788172722,
      "num_tokens": 8768039.0,
      "step": 4990
    },
    {
      "entropy": 5.314032983779907,
      "epoch": 3.914576802507837,
      "grad_norm": 1.328125,
      "learning_rate": 0.000383447289792894,
      "loss": 5.0235,
      "mean_token_accuracy": 0.18012913167476655,
      "num_tokens": 8776630.0,
      "step": 4995
    },
    {
      "entropy": 5.309368371963501,
      "epoch": 3.9184952978056424,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038318364674306036,
      "loss": 4.9807,
      "mean_token_accuracy": 0.18810795694589616,
      "num_tokens": 8785272.0,
      "step": 5000
    },
    {
      "epoch": 3.9184952978056424,
      "eval_entropy": 5.218062679896983,
      "eval_loss": 5.704412460327148,
      "eval_mean_token_accuracy": 0.16455968886671601,
      "eval_num_tokens": 8785272.0,
      "eval_runtime": 2.8524,
      "eval_samples_per_second": 1445.104,
      "eval_steps_per_second": 180.901,
      "step": 5000
    },
    {
      "entropy": 5.3158056259155275,
      "epoch": 3.9224137931034484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038291981035150883,
      "loss": 5.0094,
      "mean_token_accuracy": 0.1924430251121521,
      "num_tokens": 8794197.0,
      "step": 5005
    },
    {
      "entropy": 5.288881635665893,
      "epoch": 3.926332288401254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003826557810897579,
      "loss": 5.0374,
      "mean_token_accuracy": 0.1802823171019554,
      "num_tokens": 8803195.0,
      "step": 5010
    },
    {
      "entropy": 5.298296117782593,
      "epoch": 3.9302507836990594,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003823915594296705,
      "loss": 4.9977,
      "mean_token_accuracy": 0.18835317045450212,
      "num_tokens": 8811772.0,
      "step": 5015
    },
    {
      "entropy": 5.178708171844482,
      "epoch": 3.9341692789968654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003821271458434538,
      "loss": 4.9667,
      "mean_token_accuracy": 0.18503689020872116,
      "num_tokens": 8820482.0,
      "step": 5020
    },
    {
      "entropy": 5.261278533935547,
      "epoch": 3.938087774294671,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038186254080365756,
      "loss": 5.0487,
      "mean_token_accuracy": 0.18290771842002868,
      "num_tokens": 8828761.0,
      "step": 5025
    },
    {
      "entropy": 5.270784854888916,
      "epoch": 3.9420062695924765,
      "grad_norm": 1.34375,
      "learning_rate": 0.000381597744783174,
      "loss": 4.9266,
      "mean_token_accuracy": 0.18722235560417175,
      "num_tokens": 8837165.0,
      "step": 5030
    },
    {
      "entropy": 5.279798269271851,
      "epoch": 3.945924764890282,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038133275825523645,
      "loss": 5.0086,
      "mean_token_accuracy": 0.18682824373245238,
      "num_tokens": 8845472.0,
      "step": 5035
    },
    {
      "entropy": 5.291768026351929,
      "epoch": 3.9498432601880875,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038106758169341887,
      "loss": 5.0039,
      "mean_token_accuracy": 0.18931640833616256,
      "num_tokens": 8855246.0,
      "step": 5040
    },
    {
      "entropy": 5.399905824661255,
      "epoch": 3.9537617554858935,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003808022155716348,
      "loss": 5.0767,
      "mean_token_accuracy": 0.1806002140045166,
      "num_tokens": 8863667.0,
      "step": 5045
    },
    {
      "entropy": 5.254299020767212,
      "epoch": 3.957680250783699,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003805366603641364,
      "loss": 4.9119,
      "mean_token_accuracy": 0.19546104371547698,
      "num_tokens": 8872306.0,
      "step": 5050
    },
    {
      "entropy": 5.214466190338134,
      "epoch": 3.9615987460815045,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038027091654551406,
      "loss": 4.9485,
      "mean_token_accuracy": 0.19039267152547837,
      "num_tokens": 8882037.0,
      "step": 5055
    },
    {
      "entropy": 5.235970211029053,
      "epoch": 3.9655172413793105,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038000498459069487,
      "loss": 4.996,
      "mean_token_accuracy": 0.18853940814733505,
      "num_tokens": 8890685.0,
      "step": 5060
    },
    {
      "entropy": 5.250956630706787,
      "epoch": 3.969435736677116,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003797388649749426,
      "loss": 4.9843,
      "mean_token_accuracy": 0.1859780803322792,
      "num_tokens": 8899927.0,
      "step": 5065
    },
    {
      "entropy": 5.266615724563598,
      "epoch": 3.9733542319749215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003794725581738559,
      "loss": 5.0175,
      "mean_token_accuracy": 0.1852913275361061,
      "num_tokens": 8908571.0,
      "step": 5070
    },
    {
      "entropy": 5.24801230430603,
      "epoch": 3.9772727272727275,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037920606466336834,
      "loss": 4.9291,
      "mean_token_accuracy": 0.18991698026657106,
      "num_tokens": 8917246.0,
      "step": 5075
    },
    {
      "entropy": 5.293500089645386,
      "epoch": 3.981191222570533,
      "grad_norm": 1.203125,
      "learning_rate": 0.000378939384919747,
      "loss": 5.0368,
      "mean_token_accuracy": 0.18419442623853682,
      "num_tokens": 8926116.0,
      "step": 5080
    },
    {
      "entropy": 5.346260643005371,
      "epoch": 3.9851097178683386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003786725194195918,
      "loss": 5.082,
      "mean_token_accuracy": 0.17977986633777618,
      "num_tokens": 8934914.0,
      "step": 5085
    },
    {
      "entropy": 5.323288011550903,
      "epoch": 3.989028213166144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037840546863983484,
      "loss": 4.9771,
      "mean_token_accuracy": 0.18484600484371186,
      "num_tokens": 8943321.0,
      "step": 5090
    },
    {
      "entropy": 5.267271327972412,
      "epoch": 3.9929467084639496,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037813823305773883,
      "loss": 5.0336,
      "mean_token_accuracy": 0.18731051832437515,
      "num_tokens": 8951848.0,
      "step": 5095
    },
    {
      "entropy": 5.315811014175415,
      "epoch": 3.9968652037617556,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003778708131508974,
      "loss": 5.0166,
      "mean_token_accuracy": 0.18719226121902466,
      "num_tokens": 8961291.0,
      "step": 5100
    },
    {
      "entropy": 5.36730809211731,
      "epoch": 4.0007836990595615,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037760320939723307,
      "loss": 5.0458,
      "mean_token_accuracy": 0.18242469280958176,
      "num_tokens": 8971116.0,
      "step": 5105
    },
    {
      "entropy": 5.285284614562988,
      "epoch": 4.004702194357367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037733542227499727,
      "loss": 4.6482,
      "mean_token_accuracy": 0.21242034286260605,
      "num_tokens": 8979362.0,
      "step": 5110
    },
    {
      "entropy": 5.34524564743042,
      "epoch": 4.008620689655173,
      "grad_norm": 1.375,
      "learning_rate": 0.00037706745226276893,
      "loss": 4.8333,
      "mean_token_accuracy": 0.19892482161521913,
      "num_tokens": 8987728.0,
      "step": 5115
    },
    {
      "entropy": 5.274005556106568,
      "epoch": 4.012539184952978,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003767992998394539,
      "loss": 4.6674,
      "mean_token_accuracy": 0.20412351489067077,
      "num_tokens": 8995817.0,
      "step": 5120
    },
    {
      "entropy": 5.274462366104126,
      "epoch": 4.016457680250784,
      "grad_norm": 1.125,
      "learning_rate": 0.0003765309654842842,
      "loss": 4.7506,
      "mean_token_accuracy": 0.19884773343801498,
      "num_tokens": 9005683.0,
      "step": 5125
    },
    {
      "entropy": 5.1921546459198,
      "epoch": 4.02037617554859,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003762624496768166,
      "loss": 4.6496,
      "mean_token_accuracy": 0.21461020559072494,
      "num_tokens": 9015408.0,
      "step": 5130
    },
    {
      "entropy": 5.350738048553467,
      "epoch": 4.024294670846395,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003759937528969325,
      "loss": 4.7921,
      "mean_token_accuracy": 0.19929660111665726,
      "num_tokens": 9024381.0,
      "step": 5135
    },
    {
      "entropy": 5.259899091720581,
      "epoch": 4.028213166144201,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037572487562483666,
      "loss": 4.6883,
      "mean_token_accuracy": 0.21029362231492996,
      "num_tokens": 9033092.0,
      "step": 5140
    },
    {
      "entropy": 5.20418152809143,
      "epoch": 4.032131661442007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037545581834105623,
      "loss": 4.8226,
      "mean_token_accuracy": 0.20301758050918578,
      "num_tokens": 9042251.0,
      "step": 5145
    },
    {
      "entropy": 5.229412698745728,
      "epoch": 4.036050156739812,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037518658152644035,
      "loss": 4.6856,
      "mean_token_accuracy": 0.20453651398420333,
      "num_tokens": 9050628.0,
      "step": 5150
    },
    {
      "entropy": 5.272675085067749,
      "epoch": 4.039968652037618,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003749171656621588,
      "loss": 4.7174,
      "mean_token_accuracy": 0.21974915862083436,
      "num_tokens": 9060080.0,
      "step": 5155
    },
    {
      "entropy": 5.265033721923828,
      "epoch": 4.043887147335423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003746475712297014,
      "loss": 4.7206,
      "mean_token_accuracy": 0.2133020281791687,
      "num_tokens": 9069456.0,
      "step": 5160
    },
    {
      "entropy": 5.2450767993927006,
      "epoch": 4.047805642633229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003743777987108771,
      "loss": 4.7406,
      "mean_token_accuracy": 0.21106670647859574,
      "num_tokens": 9078790.0,
      "step": 5165
    },
    {
      "entropy": 5.164502239227295,
      "epoch": 4.051724137931035,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003741078485878132,
      "loss": 4.7427,
      "mean_token_accuracy": 0.2065027117729187,
      "num_tokens": 9086888.0,
      "step": 5170
    },
    {
      "entropy": 5.25476450920105,
      "epoch": 4.05564263322884,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003738377213429542,
      "loss": 4.7755,
      "mean_token_accuracy": 0.19382651150226593,
      "num_tokens": 9095139.0,
      "step": 5175
    },
    {
      "entropy": 5.267132139205932,
      "epoch": 4.059561128526646,
      "grad_norm": 1.25,
      "learning_rate": 0.0003735674174590614,
      "loss": 4.6822,
      "mean_token_accuracy": 0.20767621994018554,
      "num_tokens": 9103464.0,
      "step": 5180
    },
    {
      "entropy": 5.263294839859009,
      "epoch": 4.063479623824452,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037329693741921166,
      "loss": 4.7007,
      "mean_token_accuracy": 0.20521023571491243,
      "num_tokens": 9111290.0,
      "step": 5185
    },
    {
      "entropy": 5.112252950668335,
      "epoch": 4.067398119122257,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003730262817067967,
      "loss": 4.58,
      "mean_token_accuracy": 0.2119445323944092,
      "num_tokens": 9120224.0,
      "step": 5190
    },
    {
      "entropy": 5.209140205383301,
      "epoch": 4.071316614420063,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003727554508055221,
      "loss": 4.7309,
      "mean_token_accuracy": 0.2061513438820839,
      "num_tokens": 9128524.0,
      "step": 5195
    },
    {
      "entropy": 5.199201393127441,
      "epoch": 4.075235109717869,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003724844451994066,
      "loss": 4.6164,
      "mean_token_accuracy": 0.21138493120670318,
      "num_tokens": 9137026.0,
      "step": 5200
    },
    {
      "entropy": 5.222887706756592,
      "epoch": 4.079153605015674,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037221326537278113,
      "loss": 4.6883,
      "mean_token_accuracy": 0.20704406648874282,
      "num_tokens": 9146312.0,
      "step": 5205
    },
    {
      "entropy": 5.211150360107422,
      "epoch": 4.08307210031348,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037194191181028806,
      "loss": 4.7228,
      "mean_token_accuracy": 0.201760958135128,
      "num_tokens": 9155226.0,
      "step": 5210
    },
    {
      "entropy": 5.316205310821533,
      "epoch": 4.086990595611285,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037167038499688024,
      "loss": 4.8575,
      "mean_token_accuracy": 0.19546173363924027,
      "num_tokens": 9164090.0,
      "step": 5215
    },
    {
      "entropy": 5.264502143859863,
      "epoch": 4.090909090909091,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037139868541782,
      "loss": 4.6914,
      "mean_token_accuracy": 0.2117288663983345,
      "num_tokens": 9173002.0,
      "step": 5220
    },
    {
      "entropy": 5.279915857315063,
      "epoch": 4.094827586206897,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003711268135586787,
      "loss": 4.7079,
      "mean_token_accuracy": 0.20757062137126922,
      "num_tokens": 9182020.0,
      "step": 5225
    },
    {
      "entropy": 5.230730485916138,
      "epoch": 4.098746081504702,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037085476990533523,
      "loss": 4.7289,
      "mean_token_accuracy": 0.20414845943450927,
      "num_tokens": 9190848.0,
      "step": 5230
    },
    {
      "entropy": 5.155168628692627,
      "epoch": 4.102664576802508,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003705825549439759,
      "loss": 4.664,
      "mean_token_accuracy": 0.2116340771317482,
      "num_tokens": 9198808.0,
      "step": 5235
    },
    {
      "entropy": 5.283816719055176,
      "epoch": 4.106583072100314,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037031016916109295,
      "loss": 4.7879,
      "mean_token_accuracy": 0.20280979871749877,
      "num_tokens": 9207923.0,
      "step": 5240
    },
    {
      "entropy": 5.254063081741333,
      "epoch": 4.110501567398119,
      "grad_norm": 1.25,
      "learning_rate": 0.00037003761304348374,
      "loss": 4.7419,
      "mean_token_accuracy": 0.2046460911631584,
      "num_tokens": 9216759.0,
      "step": 5245
    },
    {
      "entropy": 5.1899800300598145,
      "epoch": 4.114420062695925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036976488707825056,
      "loss": 4.682,
      "mean_token_accuracy": 0.1992353618144989,
      "num_tokens": 9225951.0,
      "step": 5250
    },
    {
      "entropy": 5.178423500061035,
      "epoch": 4.118338557993731,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036949199175279876,
      "loss": 4.7521,
      "mean_token_accuracy": 0.2033326268196106,
      "num_tokens": 9234413.0,
      "step": 5255
    },
    {
      "entropy": 5.267231225967407,
      "epoch": 4.122257053291536,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003692189275548364,
      "loss": 4.8823,
      "mean_token_accuracy": 0.19916021674871445,
      "num_tokens": 9244002.0,
      "step": 5260
    },
    {
      "entropy": 5.310188817977905,
      "epoch": 4.126175548589342,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003689456949723737,
      "loss": 4.712,
      "mean_token_accuracy": 0.20442665964365006,
      "num_tokens": 9252689.0,
      "step": 5265
    },
    {
      "entropy": 5.283981084823608,
      "epoch": 4.130094043887147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003686722944937215,
      "loss": 4.6952,
      "mean_token_accuracy": 0.2058682307600975,
      "num_tokens": 9261118.0,
      "step": 5270
    },
    {
      "entropy": 5.173985576629638,
      "epoch": 4.134012539184953,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036839872660749084,
      "loss": 4.6979,
      "mean_token_accuracy": 0.20479130744934082,
      "num_tokens": 9269205.0,
      "step": 5275
    },
    {
      "entropy": 5.173965358734131,
      "epoch": 4.137931034482759,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003681249918025918,
      "loss": 4.7216,
      "mean_token_accuracy": 0.20660676509141923,
      "num_tokens": 9277770.0,
      "step": 5280
    },
    {
      "entropy": 5.217983818054199,
      "epoch": 4.141849529780564,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036785109056823297,
      "loss": 4.7612,
      "mean_token_accuracy": 0.2042287290096283,
      "num_tokens": 9286589.0,
      "step": 5285
    },
    {
      "entropy": 5.2604128360748295,
      "epoch": 4.14576802507837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036757702339392,
      "loss": 4.7113,
      "mean_token_accuracy": 0.2040370836853981,
      "num_tokens": 9295264.0,
      "step": 5290
    },
    {
      "entropy": 5.17775526046753,
      "epoch": 4.149686520376176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036730279076945574,
      "loss": 4.7566,
      "mean_token_accuracy": 0.20164868831634522,
      "num_tokens": 9303525.0,
      "step": 5295
    },
    {
      "entropy": 5.181302833557129,
      "epoch": 4.153605015673981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003670283931849382,
      "loss": 4.796,
      "mean_token_accuracy": 0.19736869633197784,
      "num_tokens": 9312839.0,
      "step": 5300
    },
    {
      "entropy": 5.265648221969604,
      "epoch": 4.157523510971787,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036675383113076023,
      "loss": 4.7344,
      "mean_token_accuracy": 0.2079014450311661,
      "num_tokens": 9321462.0,
      "step": 5305
    },
    {
      "entropy": 5.228754711151123,
      "epoch": 4.161442006269592,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00036647910509760896,
      "loss": 4.7659,
      "mean_token_accuracy": 0.20292360931634904,
      "num_tokens": 9329315.0,
      "step": 5310
    },
    {
      "entropy": 5.245350646972656,
      "epoch": 4.165360501567398,
      "grad_norm": 1.4375,
      "learning_rate": 0.00036620421557646414,
      "loss": 4.7175,
      "mean_token_accuracy": 0.20611485093832016,
      "num_tokens": 9336997.0,
      "step": 5315
    },
    {
      "entropy": 5.157686233520508,
      "epoch": 4.169278996865204,
      "grad_norm": 1.203125,
      "learning_rate": 0.000365929163058598,
      "loss": 4.6891,
      "mean_token_accuracy": 0.20539978593587876,
      "num_tokens": 9345364.0,
      "step": 5320
    },
    {
      "entropy": 5.20963773727417,
      "epoch": 4.173197492163009,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003656539480355741,
      "loss": 4.6965,
      "mean_token_accuracy": 0.2085391789674759,
      "num_tokens": 9353961.0,
      "step": 5325
    },
    {
      "entropy": 5.306543636322021,
      "epoch": 4.177115987460815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003653785709992462,
      "loss": 4.8528,
      "mean_token_accuracy": 0.19711169451475144,
      "num_tokens": 9362901.0,
      "step": 5330
    },
    {
      "entropy": 5.241455698013306,
      "epoch": 4.181034482758621,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036510303244175775,
      "loss": 4.7623,
      "mean_token_accuracy": 0.2002245917916298,
      "num_tokens": 9371800.0,
      "step": 5335
    },
    {
      "entropy": 5.204394245147705,
      "epoch": 4.184952978056426,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036482733285554073,
      "loss": 4.7403,
      "mean_token_accuracy": 0.2114669218659401,
      "num_tokens": 9380033.0,
      "step": 5340
    },
    {
      "entropy": 5.338855171203614,
      "epoch": 4.188871473354232,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036455147273331507,
      "loss": 4.818,
      "mean_token_accuracy": 0.2050924703478813,
      "num_tokens": 9389264.0,
      "step": 5345
    },
    {
      "entropy": 5.244389629364013,
      "epoch": 4.192789968652038,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003642754525680876,
      "loss": 4.7366,
      "mean_token_accuracy": 0.20819579660892487,
      "num_tokens": 9398112.0,
      "step": 5350
    },
    {
      "entropy": 5.270562744140625,
      "epoch": 4.196708463949843,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036399927285315104,
      "loss": 4.852,
      "mean_token_accuracy": 0.19091200530529023,
      "num_tokens": 9407400.0,
      "step": 5355
    },
    {
      "entropy": 5.262473392486572,
      "epoch": 4.200626959247649,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003637229340820834,
      "loss": 4.7402,
      "mean_token_accuracy": 0.1980791360139847,
      "num_tokens": 9416351.0,
      "step": 5360
    },
    {
      "entropy": 5.322093391418457,
      "epoch": 4.204545454545454,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003634464367487467,
      "loss": 4.769,
      "mean_token_accuracy": 0.19876633137464522,
      "num_tokens": 9424577.0,
      "step": 5365
    },
    {
      "entropy": 5.229406690597534,
      "epoch": 4.20846394984326,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003631697813472867,
      "loss": 4.7689,
      "mean_token_accuracy": 0.20547029227018357,
      "num_tokens": 9434242.0,
      "step": 5370
    },
    {
      "entropy": 5.135349369049072,
      "epoch": 4.212382445141066,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036289296837213136,
      "loss": 4.7403,
      "mean_token_accuracy": 0.20730825364589692,
      "num_tokens": 9443247.0,
      "step": 5375
    },
    {
      "entropy": 5.197679424285889,
      "epoch": 4.216300940438871,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036261599831799034,
      "loss": 4.6957,
      "mean_token_accuracy": 0.2066801145672798,
      "num_tokens": 9451998.0,
      "step": 5380
    },
    {
      "entropy": 5.223357629776001,
      "epoch": 4.220219435736677,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003623388716798541,
      "loss": 4.648,
      "mean_token_accuracy": 0.20918530970811844,
      "num_tokens": 9459993.0,
      "step": 5385
    },
    {
      "entropy": 5.188335752487182,
      "epoch": 4.224137931034483,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036206158895299277,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2085404023528099,
      "num_tokens": 9468036.0,
      "step": 5390
    },
    {
      "entropy": 5.173120594024658,
      "epoch": 4.228056426332288,
      "grad_norm": 1.25,
      "learning_rate": 0.00036178415063295574,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2053625300526619,
      "num_tokens": 9477031.0,
      "step": 5395
    },
    {
      "entropy": 5.218107509613037,
      "epoch": 4.231974921630094,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036150655721557,
      "loss": 4.6972,
      "mean_token_accuracy": 0.20644138306379317,
      "num_tokens": 9485687.0,
      "step": 5400
    },
    {
      "entropy": 5.200930833816528,
      "epoch": 4.2358934169279,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003612288091969402,
      "loss": 4.7309,
      "mean_token_accuracy": 0.20934922993183136,
      "num_tokens": 9494613.0,
      "step": 5405
    },
    {
      "entropy": 5.260763263702392,
      "epoch": 4.239811912225705,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036095090707344697,
      "loss": 4.7397,
      "mean_token_accuracy": 0.19972426146268846,
      "num_tokens": 9503522.0,
      "step": 5410
    },
    {
      "entropy": 5.177921104431152,
      "epoch": 4.243730407523511,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036067285134174654,
      "loss": 4.7524,
      "mean_token_accuracy": 0.20293453335762024,
      "num_tokens": 9512293.0,
      "step": 5415
    },
    {
      "entropy": 5.300930166244507,
      "epoch": 4.247648902821316,
      "grad_norm": 1.375,
      "learning_rate": 0.0003603946424987696,
      "loss": 4.8561,
      "mean_token_accuracy": 0.19625966250896454,
      "num_tokens": 9520999.0,
      "step": 5420
    },
    {
      "entropy": 5.2637711524963375,
      "epoch": 4.251567398119122,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036011628104172045,
      "loss": 4.796,
      "mean_token_accuracy": 0.19859230369329453,
      "num_tokens": 9529798.0,
      "step": 5425
    },
    {
      "entropy": 5.273455810546875,
      "epoch": 4.255485893416928,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035983776746807613,
      "loss": 4.7255,
      "mean_token_accuracy": 0.2029760554432869,
      "num_tokens": 9538668.0,
      "step": 5430
    },
    {
      "entropy": 5.234575796127319,
      "epoch": 4.259404388714733,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595591022755855,
      "loss": 4.7656,
      "mean_token_accuracy": 0.20429227501153946,
      "num_tokens": 9546973.0,
      "step": 5435
    },
    {
      "entropy": 5.135591840744018,
      "epoch": 4.263322884012539,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003592802859622684,
      "loss": 4.7292,
      "mean_token_accuracy": 0.2014813393354416,
      "num_tokens": 9555947.0,
      "step": 5440
    },
    {
      "entropy": 5.214759349822998,
      "epoch": 4.267241379310345,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000359001319026415,
      "loss": 4.7918,
      "mean_token_accuracy": 0.19812078177928924,
      "num_tokens": 9564730.0,
      "step": 5445
    },
    {
      "entropy": 5.177644729614258,
      "epoch": 4.27115987460815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035872220196658426,
      "loss": 4.7336,
      "mean_token_accuracy": 0.2008720114827156,
      "num_tokens": 9573470.0,
      "step": 5450
    },
    {
      "entropy": 5.2544333934783936,
      "epoch": 4.275078369905956,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035844293528160384,
      "loss": 4.798,
      "mean_token_accuracy": 0.1972871169447899,
      "num_tokens": 9581607.0,
      "step": 5455
    },
    {
      "entropy": 5.206086444854736,
      "epoch": 4.278996865203762,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003581635194705682,
      "loss": 4.7328,
      "mean_token_accuracy": 0.20472805202007294,
      "num_tokens": 9589801.0,
      "step": 5460
    },
    {
      "entropy": 5.250318717956543,
      "epoch": 4.282915360501567,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035788395503283914,
      "loss": 4.8063,
      "mean_token_accuracy": 0.19125736355781556,
      "num_tokens": 9599983.0,
      "step": 5465
    },
    {
      "entropy": 5.260366487503052,
      "epoch": 4.286833855799373,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035760424246804334,
      "loss": 4.6785,
      "mean_token_accuracy": 0.21201709657907486,
      "num_tokens": 9608681.0,
      "step": 5470
    },
    {
      "entropy": 5.184821128845215,
      "epoch": 4.290752351097178,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003573243822760726,
      "loss": 4.7094,
      "mean_token_accuracy": 0.20927670150995253,
      "num_tokens": 9616461.0,
      "step": 5475
    },
    {
      "entropy": 5.233028697967529,
      "epoch": 4.294670846394984,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003570443749570825,
      "loss": 4.8039,
      "mean_token_accuracy": 0.19788059294223787,
      "num_tokens": 9625480.0,
      "step": 5480
    },
    {
      "entropy": 5.208855533599854,
      "epoch": 4.29858934169279,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003567642210114914,
      "loss": 4.7728,
      "mean_token_accuracy": 0.19928020089864731,
      "num_tokens": 9634172.0,
      "step": 5485
    },
    {
      "entropy": 5.150461339950562,
      "epoch": 4.302507836990595,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035648392093998,
      "loss": 4.7509,
      "mean_token_accuracy": 0.20163060426712037,
      "num_tokens": 9643196.0,
      "step": 5490
    },
    {
      "entropy": 5.187815856933594,
      "epoch": 4.306426332288401,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003562034752434899,
      "loss": 4.6784,
      "mean_token_accuracy": 0.2101092144846916,
      "num_tokens": 9652440.0,
      "step": 5495
    },
    {
      "entropy": 5.227723073959351,
      "epoch": 4.310344827586207,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003559228844232229,
      "loss": 4.7592,
      "mean_token_accuracy": 0.1967903658747673,
      "num_tokens": 9661248.0,
      "step": 5500
    },
    {
      "epoch": 4.310344827586207,
      "eval_entropy": 5.124320444210555,
      "eval_loss": 5.706399917602539,
      "eval_mean_token_accuracy": 0.166106693027902,
      "eval_num_tokens": 9661248.0,
      "eval_runtime": 3.0454,
      "eval_samples_per_second": 1353.498,
      "eval_steps_per_second": 169.434,
      "step": 5500
    },
    {
      "entropy": 5.2364424705505375,
      "epoch": 4.314263322884012,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003556421489806406,
      "loss": 4.7873,
      "mean_token_accuracy": 0.20943569093942643,
      "num_tokens": 9670544.0,
      "step": 5505
    },
    {
      "entropy": 5.223901319503784,
      "epoch": 4.318181818181818,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003553612694174624,
      "loss": 4.7649,
      "mean_token_accuracy": 0.2051727369427681,
      "num_tokens": 9679331.0,
      "step": 5510
    },
    {
      "entropy": 5.325501489639282,
      "epoch": 4.322100313479623,
      "grad_norm": 1.25,
      "learning_rate": 0.0003550802462356659,
      "loss": 4.8164,
      "mean_token_accuracy": 0.1961141601204872,
      "num_tokens": 9688211.0,
      "step": 5515
    },
    {
      "entropy": 5.122488737106323,
      "epoch": 4.326018808777429,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035479907993748484,
      "loss": 4.6826,
      "mean_token_accuracy": 0.21221625655889512,
      "num_tokens": 9696231.0,
      "step": 5520
    },
    {
      "entropy": 5.2129480838775635,
      "epoch": 4.329937304075235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035451777102540915,
      "loss": 4.7649,
      "mean_token_accuracy": 0.19785977751016617,
      "num_tokens": 9705577.0,
      "step": 5525
    },
    {
      "entropy": 5.296283626556397,
      "epoch": 4.33385579937304,
      "grad_norm": 1.25,
      "learning_rate": 0.0003542363200021835,
      "loss": 4.7937,
      "mean_token_accuracy": 0.2032880187034607,
      "num_tokens": 9715003.0,
      "step": 5530
    },
    {
      "entropy": 5.189212846755981,
      "epoch": 4.337774294670846,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003539547273708064,
      "loss": 4.6533,
      "mean_token_accuracy": 0.2088500365614891,
      "num_tokens": 9723513.0,
      "step": 5535
    },
    {
      "entropy": 5.153302812576294,
      "epoch": 4.341692789968652,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035367299363452966,
      "loss": 4.7687,
      "mean_token_accuracy": 0.20194215774536134,
      "num_tokens": 9731800.0,
      "step": 5540
    },
    {
      "entropy": 5.2159627914428714,
      "epoch": 4.3456112852664575,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035339111929685694,
      "loss": 4.7259,
      "mean_token_accuracy": 0.20310318619012832,
      "num_tokens": 9740356.0,
      "step": 5545
    },
    {
      "entropy": 5.274029684066773,
      "epoch": 4.349529780564263,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035310910486154365,
      "loss": 4.7694,
      "mean_token_accuracy": 0.19272388219833375,
      "num_tokens": 9749529.0,
      "step": 5550
    },
    {
      "entropy": 5.241015338897705,
      "epoch": 4.353448275862069,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003528269508325951,
      "loss": 4.7479,
      "mean_token_accuracy": 0.2036553680896759,
      "num_tokens": 9758750.0,
      "step": 5555
    },
    {
      "entropy": 5.2122029781341555,
      "epoch": 4.3573667711598745,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003525446577142663,
      "loss": 4.8169,
      "mean_token_accuracy": 0.20145424455404282,
      "num_tokens": 9766992.0,
      "step": 5560
    },
    {
      "entropy": 5.269633674621582,
      "epoch": 4.36128526645768,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035226222601106105,
      "loss": 4.8432,
      "mean_token_accuracy": 0.1976443663239479,
      "num_tokens": 9777022.0,
      "step": 5565
    },
    {
      "entropy": 5.160535860061645,
      "epoch": 4.3652037617554855,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003519796562277303,
      "loss": 4.5778,
      "mean_token_accuracy": 0.21674007922410965,
      "num_tokens": 9785863.0,
      "step": 5570
    },
    {
      "entropy": 5.1979029178619385,
      "epoch": 4.3691222570532915,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003516969488692723,
      "loss": 4.743,
      "mean_token_accuracy": 0.19586774706840515,
      "num_tokens": 9794357.0,
      "step": 5575
    },
    {
      "entropy": 5.161561489105225,
      "epoch": 4.3730407523510975,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003514141044409308,
      "loss": 4.7217,
      "mean_token_accuracy": 0.2053699791431427,
      "num_tokens": 9803128.0,
      "step": 5580
    },
    {
      "entropy": 5.247322034835816,
      "epoch": 4.3769592476489025,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003511311234481947,
      "loss": 4.8241,
      "mean_token_accuracy": 0.19795375019311906,
      "num_tokens": 9812046.0,
      "step": 5585
    },
    {
      "entropy": 5.20223560333252,
      "epoch": 4.3808777429467085,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035084800639679695,
      "loss": 4.7007,
      "mean_token_accuracy": 0.20444824993610383,
      "num_tokens": 9820935.0,
      "step": 5590
    },
    {
      "entropy": 5.238837099075317,
      "epoch": 4.3847962382445145,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035056475379271356,
      "loss": 4.8195,
      "mean_token_accuracy": 0.19855652898550033,
      "num_tokens": 9829585.0,
      "step": 5595
    },
    {
      "entropy": 5.292247867584228,
      "epoch": 4.38871473354232,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003502813661421629,
      "loss": 4.8788,
      "mean_token_accuracy": 0.1960359901189804,
      "num_tokens": 9839302.0,
      "step": 5600
    },
    {
      "entropy": 5.24213433265686,
      "epoch": 4.3926332288401255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003499978439516047,
      "loss": 4.7795,
      "mean_token_accuracy": 0.20599234104156494,
      "num_tokens": 9848330.0,
      "step": 5605
    },
    {
      "entropy": 5.176373910903931,
      "epoch": 4.396551724137931,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034971418772773895,
      "loss": 4.7406,
      "mean_token_accuracy": 0.208706134557724,
      "num_tokens": 9856928.0,
      "step": 5610
    },
    {
      "entropy": 5.25033164024353,
      "epoch": 4.400470219435737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003494303979775054,
      "loss": 4.8064,
      "mean_token_accuracy": 0.19775945842266082,
      "num_tokens": 9865464.0,
      "step": 5615
    },
    {
      "entropy": 5.210266208648681,
      "epoch": 4.4043887147335425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034914647520808235,
      "loss": 4.7502,
      "mean_token_accuracy": 0.20326811522245408,
      "num_tokens": 9874345.0,
      "step": 5620
    },
    {
      "entropy": 5.117397737503052,
      "epoch": 4.408307210031348,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003488624199268859,
      "loss": 4.7194,
      "mean_token_accuracy": 0.21135797798633577,
      "num_tokens": 9883135.0,
      "step": 5625
    },
    {
      "entropy": 5.232568311691284,
      "epoch": 4.412225705329154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003485782326415687,
      "loss": 4.7949,
      "mean_token_accuracy": 0.20602937042713165,
      "num_tokens": 9892229.0,
      "step": 5630
    },
    {
      "entropy": 5.258719396591187,
      "epoch": 4.41614420062696,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003482939138600197,
      "loss": 4.7953,
      "mean_token_accuracy": 0.2047722101211548,
      "num_tokens": 9901044.0,
      "step": 5635
    },
    {
      "entropy": 5.271085977554321,
      "epoch": 4.420062695924765,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003480094640903626,
      "loss": 4.8068,
      "mean_token_accuracy": 0.19732605814933776,
      "num_tokens": 9909818.0,
      "step": 5640
    },
    {
      "entropy": 5.1796094417572025,
      "epoch": 4.423981191222571,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034772488384095517,
      "loss": 4.6924,
      "mean_token_accuracy": 0.2098855271935463,
      "num_tokens": 9919295.0,
      "step": 5645
    },
    {
      "entropy": 5.167816162109375,
      "epoch": 4.427899686520377,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034744017362038854,
      "loss": 4.8,
      "mean_token_accuracy": 0.1989780381321907,
      "num_tokens": 9927539.0,
      "step": 5650
    },
    {
      "entropy": 5.21839165687561,
      "epoch": 4.431818181818182,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034715533393748604,
      "loss": 4.8254,
      "mean_token_accuracy": 0.1952006235718727,
      "num_tokens": 9936540.0,
      "step": 5655
    },
    {
      "entropy": 5.212318086624146,
      "epoch": 4.435736677115988,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034687036530130244,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2107584744691849,
      "num_tokens": 9945065.0,
      "step": 5660
    },
    {
      "entropy": 5.204560041427612,
      "epoch": 4.439655172413793,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003465852682211227,
      "loss": 4.7149,
      "mean_token_accuracy": 0.2021802395582199,
      "num_tokens": 9954155.0,
      "step": 5665
    },
    {
      "entropy": 5.19600191116333,
      "epoch": 4.443573667711599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003463000432064617,
      "loss": 4.7824,
      "mean_token_accuracy": 0.19694020301103593,
      "num_tokens": 9963269.0,
      "step": 5670
    },
    {
      "entropy": 5.251050615310669,
      "epoch": 4.447492163009405,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034601469076706276,
      "loss": 4.8276,
      "mean_token_accuracy": 0.2033545345067978,
      "num_tokens": 9972688.0,
      "step": 5675
    },
    {
      "entropy": 5.28069658279419,
      "epoch": 4.45141065830721,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003457292114128968,
      "loss": 4.7741,
      "mean_token_accuracy": 0.2010485976934433,
      "num_tokens": 9981283.0,
      "step": 5680
    },
    {
      "entropy": 5.206014728546142,
      "epoch": 4.455329153605016,
      "grad_norm": 1.40625,
      "learning_rate": 0.000345443605654162,
      "loss": 4.7993,
      "mean_token_accuracy": 0.20102507770061492,
      "num_tokens": 9989870.0,
      "step": 5685
    },
    {
      "entropy": 5.243091201782226,
      "epoch": 4.459247648902822,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034515787400128195,
      "loss": 4.7855,
      "mean_token_accuracy": 0.20019320845603944,
      "num_tokens": 9998136.0,
      "step": 5690
    },
    {
      "entropy": 5.202512550354004,
      "epoch": 4.463166144200627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003448720169649054,
      "loss": 4.7368,
      "mean_token_accuracy": 0.20058425664901733,
      "num_tokens": 10006730.0,
      "step": 5695
    },
    {
      "entropy": 5.20303521156311,
      "epoch": 4.467084639498433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003445860350559055,
      "loss": 4.7531,
      "mean_token_accuracy": 0.2093263551592827,
      "num_tokens": 10015517.0,
      "step": 5700
    },
    {
      "entropy": 5.153885269165039,
      "epoch": 4.471003134796238,
      "grad_norm": 1.296875,
      "learning_rate": 0.000344299928785378,
      "loss": 4.7874,
      "mean_token_accuracy": 0.20614300519227982,
      "num_tokens": 10024552.0,
      "step": 5705
    },
    {
      "entropy": 5.192214012145996,
      "epoch": 4.474921630094044,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034401369866464133,
      "loss": 4.7658,
      "mean_token_accuracy": 0.20103832185268403,
      "num_tokens": 10033523.0,
      "step": 5710
    },
    {
      "entropy": 5.164818048477173,
      "epoch": 4.47884012539185,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034372734520523505,
      "loss": 4.6417,
      "mean_token_accuracy": 0.20695265233516694,
      "num_tokens": 10041988.0,
      "step": 5715
    },
    {
      "entropy": 5.17066764831543,
      "epoch": 4.482758620689655,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003434408689189193,
      "loss": 4.7474,
      "mean_token_accuracy": 0.20707701444625853,
      "num_tokens": 10050543.0,
      "step": 5720
    },
    {
      "entropy": 5.1972681999206545,
      "epoch": 4.486677115987461,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003431542703176737,
      "loss": 4.7547,
      "mean_token_accuracy": 0.20500834435224533,
      "num_tokens": 10059300.0,
      "step": 5725
    },
    {
      "entropy": 5.163868856430054,
      "epoch": 4.490595611285267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003428675499136963,
      "loss": 4.7862,
      "mean_token_accuracy": 0.20483419448137283,
      "num_tokens": 10068535.0,
      "step": 5730
    },
    {
      "entropy": 5.285054397583008,
      "epoch": 4.494514106583072,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000342580708219403,
      "loss": 4.8345,
      "mean_token_accuracy": 0.20355326384305955,
      "num_tokens": 10077731.0,
      "step": 5735
    },
    {
      "entropy": 5.230941438674927,
      "epoch": 4.498432601880878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003422937457474264,
      "loss": 4.7532,
      "mean_token_accuracy": 0.20394681096076966,
      "num_tokens": 10087146.0,
      "step": 5740
    },
    {
      "entropy": 5.185268449783325,
      "epoch": 4.502351097178684,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034200666301061495,
      "loss": 4.7336,
      "mean_token_accuracy": 0.20517101138830185,
      "num_tokens": 10096659.0,
      "step": 5745
    },
    {
      "entropy": 5.226880502700806,
      "epoch": 4.506269592476489,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034171946052203217,
      "loss": 4.837,
      "mean_token_accuracy": 0.1966334119439125,
      "num_tokens": 10105403.0,
      "step": 5750
    },
    {
      "entropy": 5.292223024368286,
      "epoch": 4.510188087774295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034143213879495526,
      "loss": 4.7733,
      "mean_token_accuracy": 0.20389281809329987,
      "num_tokens": 10115410.0,
      "step": 5755
    },
    {
      "entropy": 5.1936118602752686,
      "epoch": 4.514106583072101,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000341144698342875,
      "loss": 4.7916,
      "mean_token_accuracy": 0.20102183967828752,
      "num_tokens": 10123800.0,
      "step": 5760
    },
    {
      "entropy": 5.259945249557495,
      "epoch": 4.518025078369906,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003408571396794939,
      "loss": 4.8273,
      "mean_token_accuracy": 0.1977522239089012,
      "num_tokens": 10132775.0,
      "step": 5765
    },
    {
      "entropy": 5.21550874710083,
      "epoch": 4.521943573667712,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003405694633187259,
      "loss": 4.7113,
      "mean_token_accuracy": 0.20720769613981246,
      "num_tokens": 10141811.0,
      "step": 5770
    },
    {
      "entropy": 5.13583197593689,
      "epoch": 4.525862068965517,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034028166977469545,
      "loss": 4.7283,
      "mean_token_accuracy": 0.20180071294307708,
      "num_tokens": 10150823.0,
      "step": 5775
    },
    {
      "entropy": 5.210734462738037,
      "epoch": 4.529780564263323,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033999375956173607,
      "loss": 4.7244,
      "mean_token_accuracy": 0.20741142630577086,
      "num_tokens": 10158983.0,
      "step": 5780
    },
    {
      "entropy": 5.25197024345398,
      "epoch": 4.533699059561129,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003397057331943902,
      "loss": 4.738,
      "mean_token_accuracy": 0.20216021984815596,
      "num_tokens": 10167596.0,
      "step": 5785
    },
    {
      "entropy": 5.211240625381469,
      "epoch": 4.537617554858934,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003394175911874076,
      "loss": 4.815,
      "mean_token_accuracy": 0.20349200069904327,
      "num_tokens": 10176216.0,
      "step": 5790
    },
    {
      "entropy": 5.167293882369995,
      "epoch": 4.54153605015674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003391293340557446,
      "loss": 4.721,
      "mean_token_accuracy": 0.20417336821556092,
      "num_tokens": 10185685.0,
      "step": 5795
    },
    {
      "entropy": 5.191263675689697,
      "epoch": 4.545454545454545,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003388409623145637,
      "loss": 4.7133,
      "mean_token_accuracy": 0.20614122599363327,
      "num_tokens": 10193679.0,
      "step": 5800
    },
    {
      "entropy": 5.194652891159057,
      "epoch": 4.549373040752351,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00033855247647923177,
      "loss": 4.7633,
      "mean_token_accuracy": 0.2070121005177498,
      "num_tokens": 10201154.0,
      "step": 5805
    },
    {
      "entropy": 5.180051946640015,
      "epoch": 4.553291536050157,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003382638770653198,
      "loss": 4.6943,
      "mean_token_accuracy": 0.2105014681816101,
      "num_tokens": 10209180.0,
      "step": 5810
    },
    {
      "entropy": 5.22576003074646,
      "epoch": 4.557210031347962,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003379751645886017,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2047353833913803,
      "num_tokens": 10217430.0,
      "step": 5815
    },
    {
      "entropy": 5.198554801940918,
      "epoch": 4.561128526645768,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0003376863395650537,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2034112185239792,
      "num_tokens": 10226453.0,
      "step": 5820
    },
    {
      "entropy": 5.309712648391724,
      "epoch": 4.565047021943574,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033739740251085263,
      "loss": 4.8673,
      "mean_token_accuracy": 0.19931832253932952,
      "num_tokens": 10235005.0,
      "step": 5825
    },
    {
      "entropy": 5.25501708984375,
      "epoch": 4.568965517241379,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033710835394237603,
      "loss": 4.7918,
      "mean_token_accuracy": 0.20499035269021987,
      "num_tokens": 10243800.0,
      "step": 5830
    },
    {
      "entropy": 5.199924373626709,
      "epoch": 4.572884012539185,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033681919437620066,
      "loss": 4.7631,
      "mean_token_accuracy": 0.20323984026908876,
      "num_tokens": 10252370.0,
      "step": 5835
    },
    {
      "entropy": 5.2229842185974125,
      "epoch": 4.576802507836991,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003365299243291014,
      "loss": 4.8283,
      "mean_token_accuracy": 0.19584422260522844,
      "num_tokens": 10260777.0,
      "step": 5840
    },
    {
      "entropy": 5.1856544494628904,
      "epoch": 4.580721003134796,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033624054431805067,
      "loss": 4.8259,
      "mean_token_accuracy": 0.1929199680685997,
      "num_tokens": 10270571.0,
      "step": 5845
    },
    {
      "entropy": 5.265431880950928,
      "epoch": 4.584639498432602,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033595105486021763,
      "loss": 4.7976,
      "mean_token_accuracy": 0.20136764347553254,
      "num_tokens": 10279284.0,
      "step": 5850
    },
    {
      "entropy": 5.174901247024536,
      "epoch": 4.588557993730408,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033566145647296675,
      "loss": 4.7141,
      "mean_token_accuracy": 0.20317427664995194,
      "num_tokens": 10288610.0,
      "step": 5855
    },
    {
      "entropy": 5.170123672485351,
      "epoch": 4.592476489028213,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033537174967385744,
      "loss": 4.6938,
      "mean_token_accuracy": 0.207176510989666,
      "num_tokens": 10296603.0,
      "step": 5860
    },
    {
      "entropy": 5.1325733184814455,
      "epoch": 4.596394984326019,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033508193498064254,
      "loss": 4.7024,
      "mean_token_accuracy": 0.20882961004972458,
      "num_tokens": 10305199.0,
      "step": 5865
    },
    {
      "entropy": 5.133954334259033,
      "epoch": 4.600313479623824,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033479201291126807,
      "loss": 4.6442,
      "mean_token_accuracy": 0.21270055323839188,
      "num_tokens": 10313967.0,
      "step": 5870
    },
    {
      "entropy": 5.216064834594727,
      "epoch": 4.60423197492163,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003345019839838717,
      "loss": 4.7632,
      "mean_token_accuracy": 0.20564036667346955,
      "num_tokens": 10322532.0,
      "step": 5875
    },
    {
      "entropy": 5.147199583053589,
      "epoch": 4.608150470219436,
      "grad_norm": 1.25,
      "learning_rate": 0.0003342118487167822,
      "loss": 4.6295,
      "mean_token_accuracy": 0.21614251285791397,
      "num_tokens": 10330631.0,
      "step": 5880
    },
    {
      "entropy": 5.092808675765991,
      "epoch": 4.612068965517241,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033392160762851813,
      "loss": 4.6467,
      "mean_token_accuracy": 0.21375302076339722,
      "num_tokens": 10339229.0,
      "step": 5885
    },
    {
      "entropy": 5.2239381790161135,
      "epoch": 4.615987460815047,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003336312612377877,
      "loss": 4.7709,
      "mean_token_accuracy": 0.19939426928758622,
      "num_tokens": 10348882.0,
      "step": 5890
    },
    {
      "entropy": 5.308163690567016,
      "epoch": 4.619905956112853,
      "grad_norm": 1.421875,
      "learning_rate": 0.00033334081006348683,
      "loss": 4.9049,
      "mean_token_accuracy": 0.1920166552066803,
      "num_tokens": 10358154.0,
      "step": 5895
    },
    {
      "entropy": 5.231937742233276,
      "epoch": 4.623824451410658,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003330502546246988,
      "loss": 4.8285,
      "mean_token_accuracy": 0.20040331333875655,
      "num_tokens": 10367100.0,
      "step": 5900
    },
    {
      "entropy": 5.095721673965454,
      "epoch": 4.627742946708464,
      "grad_norm": 1.25,
      "learning_rate": 0.00033275959544069335,
      "loss": 4.6292,
      "mean_token_accuracy": 0.21857758164405822,
      "num_tokens": 10375711.0,
      "step": 5905
    },
    {
      "entropy": 5.186459970474243,
      "epoch": 4.631661442006269,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033246883303092564,
      "loss": 4.7357,
      "mean_token_accuracy": 0.20545791387557982,
      "num_tokens": 10384513.0,
      "step": 5910
    },
    {
      "entropy": 5.2376487255096436,
      "epoch": 4.635579937304075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033217796791503503,
      "loss": 4.8019,
      "mean_token_accuracy": 0.1997740313410759,
      "num_tokens": 10393564.0,
      "step": 5915
    },
    {
      "entropy": 5.241690254211425,
      "epoch": 4.639498432601881,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003318870006128449,
      "loss": 4.804,
      "mean_token_accuracy": 0.19869185239076614,
      "num_tokens": 10402042.0,
      "step": 5920
    },
    {
      "entropy": 5.167234230041504,
      "epoch": 4.643416927899686,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003315959316443608,
      "loss": 4.6908,
      "mean_token_accuracy": 0.20043910890817643,
      "num_tokens": 10410847.0,
      "step": 5925
    },
    {
      "entropy": 5.230126237869262,
      "epoch": 4.647335423197492,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003313047615297703,
      "loss": 4.8232,
      "mean_token_accuracy": 0.19003514498472213,
      "num_tokens": 10420366.0,
      "step": 5930
    },
    {
      "entropy": 5.181274890899658,
      "epoch": 4.651253918495298,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033101349078944165,
      "loss": 4.7225,
      "mean_token_accuracy": 0.2050636738538742,
      "num_tokens": 10428870.0,
      "step": 5935
    },
    {
      "entropy": 5.136593055725098,
      "epoch": 4.655172413793103,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003307221199439227,
      "loss": 4.7725,
      "mean_token_accuracy": 0.19795012176036836,
      "num_tokens": 10437714.0,
      "step": 5940
    },
    {
      "entropy": 5.144877290725708,
      "epoch": 4.659090909090909,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033043064951394045,
      "loss": 4.6812,
      "mean_token_accuracy": 0.20986449718475342,
      "num_tokens": 10447198.0,
      "step": 5945
    },
    {
      "entropy": 5.223510456085205,
      "epoch": 4.663009404388715,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00033013908002039994,
      "loss": 4.7311,
      "mean_token_accuracy": 0.20731364637613298,
      "num_tokens": 10455790.0,
      "step": 5950
    },
    {
      "entropy": 5.245111513137817,
      "epoch": 4.66692789968652,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032984741198438305,
      "loss": 4.8189,
      "mean_token_accuracy": 0.2073121890425682,
      "num_tokens": 10464094.0,
      "step": 5955
    },
    {
      "entropy": 5.175472354888916,
      "epoch": 4.670846394984326,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003295556459271478,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2015923261642456,
      "num_tokens": 10473201.0,
      "step": 5960
    },
    {
      "entropy": 5.256532144546509,
      "epoch": 4.674764890282132,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003292637823701276,
      "loss": 4.8064,
      "mean_token_accuracy": 0.20188169032335282,
      "num_tokens": 10482617.0,
      "step": 5965
    },
    {
      "entropy": 5.219484186172485,
      "epoch": 4.678683385579937,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003289718218349298,
      "loss": 4.6915,
      "mean_token_accuracy": 0.21117063313722612,
      "num_tokens": 10491702.0,
      "step": 5970
    },
    {
      "entropy": 5.256941509246826,
      "epoch": 4.682601880877743,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003286797648433351,
      "loss": 4.8795,
      "mean_token_accuracy": 0.19534407705068588,
      "num_tokens": 10501189.0,
      "step": 5975
    },
    {
      "entropy": 5.191068315505982,
      "epoch": 4.686520376175548,
      "grad_norm": 1.21875,
      "learning_rate": 0.000328387611917297,
      "loss": 4.7464,
      "mean_token_accuracy": 0.2013760656118393,
      "num_tokens": 10510272.0,
      "step": 5980
    },
    {
      "entropy": 5.243597173690796,
      "epoch": 4.690438871473354,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003280953635789401,
      "loss": 4.8216,
      "mean_token_accuracy": 0.20551169365644456,
      "num_tokens": 10518782.0,
      "step": 5985
    },
    {
      "entropy": 5.139669942855835,
      "epoch": 4.69435736677116,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003278030203505594,
      "loss": 4.7656,
      "mean_token_accuracy": 0.2065542533993721,
      "num_tokens": 10528262.0,
      "step": 5990
    },
    {
      "entropy": 5.116983032226562,
      "epoch": 4.698275862068965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003275105827546197,
      "loss": 4.737,
      "mean_token_accuracy": 0.2054427146911621,
      "num_tokens": 10536399.0,
      "step": 5995
    },
    {
      "entropy": 5.238217115402222,
      "epoch": 4.702194357366771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032721805131375446,
      "loss": 4.8068,
      "mean_token_accuracy": 0.1975304141640663,
      "num_tokens": 10544842.0,
      "step": 6000
    },
    {
      "epoch": 4.702194357366771,
      "eval_entropy": 5.1561952533648,
      "eval_loss": 5.674187183380127,
      "eval_mean_token_accuracy": 0.16827263257126937,
      "eval_num_tokens": 10544842.0,
      "eval_runtime": 2.8453,
      "eval_samples_per_second": 1448.704,
      "eval_steps_per_second": 181.352,
      "step": 6000
    },
    {
      "entropy": 5.218733882904052,
      "epoch": 4.706112852664576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003269254265507647,
      "loss": 4.7267,
      "mean_token_accuracy": 0.20571708232164382,
      "num_tokens": 10553630.0,
      "step": 6005
    },
    {
      "entropy": 5.138806295394898,
      "epoch": 4.710031347962382,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032663270898861845,
      "loss": 4.754,
      "mean_token_accuracy": 0.20039431750774384,
      "num_tokens": 10561714.0,
      "step": 6010
    },
    {
      "entropy": 5.17672438621521,
      "epoch": 4.713949843260188,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032633989915044944,
      "loss": 4.8446,
      "mean_token_accuracy": 0.193728069961071,
      "num_tokens": 10571075.0,
      "step": 6015
    },
    {
      "entropy": 5.261137962341309,
      "epoch": 4.717868338557993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032604699755955615,
      "loss": 4.6936,
      "mean_token_accuracy": 0.21206937581300736,
      "num_tokens": 10579698.0,
      "step": 6020
    },
    {
      "entropy": 5.144515323638916,
      "epoch": 4.721786833855799,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032575400473940135,
      "loss": 4.7364,
      "mean_token_accuracy": 0.20617617517709733,
      "num_tokens": 10588060.0,
      "step": 6025
    },
    {
      "entropy": 5.124079751968384,
      "epoch": 4.725705329153605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003254609212136108,
      "loss": 4.7834,
      "mean_token_accuracy": 0.2007227286696434,
      "num_tokens": 10596482.0,
      "step": 6030
    },
    {
      "entropy": 5.19301061630249,
      "epoch": 4.72962382445141,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003251677475059721,
      "loss": 4.8014,
      "mean_token_accuracy": 0.19893116503953934,
      "num_tokens": 10605630.0,
      "step": 6035
    },
    {
      "entropy": 5.239860773086548,
      "epoch": 4.733542319749216,
      "grad_norm": 1.25,
      "learning_rate": 0.00032487448414043435,
      "loss": 4.7926,
      "mean_token_accuracy": 0.20622722059488297,
      "num_tokens": 10614772.0,
      "step": 6040
    },
    {
      "entropy": 5.2177345752716064,
      "epoch": 4.737460815047022,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032458113164110665,
      "loss": 4.7594,
      "mean_token_accuracy": 0.211601360142231,
      "num_tokens": 10623583.0,
      "step": 6045
    },
    {
      "entropy": 5.164374828338623,
      "epoch": 4.741379310344827,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003242876905322575,
      "loss": 4.7221,
      "mean_token_accuracy": 0.21500681787729264,
      "num_tokens": 10632136.0,
      "step": 6050
    },
    {
      "entropy": 5.221694755554199,
      "epoch": 4.745297805642633,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003239941613383138,
      "loss": 4.8312,
      "mean_token_accuracy": 0.1959396034479141,
      "num_tokens": 10641306.0,
      "step": 6055
    },
    {
      "entropy": 5.210137796401978,
      "epoch": 4.749216300940439,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032370054458385973,
      "loss": 4.7842,
      "mean_token_accuracy": 0.20270317792892456,
      "num_tokens": 10650866.0,
      "step": 6060
    },
    {
      "entropy": 5.249972105026245,
      "epoch": 4.753134796238244,
      "grad_norm": 1.265625,
      "learning_rate": 0.000323406840793636,
      "loss": 4.7874,
      "mean_token_accuracy": 0.21086941212415694,
      "num_tokens": 10659252.0,
      "step": 6065
    },
    {
      "entropy": 5.192861652374267,
      "epoch": 4.75705329153605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000323113050492539,
      "loss": 4.677,
      "mean_token_accuracy": 0.20288807153701782,
      "num_tokens": 10667987.0,
      "step": 6070
    },
    {
      "entropy": 5.189536762237549,
      "epoch": 4.7609717868338555,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032281917420561967,
      "loss": 4.8606,
      "mean_token_accuracy": 0.19448111653327943,
      "num_tokens": 10676524.0,
      "step": 6075
    },
    {
      "entropy": 5.181484365463257,
      "epoch": 4.764890282131661,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003225252124580825,
      "loss": 4.7868,
      "mean_token_accuracy": 0.2006719544529915,
      "num_tokens": 10685940.0,
      "step": 6080
    },
    {
      "entropy": 5.230218982696533,
      "epoch": 4.768808777429467,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032223116577528475,
      "loss": 4.7867,
      "mean_token_accuracy": 0.20605482459068297,
      "num_tokens": 10694522.0,
      "step": 6085
    },
    {
      "entropy": 5.217789554595948,
      "epoch": 4.7727272727272725,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003219370346827356,
      "loss": 4.7164,
      "mean_token_accuracy": 0.205253966152668,
      "num_tokens": 10703868.0,
      "step": 6090
    },
    {
      "entropy": 5.1508471965789795,
      "epoch": 4.7766457680250785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000321642819706095,
      "loss": 4.7281,
      "mean_token_accuracy": 0.20630497932434083,
      "num_tokens": 10712044.0,
      "step": 6095
    },
    {
      "entropy": 5.295997190475464,
      "epoch": 4.7805642633228835,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003213485213711729,
      "loss": 4.9036,
      "mean_token_accuracy": 0.19497592896223068,
      "num_tokens": 10720806.0,
      "step": 6100
    },
    {
      "entropy": 5.149513578414917,
      "epoch": 4.7844827586206895,
      "grad_norm": 1.28125,
      "learning_rate": 0.000321054140203928,
      "loss": 4.6396,
      "mean_token_accuracy": 0.2187992066144943,
      "num_tokens": 10729238.0,
      "step": 6105
    },
    {
      "entropy": 5.117345952987671,
      "epoch": 4.7884012539184955,
      "grad_norm": 1.25,
      "learning_rate": 0.0003207596767304672,
      "loss": 4.7131,
      "mean_token_accuracy": 0.20769082009792328,
      "num_tokens": 10738175.0,
      "step": 6110
    },
    {
      "entropy": 5.244709253311157,
      "epoch": 4.7923197492163006,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032046513147704456,
      "loss": 4.8043,
      "mean_token_accuracy": 0.19995521306991576,
      "num_tokens": 10747475.0,
      "step": 6115
    },
    {
      "entropy": 5.216456365585327,
      "epoch": 4.7962382445141065,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032017050497006,
      "loss": 4.7597,
      "mean_token_accuracy": 0.2049367159605026,
      "num_tokens": 10756739.0,
      "step": 6120
    },
    {
      "entropy": 5.139582633972168,
      "epoch": 4.8001567398119125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031987579773605913,
      "loss": 4.7617,
      "mean_token_accuracy": 0.20457471609115602,
      "num_tokens": 10765141.0,
      "step": 6125
    },
    {
      "entropy": 5.21734356880188,
      "epoch": 4.804075235109718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003195810103017314,
      "loss": 4.8477,
      "mean_token_accuracy": 0.19864890724420547,
      "num_tokens": 10774759.0,
      "step": 6130
    },
    {
      "entropy": 5.231565618515015,
      "epoch": 4.8079937304075235,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003192861431939098,
      "loss": 4.8218,
      "mean_token_accuracy": 0.20146313458681106,
      "num_tokens": 10783717.0,
      "step": 6135
    },
    {
      "entropy": 5.184237909317017,
      "epoch": 4.8119122257053295,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003189911969395697,
      "loss": 4.7988,
      "mean_token_accuracy": 0.20226143151521683,
      "num_tokens": 10792797.0,
      "step": 6140
    },
    {
      "entropy": 5.097917079925537,
      "epoch": 4.815830721003135,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031869617206582777,
      "loss": 4.6654,
      "mean_token_accuracy": 0.21379477232694627,
      "num_tokens": 10802249.0,
      "step": 6145
    },
    {
      "entropy": 5.156694746017456,
      "epoch": 4.8197492163009406,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003184010690999413,
      "loss": 4.7751,
      "mean_token_accuracy": 0.20006383806467057,
      "num_tokens": 10810722.0,
      "step": 6150
    },
    {
      "entropy": 5.265868663787842,
      "epoch": 4.8236677115987465,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003181058885693073,
      "loss": 4.8252,
      "mean_token_accuracy": 0.199437914788723,
      "num_tokens": 10819115.0,
      "step": 6155
    },
    {
      "entropy": 5.257183170318603,
      "epoch": 4.827586206896552,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003178106310014612,
      "loss": 4.8359,
      "mean_token_accuracy": 0.1961295336484909,
      "num_tokens": 10827419.0,
      "step": 6160
    },
    {
      "entropy": 5.125362539291382,
      "epoch": 4.831504702194358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003175152969240759,
      "loss": 4.7109,
      "mean_token_accuracy": 0.2070888176560402,
      "num_tokens": 10836610.0,
      "step": 6165
    },
    {
      "entropy": 5.182009649276734,
      "epoch": 4.835423197492163,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031721988686496156,
      "loss": 4.793,
      "mean_token_accuracy": 0.20870455205440522,
      "num_tokens": 10845127.0,
      "step": 6170
    },
    {
      "entropy": 5.22943959236145,
      "epoch": 4.839341692789969,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003169244013520638,
      "loss": 4.7808,
      "mean_token_accuracy": 0.20789557695388794,
      "num_tokens": 10853458.0,
      "step": 6175
    },
    {
      "entropy": 5.2488127708435055,
      "epoch": 4.843260188087775,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000316628840913463,
      "loss": 4.8979,
      "mean_token_accuracy": 0.19310447275638581,
      "num_tokens": 10862134.0,
      "step": 6180
    },
    {
      "entropy": 5.238654947280883,
      "epoch": 4.84717868338558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003163332060773738,
      "loss": 4.9089,
      "mean_token_accuracy": 0.19707036167383193,
      "num_tokens": 10872127.0,
      "step": 6185
    },
    {
      "entropy": 5.217839908599854,
      "epoch": 4.851097178683386,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003160374973721434,
      "loss": 4.8143,
      "mean_token_accuracy": 0.20148200541734695,
      "num_tokens": 10880617.0,
      "step": 6190
    },
    {
      "entropy": 5.224712800979614,
      "epoch": 4.855015673981192,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003157417153262513,
      "loss": 4.7822,
      "mean_token_accuracy": 0.2006895959377289,
      "num_tokens": 10888769.0,
      "step": 6195
    },
    {
      "entropy": 5.204051113128662,
      "epoch": 4.858934169278997,
      "grad_norm": 1.25,
      "learning_rate": 0.00031544586046830796,
      "loss": 4.7578,
      "mean_token_accuracy": 0.20031883716583251,
      "num_tokens": 10897638.0,
      "step": 6200
    },
    {
      "entropy": 5.199450588226318,
      "epoch": 4.862852664576803,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000315149933327054,
      "loss": 4.7039,
      "mean_token_accuracy": 0.20880066156387328,
      "num_tokens": 10906224.0,
      "step": 6205
    },
    {
      "entropy": 5.102503061294556,
      "epoch": 4.866771159874608,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031485393443135914,
      "loss": 4.711,
      "mean_token_accuracy": 0.21036999374628068,
      "num_tokens": 10915390.0,
      "step": 6210
    },
    {
      "entropy": 5.222969579696655,
      "epoch": 4.870689655172414,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003145578643102214,
      "loss": 4.7993,
      "mean_token_accuracy": 0.19999231696128844,
      "num_tokens": 10923719.0,
      "step": 6215
    },
    {
      "entropy": 5.190935707092285,
      "epoch": 4.87460815047022,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003142617234927662,
      "loss": 4.7413,
      "mean_token_accuracy": 0.203766268491745,
      "num_tokens": 10932610.0,
      "step": 6220
    },
    {
      "entropy": 5.227710103988647,
      "epoch": 4.878526645768025,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031396551250824513,
      "loss": 4.7305,
      "mean_token_accuracy": 0.20592274218797685,
      "num_tokens": 10941315.0,
      "step": 6225
    },
    {
      "entropy": 5.075382804870605,
      "epoch": 4.882445141065831,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003136692318860352,
      "loss": 4.6942,
      "mean_token_accuracy": 0.21001509130001067,
      "num_tokens": 10949781.0,
      "step": 6230
    },
    {
      "entropy": 5.228602743148803,
      "epoch": 4.886363636363637,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003133728821556381,
      "loss": 4.8794,
      "mean_token_accuracy": 0.1993560090661049,
      "num_tokens": 10958207.0,
      "step": 6235
    },
    {
      "entropy": 5.194638681411743,
      "epoch": 4.890282131661442,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031307646384667854,
      "loss": 4.7789,
      "mean_token_accuracy": 0.19304061233997344,
      "num_tokens": 10967162.0,
      "step": 6240
    },
    {
      "entropy": 5.247672271728516,
      "epoch": 4.894200626959248,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031277997748890433,
      "loss": 4.8651,
      "mean_token_accuracy": 0.19524385035037994,
      "num_tokens": 10975517.0,
      "step": 6245
    },
    {
      "entropy": 5.192938327789307,
      "epoch": 4.898119122257054,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003124834236121847,
      "loss": 4.7313,
      "mean_token_accuracy": 0.20214771032333373,
      "num_tokens": 10984668.0,
      "step": 6250
    },
    {
      "entropy": 5.120031833648682,
      "epoch": 4.902037617554859,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031218680274650934,
      "loss": 4.7348,
      "mean_token_accuracy": 0.1996182084083557,
      "num_tokens": 10993475.0,
      "step": 6255
    },
    {
      "entropy": 5.105472803115845,
      "epoch": 4.905956112852665,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031189011542198794,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2163337990641594,
      "num_tokens": 11002419.0,
      "step": 6260
    },
    {
      "entropy": 5.130008935928345,
      "epoch": 4.90987460815047,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003115933621688488,
      "loss": 4.7531,
      "mean_token_accuracy": 0.20833301693201065,
      "num_tokens": 11011271.0,
      "step": 6265
    },
    {
      "entropy": 5.249984979629517,
      "epoch": 4.913793103448276,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031129654351743816,
      "loss": 4.796,
      "mean_token_accuracy": 0.19797551929950713,
      "num_tokens": 11020461.0,
      "step": 6270
    },
    {
      "entropy": 5.2206583499908445,
      "epoch": 4.917711598746082,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031099965999821906,
      "loss": 4.7076,
      "mean_token_accuracy": 0.20835764557123185,
      "num_tokens": 11029251.0,
      "step": 6275
    },
    {
      "entropy": 5.2054595947265625,
      "epoch": 4.921630094043887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003107027121417704,
      "loss": 4.7165,
      "mean_token_accuracy": 0.21009994745254518,
      "num_tokens": 11038149.0,
      "step": 6280
    },
    {
      "entropy": 5.22030816078186,
      "epoch": 4.925548589341693,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031040570047878634,
      "loss": 4.7967,
      "mean_token_accuracy": 0.20408463776111602,
      "num_tokens": 11046262.0,
      "step": 6285
    },
    {
      "entropy": 5.239262056350708,
      "epoch": 4.929467084639499,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031010862554007473,
      "loss": 4.8428,
      "mean_token_accuracy": 0.19639027267694473,
      "num_tokens": 11055755.0,
      "step": 6290
    },
    {
      "entropy": 5.141078281402588,
      "epoch": 4.933385579937304,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003098114878565567,
      "loss": 4.7642,
      "mean_token_accuracy": 0.20326332300901412,
      "num_tokens": 11064534.0,
      "step": 6295
    },
    {
      "entropy": 5.185782194137573,
      "epoch": 4.93730407523511,
      "grad_norm": 1.25,
      "learning_rate": 0.00030951428795926536,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2026752546429634,
      "num_tokens": 11073768.0,
      "step": 6300
    },
    {
      "entropy": 5.190009880065918,
      "epoch": 4.941222570532915,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003092170263793452,
      "loss": 4.7823,
      "mean_token_accuracy": 0.20681187212467195,
      "num_tokens": 11083683.0,
      "step": 6305
    },
    {
      "entropy": 5.224541282653808,
      "epoch": 4.945141065830721,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030891970364805096,
      "loss": 4.8471,
      "mean_token_accuracy": 0.1998303085565567,
      "num_tokens": 11092368.0,
      "step": 6310
    },
    {
      "entropy": 5.155054664611816,
      "epoch": 4.949059561128527,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003086223202967464,
      "loss": 4.7262,
      "mean_token_accuracy": 0.20833810567855834,
      "num_tokens": 11100544.0,
      "step": 6315
    },
    {
      "entropy": 5.192977857589722,
      "epoch": 4.952978056426332,
      "grad_norm": 1.34375,
      "learning_rate": 0.000308324876856904,
      "loss": 4.7722,
      "mean_token_accuracy": 0.2010019913315773,
      "num_tokens": 11108962.0,
      "step": 6320
    },
    {
      "entropy": 5.265523529052734,
      "epoch": 4.956896551724138,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030802737386010314,
      "loss": 4.8341,
      "mean_token_accuracy": 0.20120722949504852,
      "num_tokens": 11117011.0,
      "step": 6325
    },
    {
      "entropy": 5.203204917907715,
      "epoch": 4.960815047021944,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003077298118380301,
      "loss": 4.7898,
      "mean_token_accuracy": 0.2047535538673401,
      "num_tokens": 11126009.0,
      "step": 6330
    },
    {
      "entropy": 5.193589115142823,
      "epoch": 4.964733542319749,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003074321913224765,
      "loss": 4.668,
      "mean_token_accuracy": 0.20852853804826738,
      "num_tokens": 11134404.0,
      "step": 6335
    },
    {
      "entropy": 5.244256019592285,
      "epoch": 4.968652037617555,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003071345128453382,
      "loss": 4.8137,
      "mean_token_accuracy": 0.2018840491771698,
      "num_tokens": 11142245.0,
      "step": 6340
    },
    {
      "entropy": 5.1453667163848875,
      "epoch": 4.972570532915361,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003068367769386153,
      "loss": 4.7879,
      "mean_token_accuracy": 0.20511994063854216,
      "num_tokens": 11150881.0,
      "step": 6345
    },
    {
      "entropy": 5.135769939422607,
      "epoch": 4.976489028213166,
      "grad_norm": 1.25,
      "learning_rate": 0.00030653898413440973,
      "loss": 4.7301,
      "mean_token_accuracy": 0.2066554382443428,
      "num_tokens": 11159421.0,
      "step": 6350
    },
    {
      "entropy": 5.190818357467651,
      "epoch": 4.980407523510972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003062411349649258,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2039031893014908,
      "num_tokens": 11168633.0,
      "step": 6355
    },
    {
      "entropy": 5.1425763130187985,
      "epoch": 4.984326018808778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030594322996246816,
      "loss": 4.6476,
      "mean_token_accuracy": 0.21405357420444487,
      "num_tokens": 11177435.0,
      "step": 6360
    },
    {
      "entropy": 5.225773048400879,
      "epoch": 4.988244514106583,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030564526965944127,
      "loss": 4.8873,
      "mean_token_accuracy": 0.19553401172161103,
      "num_tokens": 11186195.0,
      "step": 6365
    },
    {
      "entropy": 5.1790283203125,
      "epoch": 4.992163009404389,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003053472545883488,
      "loss": 4.7051,
      "mean_token_accuracy": 0.2087102323770523,
      "num_tokens": 11194087.0,
      "step": 6370
    },
    {
      "entropy": 5.201116228103638,
      "epoch": 4.996081504702194,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030504918528179164,
      "loss": 4.7901,
      "mean_token_accuracy": 0.20314509570598602,
      "num_tokens": 11202677.0,
      "step": 6375
    },
    {
      "entropy": 5.204102659225464,
      "epoch": 5.0,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030475106227246824,
      "loss": 4.8149,
      "mean_token_accuracy": 0.1943760871887207,
      "num_tokens": 11211560.0,
      "step": 6380
    },
    {
      "entropy": 5.154899835586548,
      "epoch": 5.003918495297806,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003044528860931726,
      "loss": 4.531,
      "mean_token_accuracy": 0.2282404899597168,
      "num_tokens": 11220186.0,
      "step": 6385
    },
    {
      "entropy": 5.212333393096924,
      "epoch": 5.007836990595611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003041546572767939,
      "loss": 4.556,
      "mean_token_accuracy": 0.21682643443346022,
      "num_tokens": 11229148.0,
      "step": 6390
    },
    {
      "entropy": 5.180864048004151,
      "epoch": 5.011755485893417,
      "grad_norm": 1.375,
      "learning_rate": 0.0003038563763563156,
      "loss": 4.4908,
      "mean_token_accuracy": 0.23659077882766724,
      "num_tokens": 11237585.0,
      "step": 6395
    },
    {
      "entropy": 5.137466716766357,
      "epoch": 5.015673981191223,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030355804386481374,
      "loss": 4.445,
      "mean_token_accuracy": 0.22268743813037872,
      "num_tokens": 11246203.0,
      "step": 6400
    },
    {
      "entropy": 5.18770751953125,
      "epoch": 5.019592476489028,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030325966033545716,
      "loss": 4.5714,
      "mean_token_accuracy": 0.21903689652681352,
      "num_tokens": 11254999.0,
      "step": 6405
    },
    {
      "entropy": 5.151151323318482,
      "epoch": 5.023510971786834,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030296122630150534,
      "loss": 4.5089,
      "mean_token_accuracy": 0.2254614621400833,
      "num_tokens": 11264110.0,
      "step": 6410
    },
    {
      "entropy": 5.1392223834991455,
      "epoch": 5.027429467084639,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030266274229630844,
      "loss": 4.4521,
      "mean_token_accuracy": 0.23787928223609925,
      "num_tokens": 11273452.0,
      "step": 6415
    },
    {
      "entropy": 5.1035181999206545,
      "epoch": 5.031347962382445,
      "grad_norm": 1.375,
      "learning_rate": 0.00030236420885330586,
      "loss": 4.4611,
      "mean_token_accuracy": 0.2261319100856781,
      "num_tokens": 11281926.0,
      "step": 6420
    },
    {
      "entropy": 5.188841247558594,
      "epoch": 5.035266457680251,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003020656265060251,
      "loss": 4.535,
      "mean_token_accuracy": 0.22770574390888215,
      "num_tokens": 11290630.0,
      "step": 6425
    },
    {
      "entropy": 5.215688943862915,
      "epoch": 5.039184952978056,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003017669957880813,
      "loss": 4.5484,
      "mean_token_accuracy": 0.2218847393989563,
      "num_tokens": 11299126.0,
      "step": 6430
    },
    {
      "entropy": 5.188980531692505,
      "epoch": 5.043103448275862,
      "grad_norm": 1.3125,
      "learning_rate": 0.000301468317233176,
      "loss": 4.6266,
      "mean_token_accuracy": 0.21058132797479628,
      "num_tokens": 11308331.0,
      "step": 6435
    },
    {
      "entropy": 5.168396615982056,
      "epoch": 5.047021943573668,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030116959137509614,
      "loss": 4.5199,
      "mean_token_accuracy": 0.2172871857881546,
      "num_tokens": 11316245.0,
      "step": 6440
    },
    {
      "entropy": 5.177174758911133,
      "epoch": 5.050940438871473,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030087081874771325,
      "loss": 4.5142,
      "mean_token_accuracy": 0.21957987844944,
      "num_tokens": 11324994.0,
      "step": 6445
    },
    {
      "entropy": 5.106066465377808,
      "epoch": 5.054858934169279,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030057199988498245,
      "loss": 4.4695,
      "mean_token_accuracy": 0.22277211397886276,
      "num_tokens": 11333255.0,
      "step": 6450
    },
    {
      "entropy": 5.106308031082153,
      "epoch": 5.058777429467085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003002731353209416,
      "loss": 4.5079,
      "mean_token_accuracy": 0.22293618470430374,
      "num_tokens": 11341857.0,
      "step": 6455
    },
    {
      "entropy": 5.166507863998413,
      "epoch": 5.06269592476489,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002999742255897102,
      "loss": 4.6156,
      "mean_token_accuracy": 0.21915097981691362,
      "num_tokens": 11350933.0,
      "step": 6460
    },
    {
      "entropy": 5.157345485687256,
      "epoch": 5.066614420062696,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029967527122548826,
      "loss": 4.5076,
      "mean_token_accuracy": 0.21756283789873124,
      "num_tokens": 11360346.0,
      "step": 6465
    },
    {
      "entropy": 5.173162269592285,
      "epoch": 5.070532915360501,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002993762727625556,
      "loss": 4.5656,
      "mean_token_accuracy": 0.22167657315731049,
      "num_tokens": 11369401.0,
      "step": 6470
    },
    {
      "entropy": 5.149011516571045,
      "epoch": 5.074451410658307,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002990772307352712,
      "loss": 4.517,
      "mean_token_accuracy": 0.21981369853019714,
      "num_tokens": 11377882.0,
      "step": 6475
    },
    {
      "entropy": 5.130825185775757,
      "epoch": 5.078369905956113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002987781456780715,
      "loss": 4.5322,
      "mean_token_accuracy": 0.22399216741323472,
      "num_tokens": 11386934.0,
      "step": 6480
    },
    {
      "entropy": 5.119526624679565,
      "epoch": 5.082288401253918,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029847901812547,
      "loss": 4.5482,
      "mean_token_accuracy": 0.21710938960313797,
      "num_tokens": 11395929.0,
      "step": 6485
    },
    {
      "entropy": 5.164478731155396,
      "epoch": 5.086206896551724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029817984861205626,
      "loss": 4.5043,
      "mean_token_accuracy": 0.2229900598526001,
      "num_tokens": 11405664.0,
      "step": 6490
    },
    {
      "entropy": 5.164281606674194,
      "epoch": 5.09012539184953,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002978806376724945,
      "loss": 4.5929,
      "mean_token_accuracy": 0.22421531081199647,
      "num_tokens": 11414667.0,
      "step": 6495
    },
    {
      "entropy": 5.169739818572998,
      "epoch": 5.094043887147335,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029758138584152333,
      "loss": 4.5655,
      "mean_token_accuracy": 0.2156837910413742,
      "num_tokens": 11423907.0,
      "step": 6500
    },
    {
      "epoch": 5.094043887147335,
      "eval_entropy": 5.035524445910786,
      "eval_loss": 5.67020320892334,
      "eval_mean_token_accuracy": 0.17033282251552093,
      "eval_num_tokens": 11423907.0,
      "eval_runtime": 2.854,
      "eval_samples_per_second": 1444.304,
      "eval_steps_per_second": 180.801,
      "step": 6500
    },
    {
      "entropy": 5.1778112888336185,
      "epoch": 5.097962382445141,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029728209365395433,
      "loss": 4.5913,
      "mean_token_accuracy": 0.22655981183052062,
      "num_tokens": 11433824.0,
      "step": 6505
    },
    {
      "entropy": 5.176232147216797,
      "epoch": 5.101880877742946,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029698276164467105,
      "loss": 4.4814,
      "mean_token_accuracy": 0.2242218405008316,
      "num_tokens": 11442371.0,
      "step": 6510
    },
    {
      "entropy": 5.210544586181641,
      "epoch": 5.105799373040752,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002966833903486284,
      "loss": 4.5073,
      "mean_token_accuracy": 0.2194698229432106,
      "num_tokens": 11451657.0,
      "step": 6515
    },
    {
      "entropy": 5.080518531799316,
      "epoch": 5.109717868338558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029638398030085134,
      "loss": 4.4529,
      "mean_token_accuracy": 0.22430081963539122,
      "num_tokens": 11460064.0,
      "step": 6520
    },
    {
      "entropy": 5.085877704620361,
      "epoch": 5.113636363636363,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029608453203643434,
      "loss": 4.5618,
      "mean_token_accuracy": 0.21461535543203353,
      "num_tokens": 11469184.0,
      "step": 6525
    },
    {
      "entropy": 5.199899053573608,
      "epoch": 5.117554858934169,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002957850460905398,
      "loss": 4.4839,
      "mean_token_accuracy": 0.23056766092777253,
      "num_tokens": 11478069.0,
      "step": 6530
    },
    {
      "entropy": 5.1898736476898195,
      "epoch": 5.121473354231975,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029548552299839774,
      "loss": 4.5183,
      "mean_token_accuracy": 0.22824599146842955,
      "num_tokens": 11486572.0,
      "step": 6535
    },
    {
      "entropy": 5.131806707382202,
      "epoch": 5.12539184952978,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002951859632953046,
      "loss": 4.5626,
      "mean_token_accuracy": 0.2170782282948494,
      "num_tokens": 11495857.0,
      "step": 6540
    },
    {
      "entropy": 5.124344348907471,
      "epoch": 5.129310344827586,
      "grad_norm": 1.25,
      "learning_rate": 0.00029488636751662196,
      "loss": 4.5487,
      "mean_token_accuracy": 0.22196327298879623,
      "num_tokens": 11504875.0,
      "step": 6545
    },
    {
      "entropy": 5.1579893112182615,
      "epoch": 5.133228840125392,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002945867361977762,
      "loss": 4.5475,
      "mean_token_accuracy": 0.21765585392713546,
      "num_tokens": 11514509.0,
      "step": 6550
    },
    {
      "entropy": 5.15773377418518,
      "epoch": 5.137147335423197,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029428706987425705,
      "loss": 4.5819,
      "mean_token_accuracy": 0.21433132886886597,
      "num_tokens": 11523046.0,
      "step": 6555
    },
    {
      "entropy": 5.156280374526977,
      "epoch": 5.141065830721003,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002939873690816167,
      "loss": 4.4813,
      "mean_token_accuracy": 0.22801015228033067,
      "num_tokens": 11531270.0,
      "step": 6560
    },
    {
      "entropy": 5.173916006088257,
      "epoch": 5.144984326018808,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002936876343554692,
      "loss": 4.5315,
      "mean_token_accuracy": 0.22209583967924118,
      "num_tokens": 11539722.0,
      "step": 6565
    },
    {
      "entropy": 5.136562013626099,
      "epoch": 5.148902821316614,
      "grad_norm": 1.3125,
      "learning_rate": 0.000293387866231489,
      "loss": 4.5801,
      "mean_token_accuracy": 0.2217647135257721,
      "num_tokens": 11549614.0,
      "step": 6570
    },
    {
      "entropy": 5.095829820632934,
      "epoch": 5.15282131661442,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029308806524541047,
      "loss": 4.4162,
      "mean_token_accuracy": 0.2249612733721733,
      "num_tokens": 11557674.0,
      "step": 6575
    },
    {
      "entropy": 5.121955013275146,
      "epoch": 5.156739811912225,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029278823193302643,
      "loss": 4.4983,
      "mean_token_accuracy": 0.21971090584993364,
      "num_tokens": 11566691.0,
      "step": 6580
    },
    {
      "entropy": 5.157381772994995,
      "epoch": 5.160658307210031,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029248836683018774,
      "loss": 4.475,
      "mean_token_accuracy": 0.22810071408748628,
      "num_tokens": 11574824.0,
      "step": 6585
    },
    {
      "entropy": 5.081808376312256,
      "epoch": 5.164576802507837,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029218847047280197,
      "loss": 4.4903,
      "mean_token_accuracy": 0.22595936954021453,
      "num_tokens": 11582988.0,
      "step": 6590
    },
    {
      "entropy": 5.125911521911621,
      "epoch": 5.168495297805642,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002918885433968323,
      "loss": 4.5663,
      "mean_token_accuracy": 0.21752714216709138,
      "num_tokens": 11591553.0,
      "step": 6595
    },
    {
      "entropy": 5.115233993530273,
      "epoch": 5.172413793103448,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002915885861382974,
      "loss": 4.5639,
      "mean_token_accuracy": 0.2152484744787216,
      "num_tokens": 11600347.0,
      "step": 6600
    },
    {
      "entropy": 5.178760385513305,
      "epoch": 5.1763322884012535,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029128859923326935,
      "loss": 4.5899,
      "mean_token_accuracy": 0.22104137390851974,
      "num_tokens": 11609217.0,
      "step": 6605
    },
    {
      "entropy": 5.188550996780395,
      "epoch": 5.1802507836990594,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029098858321787336,
      "loss": 4.5628,
      "mean_token_accuracy": 0.22103688716888428,
      "num_tokens": 11617355.0,
      "step": 6610
    },
    {
      "entropy": 5.153822565078736,
      "epoch": 5.184169278996865,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029068853862828677,
      "loss": 4.5273,
      "mean_token_accuracy": 0.22460360676050187,
      "num_tokens": 11626152.0,
      "step": 6615
    },
    {
      "entropy": 5.103583812713623,
      "epoch": 5.1880877742946705,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002903884660007378,
      "loss": 4.5066,
      "mean_token_accuracy": 0.21820321083068847,
      "num_tokens": 11634868.0,
      "step": 6620
    },
    {
      "entropy": 5.119492769241333,
      "epoch": 5.1920062695924765,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029008836587150505,
      "loss": 4.6023,
      "mean_token_accuracy": 0.21758635044097902,
      "num_tokens": 11644199.0,
      "step": 6625
    },
    {
      "entropy": 5.22194538116455,
      "epoch": 5.195924764890282,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002897882387769159,
      "loss": 4.6416,
      "mean_token_accuracy": 0.21393326222896575,
      "num_tokens": 11653227.0,
      "step": 6630
    },
    {
      "entropy": 5.101305913925171,
      "epoch": 5.1998432601880875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002894880852533464,
      "loss": 4.4465,
      "mean_token_accuracy": 0.23457471579313277,
      "num_tokens": 11661628.0,
      "step": 6635
    },
    {
      "entropy": 5.1236889362335205,
      "epoch": 5.2037617554858935,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028918790583721936,
      "loss": 4.5054,
      "mean_token_accuracy": 0.22735761404037474,
      "num_tokens": 11670616.0,
      "step": 6640
    },
    {
      "entropy": 5.065277767181397,
      "epoch": 5.2076802507836994,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002888877010650041,
      "loss": 4.5189,
      "mean_token_accuracy": 0.22657260149717331,
      "num_tokens": 11679069.0,
      "step": 6645
    },
    {
      "entropy": 5.163487434387207,
      "epoch": 5.2115987460815045,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028858747147321527,
      "loss": 4.4979,
      "mean_token_accuracy": 0.22687303125858307,
      "num_tokens": 11688184.0,
      "step": 6650
    },
    {
      "entropy": 5.121678686141967,
      "epoch": 5.2155172413793105,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002882872175984118,
      "loss": 4.5607,
      "mean_token_accuracy": 0.21819736510515214,
      "num_tokens": 11696389.0,
      "step": 6655
    },
    {
      "entropy": 5.099625778198242,
      "epoch": 5.219435736677116,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002879869399771962,
      "loss": 4.584,
      "mean_token_accuracy": 0.21435199528932572,
      "num_tokens": 11704881.0,
      "step": 6660
    },
    {
      "entropy": 5.174586153030395,
      "epoch": 5.2233542319749215,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028768663914621295,
      "loss": 4.6174,
      "mean_token_accuracy": 0.21966575533151628,
      "num_tokens": 11713566.0,
      "step": 6665
    },
    {
      "entropy": 5.165548419952392,
      "epoch": 5.2272727272727275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002873863156421486,
      "loss": 4.5766,
      "mean_token_accuracy": 0.2228282243013382,
      "num_tokens": 11722427.0,
      "step": 6670
    },
    {
      "entropy": 5.178943300247193,
      "epoch": 5.231191222570533,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028708597000172984,
      "loss": 4.537,
      "mean_token_accuracy": 0.22365642338991165,
      "num_tokens": 11731010.0,
      "step": 6675
    },
    {
      "entropy": 5.102612018585205,
      "epoch": 5.235109717868339,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002867856027617229,
      "loss": 4.5701,
      "mean_token_accuracy": 0.21524299830198287,
      "num_tokens": 11739569.0,
      "step": 6680
    },
    {
      "entropy": 5.131764888763428,
      "epoch": 5.2390282131661445,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028648521445893304,
      "loss": 4.511,
      "mean_token_accuracy": 0.22026837319135667,
      "num_tokens": 11747687.0,
      "step": 6685
    },
    {
      "entropy": 5.134809732437134,
      "epoch": 5.24294670846395,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002861848056302026,
      "loss": 4.5897,
      "mean_token_accuracy": 0.22363103926181793,
      "num_tokens": 11756713.0,
      "step": 6690
    },
    {
      "entropy": 5.135261392593383,
      "epoch": 5.246865203761756,
      "grad_norm": 1.25,
      "learning_rate": 0.00028588437681241106,
      "loss": 4.5013,
      "mean_token_accuracy": 0.22841547280550004,
      "num_tokens": 11765625.0,
      "step": 6695
    },
    {
      "entropy": 5.17015643119812,
      "epoch": 5.250783699059561,
      "grad_norm": 1.25,
      "learning_rate": 0.00028558392854247333,
      "loss": 4.5977,
      "mean_token_accuracy": 0.21586582213640212,
      "num_tokens": 11774594.0,
      "step": 6700
    },
    {
      "entropy": 5.121698808670044,
      "epoch": 5.254702194357367,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002852834613573391,
      "loss": 4.5375,
      "mean_token_accuracy": 0.23027369230985642,
      "num_tokens": 11782983.0,
      "step": 6705
    },
    {
      "entropy": 5.094813823699951,
      "epoch": 5.258620689655173,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002849829757939922,
      "loss": 4.5169,
      "mean_token_accuracy": 0.22302103638648987,
      "num_tokens": 11791940.0,
      "step": 6710
    },
    {
      "entropy": 5.18468976020813,
      "epoch": 5.262539184952978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002846824723894487,
      "loss": 4.5443,
      "mean_token_accuracy": 0.21930090934038163,
      "num_tokens": 11800357.0,
      "step": 6715
    },
    {
      "entropy": 5.1937497615814205,
      "epoch": 5.266457680250784,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002843819516807572,
      "loss": 4.6034,
      "mean_token_accuracy": 0.21227792054414749,
      "num_tokens": 11809005.0,
      "step": 6720
    },
    {
      "entropy": 5.181586456298828,
      "epoch": 5.27037617554859,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028408141420499685,
      "loss": 4.5508,
      "mean_token_accuracy": 0.21465314030647278,
      "num_tokens": 11818021.0,
      "step": 6725
    },
    {
      "entropy": 5.1822703838348385,
      "epoch": 5.274294670846395,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002837808604992768,
      "loss": 4.5627,
      "mean_token_accuracy": 0.2147682264447212,
      "num_tokens": 11826791.0,
      "step": 6730
    },
    {
      "entropy": 5.090599107742309,
      "epoch": 5.278213166144201,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028348029110073533,
      "loss": 4.472,
      "mean_token_accuracy": 0.2284413293004036,
      "num_tokens": 11835110.0,
      "step": 6735
    },
    {
      "entropy": 5.125855779647827,
      "epoch": 5.282131661442007,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002831797065465386,
      "loss": 4.6032,
      "mean_token_accuracy": 0.21895151436328888,
      "num_tokens": 11842658.0,
      "step": 6740
    },
    {
      "entropy": 5.110217618942261,
      "epoch": 5.286050156739812,
      "grad_norm": 1.375,
      "learning_rate": 0.00028287910737388,
      "loss": 4.5759,
      "mean_token_accuracy": 0.21876311749219896,
      "num_tokens": 11851028.0,
      "step": 6745
    },
    {
      "entropy": 5.153445434570313,
      "epoch": 5.289968652037618,
      "grad_norm": 1.265625,
      "learning_rate": 0.000282578494119979,
      "loss": 4.5111,
      "mean_token_accuracy": 0.23051988631486892,
      "num_tokens": 11859471.0,
      "step": 6750
    },
    {
      "entropy": 5.120153617858887,
      "epoch": 5.293887147335424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028227786732208014,
      "loss": 4.4674,
      "mean_token_accuracy": 0.222701957821846,
      "num_tokens": 11868625.0,
      "step": 6755
    },
    {
      "entropy": 5.090478944778442,
      "epoch": 5.297805642633229,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028197722751745246,
      "loss": 4.5225,
      "mean_token_accuracy": 0.21772418171167374,
      "num_tokens": 11877744.0,
      "step": 6760
    },
    {
      "entropy": 5.110264873504638,
      "epoch": 5.301724137931035,
      "grad_norm": 1.375,
      "learning_rate": 0.0002816765752433879,
      "loss": 4.4944,
      "mean_token_accuracy": 0.22541847229003906,
      "num_tokens": 11886445.0,
      "step": 6765
    },
    {
      "entropy": 5.059976434707641,
      "epoch": 5.30564263322884,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028137591103720075,
      "loss": 4.5257,
      "mean_token_accuracy": 0.22091928869485855,
      "num_tokens": 11896040.0,
      "step": 6770
    },
    {
      "entropy": 5.08160252571106,
      "epoch": 5.309561128526646,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002810752354362267,
      "loss": 4.5598,
      "mean_token_accuracy": 0.210668183863163,
      "num_tokens": 11904996.0,
      "step": 6775
    },
    {
      "entropy": 5.180131530761718,
      "epoch": 5.313479623824452,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000280774548977822,
      "loss": 4.5475,
      "mean_token_accuracy": 0.22187484353780745,
      "num_tokens": 11913357.0,
      "step": 6780
    },
    {
      "entropy": 5.159757232666015,
      "epoch": 5.317398119122257,
      "grad_norm": 1.25,
      "learning_rate": 0.00028047385219936196,
      "loss": 4.5802,
      "mean_token_accuracy": 0.22327842116355895,
      "num_tokens": 11922196.0,
      "step": 6785
    },
    {
      "entropy": 5.087998962402343,
      "epoch": 5.321316614420063,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028017314563824044,
      "loss": 4.4488,
      "mean_token_accuracy": 0.22912188917398452,
      "num_tokens": 11930682.0,
      "step": 6790
    },
    {
      "entropy": 5.048047924041748,
      "epoch": 5.325235109717869,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027987242983186896,
      "loss": 4.5547,
      "mean_token_accuracy": 0.21946143805980683,
      "num_tokens": 11939470.0,
      "step": 6795
    },
    {
      "entropy": 5.110309028625489,
      "epoch": 5.329153605015674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027957170531767525,
      "loss": 4.5238,
      "mean_token_accuracy": 0.21987996697425843,
      "num_tokens": 11948183.0,
      "step": 6800
    },
    {
      "entropy": 5.2082374572753904,
      "epoch": 5.33307210031348,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027927097263310294,
      "loss": 4.5802,
      "mean_token_accuracy": 0.21753987818956375,
      "num_tokens": 11957027.0,
      "step": 6805
    },
    {
      "entropy": 5.103675174713135,
      "epoch": 5.336990595611285,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002789702323156099,
      "loss": 4.5366,
      "mean_token_accuracy": 0.21988280564546586,
      "num_tokens": 11965871.0,
      "step": 6810
    },
    {
      "entropy": 5.1487571716308596,
      "epoch": 5.340909090909091,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027866948490266815,
      "loss": 4.6474,
      "mean_token_accuracy": 0.21362998187541962,
      "num_tokens": 11974384.0,
      "step": 6815
    },
    {
      "entropy": 5.183628988265991,
      "epoch": 5.344827586206897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002783687309317618,
      "loss": 4.5589,
      "mean_token_accuracy": 0.21759943217039107,
      "num_tokens": 11983892.0,
      "step": 6820
    },
    {
      "entropy": 5.262387800216675,
      "epoch": 5.348746081504702,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002780679709403871,
      "loss": 4.607,
      "mean_token_accuracy": 0.2169151112437248,
      "num_tokens": 11992990.0,
      "step": 6825
    },
    {
      "entropy": 5.0500494003295895,
      "epoch": 5.352664576802508,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027776720546605086,
      "loss": 4.4095,
      "mean_token_accuracy": 0.23493760377168654,
      "num_tokens": 12001679.0,
      "step": 6830
    },
    {
      "entropy": 5.093603801727295,
      "epoch": 5.356583072100314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002774664350462697,
      "loss": 4.5783,
      "mean_token_accuracy": 0.21829910278320314,
      "num_tokens": 12011300.0,
      "step": 6835
    },
    {
      "entropy": 5.100832414627075,
      "epoch": 5.360501567398119,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027716566021856933,
      "loss": 4.5336,
      "mean_token_accuracy": 0.21962115168571472,
      "num_tokens": 12019776.0,
      "step": 6840
    },
    {
      "entropy": 5.124850225448609,
      "epoch": 5.364420062695925,
      "grad_norm": 1.171875,
      "learning_rate": 0.000276864881520483,
      "loss": 4.5669,
      "mean_token_accuracy": 0.22404860109090804,
      "num_tokens": 12028838.0,
      "step": 6845
    },
    {
      "entropy": 5.196340036392212,
      "epoch": 5.368338557993731,
      "grad_norm": 1.375,
      "learning_rate": 0.0002765640994895509,
      "loss": 4.5319,
      "mean_token_accuracy": 0.2215167835354805,
      "num_tokens": 12036903.0,
      "step": 6850
    },
    {
      "entropy": 5.132545232772827,
      "epoch": 5.372257053291536,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027626331466331955,
      "loss": 4.5141,
      "mean_token_accuracy": 0.2211013212800026,
      "num_tokens": 12045838.0,
      "step": 6855
    },
    {
      "entropy": 5.170269584655761,
      "epoch": 5.376175548589342,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027596252757933995,
      "loss": 4.6666,
      "mean_token_accuracy": 0.2095175564289093,
      "num_tokens": 12055791.0,
      "step": 6860
    },
    {
      "entropy": 5.091013860702515,
      "epoch": 5.380094043887147,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002756617387751674,
      "loss": 4.4652,
      "mean_token_accuracy": 0.225431190431118,
      "num_tokens": 12064842.0,
      "step": 6865
    },
    {
      "entropy": 5.148637580871582,
      "epoch": 5.384012539184953,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002753609487883606,
      "loss": 4.5325,
      "mean_token_accuracy": 0.22105031758546828,
      "num_tokens": 12073487.0,
      "step": 6870
    },
    {
      "entropy": 5.07039794921875,
      "epoch": 5.387931034482759,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027506015815647965,
      "loss": 4.5025,
      "mean_token_accuracy": 0.2194417655467987,
      "num_tokens": 12082000.0,
      "step": 6875
    },
    {
      "entropy": 5.119787979125976,
      "epoch": 5.391849529780564,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027475936741708636,
      "loss": 4.5473,
      "mean_token_accuracy": 0.2210903525352478,
      "num_tokens": 12090347.0,
      "step": 6880
    },
    {
      "entropy": 5.119500589370728,
      "epoch": 5.39576802507837,
      "grad_norm": 1.375,
      "learning_rate": 0.0002744585771077425,
      "loss": 4.5218,
      "mean_token_accuracy": 0.22310981452465056,
      "num_tokens": 12098340.0,
      "step": 6885
    },
    {
      "entropy": 5.144648838043213,
      "epoch": 5.399686520376176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027415778776600913,
      "loss": 4.5516,
      "mean_token_accuracy": 0.22398447543382644,
      "num_tokens": 12106771.0,
      "step": 6890
    },
    {
      "entropy": 5.170363426208496,
      "epoch": 5.403605015673981,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002738569999294456,
      "loss": 4.5597,
      "mean_token_accuracy": 0.22036366313695907,
      "num_tokens": 12115436.0,
      "step": 6895
    },
    {
      "entropy": 5.105395698547364,
      "epoch": 5.407523510971787,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002735562141356085,
      "loss": 4.4385,
      "mean_token_accuracy": 0.22388458847999573,
      "num_tokens": 12123939.0,
      "step": 6900
    },
    {
      "entropy": 5.232825183868409,
      "epoch": 5.411442006269592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002732554309220509,
      "loss": 4.6824,
      "mean_token_accuracy": 0.2135616272687912,
      "num_tokens": 12132803.0,
      "step": 6905
    },
    {
      "entropy": 5.127811336517334,
      "epoch": 5.415360501567398,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002729546508263211,
      "loss": 4.6092,
      "mean_token_accuracy": 0.21659033447504045,
      "num_tokens": 12141875.0,
      "step": 6910
    },
    {
      "entropy": 5.154586887359619,
      "epoch": 5.419278996865204,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002726538743859618,
      "loss": 4.5276,
      "mean_token_accuracy": 0.22410794347524643,
      "num_tokens": 12150685.0,
      "step": 6915
    },
    {
      "entropy": 5.1050256252288815,
      "epoch": 5.423197492163009,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002723531021385095,
      "loss": 4.5133,
      "mean_token_accuracy": 0.2231899991631508,
      "num_tokens": 12159070.0,
      "step": 6920
    },
    {
      "entropy": 5.117692232131958,
      "epoch": 5.427115987460815,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002720523346214928,
      "loss": 4.5628,
      "mean_token_accuracy": 0.22379320710897446,
      "num_tokens": 12167676.0,
      "step": 6925
    },
    {
      "entropy": 5.0910866260528564,
      "epoch": 5.431034482758621,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027175157237243204,
      "loss": 4.5381,
      "mean_token_accuracy": 0.21880834251642228,
      "num_tokens": 12176222.0,
      "step": 6930
    },
    {
      "entropy": 5.140155935287476,
      "epoch": 5.434952978056426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002714508159288382,
      "loss": 4.5344,
      "mean_token_accuracy": 0.22697316855192184,
      "num_tokens": 12185171.0,
      "step": 6935
    },
    {
      "entropy": 5.17184624671936,
      "epoch": 5.438871473354232,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002711500658282118,
      "loss": 4.6099,
      "mean_token_accuracy": 0.22324346303939818,
      "num_tokens": 12193875.0,
      "step": 6940
    },
    {
      "entropy": 5.08806586265564,
      "epoch": 5.442789968652038,
      "grad_norm": 1.25,
      "learning_rate": 0.00027084932260804193,
      "loss": 4.5282,
      "mean_token_accuracy": 0.225925575196743,
      "num_tokens": 12202681.0,
      "step": 6945
    },
    {
      "entropy": 5.122277021408081,
      "epoch": 5.446708463949843,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002705485868058056,
      "loss": 4.5097,
      "mean_token_accuracy": 0.22467216849327087,
      "num_tokens": 12210854.0,
      "step": 6950
    },
    {
      "entropy": 5.211093759536743,
      "epoch": 5.450626959247649,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027024785895896644,
      "loss": 4.5468,
      "mean_token_accuracy": 0.22023258805274964,
      "num_tokens": 12219643.0,
      "step": 6955
    },
    {
      "entropy": 5.157536554336548,
      "epoch": 5.454545454545454,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026994713960497383,
      "loss": 4.549,
      "mean_token_accuracy": 0.2246282696723938,
      "num_tokens": 12228767.0,
      "step": 6960
    },
    {
      "entropy": 5.141501092910767,
      "epoch": 5.45846394984326,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026964642928126197,
      "loss": 4.6378,
      "mean_token_accuracy": 0.21677130162715913,
      "num_tokens": 12237468.0,
      "step": 6965
    },
    {
      "entropy": 5.098905849456787,
      "epoch": 5.462382445141066,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00026934572852524907,
      "loss": 4.5918,
      "mean_token_accuracy": 0.21876722425222397,
      "num_tokens": 12246040.0,
      "step": 6970
    },
    {
      "entropy": 5.186588716506958,
      "epoch": 5.466300940438871,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026904503787433614,
      "loss": 4.6731,
      "mean_token_accuracy": 0.2099030300974846,
      "num_tokens": 12254646.0,
      "step": 6975
    },
    {
      "entropy": 5.147920846939087,
      "epoch": 5.470219435736677,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000268744357865906,
      "loss": 4.5577,
      "mean_token_accuracy": 0.2162935733795166,
      "num_tokens": 12263306.0,
      "step": 6980
    },
    {
      "entropy": 5.082468795776367,
      "epoch": 5.474137931034483,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026844368903732263,
      "loss": 4.5561,
      "mean_token_accuracy": 0.22090425342321396,
      "num_tokens": 12272498.0,
      "step": 6985
    },
    {
      "entropy": 5.029184722900391,
      "epoch": 5.478056426332288,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026814303192593015,
      "loss": 4.436,
      "mean_token_accuracy": 0.22902955263853073,
      "num_tokens": 12281592.0,
      "step": 6990
    },
    {
      "entropy": 5.161617136001587,
      "epoch": 5.481974921630094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026784238706905136,
      "loss": 4.5615,
      "mean_token_accuracy": 0.22017715126276016,
      "num_tokens": 12290112.0,
      "step": 6995
    },
    {
      "entropy": 5.197589254379272,
      "epoch": 5.485893416927899,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002675417550039874,
      "loss": 4.6001,
      "mean_token_accuracy": 0.21694946587085723,
      "num_tokens": 12299868.0,
      "step": 7000
    },
    {
      "epoch": 5.485893416927899,
      "eval_entropy": 5.060551832812701,
      "eval_loss": 5.655725002288818,
      "eval_mean_token_accuracy": 0.17130278738201127,
      "eval_num_tokens": 12299868.0,
      "eval_runtime": 2.8554,
      "eval_samples_per_second": 1443.555,
      "eval_steps_per_second": 180.707,
      "step": 7000
    },
    {
      "entropy": 5.139360666275024,
      "epoch": 5.489811912225705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026724113626801655,
      "loss": 4.5249,
      "mean_token_accuracy": 0.21910066455602645,
      "num_tokens": 12307903.0,
      "step": 7005
    },
    {
      "entropy": 5.1920534610748295,
      "epoch": 5.493730407523511,
      "grad_norm": 1.25,
      "learning_rate": 0.0002669405313983932,
      "loss": 4.6576,
      "mean_token_accuracy": 0.21478118002414703,
      "num_tokens": 12317287.0,
      "step": 7010
    },
    {
      "entropy": 5.222747278213501,
      "epoch": 5.497648902821316,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000266639940932347,
      "loss": 4.6384,
      "mean_token_accuracy": 0.21346164792776107,
      "num_tokens": 12325770.0,
      "step": 7015
    },
    {
      "entropy": 5.152126836776733,
      "epoch": 5.501567398119122,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002663393654070819,
      "loss": 4.574,
      "mean_token_accuracy": 0.21479832231998444,
      "num_tokens": 12334800.0,
      "step": 7020
    },
    {
      "entropy": 5.185343599319458,
      "epoch": 5.505485893416928,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002660388053597752,
      "loss": 4.5845,
      "mean_token_accuracy": 0.21623861342668532,
      "num_tokens": 12343362.0,
      "step": 7025
    },
    {
      "entropy": 5.16254153251648,
      "epoch": 5.509404388714733,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026573826132757623,
      "loss": 4.5966,
      "mean_token_accuracy": 0.21032341718673705,
      "num_tokens": 12352277.0,
      "step": 7030
    },
    {
      "entropy": 5.105849647521973,
      "epoch": 5.513322884012539,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026543773384760593,
      "loss": 4.4541,
      "mean_token_accuracy": 0.23194241225719453,
      "num_tokens": 12361244.0,
      "step": 7035
    },
    {
      "entropy": 5.136973857879639,
      "epoch": 5.517241379310345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002651372234569557,
      "loss": 4.5435,
      "mean_token_accuracy": 0.2155814364552498,
      "num_tokens": 12370228.0,
      "step": 7040
    },
    {
      "entropy": 5.086033725738526,
      "epoch": 5.52115987460815,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002648367306926864,
      "loss": 4.5109,
      "mean_token_accuracy": 0.22401620149612428,
      "num_tokens": 12379392.0,
      "step": 7045
    },
    {
      "entropy": 5.143168878555298,
      "epoch": 5.525078369905956,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002645362560918271,
      "loss": 4.5948,
      "mean_token_accuracy": 0.21438047736883165,
      "num_tokens": 12387334.0,
      "step": 7050
    },
    {
      "entropy": 5.135802936553955,
      "epoch": 5.528996865203762,
      "grad_norm": 1.21875,
      "learning_rate": 0.000264235800191375,
      "loss": 4.4948,
      "mean_token_accuracy": 0.22569056898355483,
      "num_tokens": 12396875.0,
      "step": 7055
    },
    {
      "entropy": 5.179565048217773,
      "epoch": 5.532915360501567,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002639353635282932,
      "loss": 4.509,
      "mean_token_accuracy": 0.23013535439968108,
      "num_tokens": 12405123.0,
      "step": 7060
    },
    {
      "entropy": 5.2452630519866945,
      "epoch": 5.536833855799373,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026363494663951076,
      "loss": 4.6446,
      "mean_token_accuracy": 0.20843614935874938,
      "num_tokens": 12413208.0,
      "step": 7065
    },
    {
      "entropy": 5.179215764999389,
      "epoch": 5.540752351097178,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026333455006192146,
      "loss": 4.6496,
      "mean_token_accuracy": 0.21053754538297653,
      "num_tokens": 12421704.0,
      "step": 7070
    },
    {
      "entropy": 5.162926721572876,
      "epoch": 5.544670846394984,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026303417433238267,
      "loss": 4.583,
      "mean_token_accuracy": 0.2139764368534088,
      "num_tokens": 12430610.0,
      "step": 7075
    },
    {
      "entropy": 5.1303974151611325,
      "epoch": 5.54858934169279,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002627338199877146,
      "loss": 4.4858,
      "mean_token_accuracy": 0.2228301152586937,
      "num_tokens": 12439867.0,
      "step": 7080
    },
    {
      "entropy": 5.1336242198944095,
      "epoch": 5.552507836990595,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002624334875646989,
      "loss": 4.5695,
      "mean_token_accuracy": 0.2183583900332451,
      "num_tokens": 12449540.0,
      "step": 7085
    },
    {
      "entropy": 5.141629981994629,
      "epoch": 5.556426332288401,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026213317760007864,
      "loss": 4.5903,
      "mean_token_accuracy": 0.21773380786180496,
      "num_tokens": 12458138.0,
      "step": 7090
    },
    {
      "entropy": 5.222535514831543,
      "epoch": 5.560344827586206,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002618328906305562,
      "loss": 4.6817,
      "mean_token_accuracy": 0.20805475413799285,
      "num_tokens": 12467921.0,
      "step": 7095
    },
    {
      "entropy": 5.228331518173218,
      "epoch": 5.564263322884012,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002615326271927931,
      "loss": 4.528,
      "mean_token_accuracy": 0.2176910936832428,
      "num_tokens": 12476873.0,
      "step": 7100
    },
    {
      "entropy": 5.186580991744995,
      "epoch": 5.568181818181818,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026123238782340893,
      "loss": 4.5698,
      "mean_token_accuracy": 0.2192312628030777,
      "num_tokens": 12485924.0,
      "step": 7105
    },
    {
      "entropy": 5.102415132522583,
      "epoch": 5.572100313479623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026093217305898003,
      "loss": 4.5085,
      "mean_token_accuracy": 0.22322235405445098,
      "num_tokens": 12494533.0,
      "step": 7110
    },
    {
      "entropy": 5.076966428756714,
      "epoch": 5.576018808777429,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026063198343603887,
      "loss": 4.4618,
      "mean_token_accuracy": 0.23291003704071045,
      "num_tokens": 12503137.0,
      "step": 7115
    },
    {
      "entropy": 5.11033296585083,
      "epoch": 5.579937304075235,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026033181949107286,
      "loss": 4.5797,
      "mean_token_accuracy": 0.22259340584278106,
      "num_tokens": 12511701.0,
      "step": 7120
    },
    {
      "entropy": 5.089531278610229,
      "epoch": 5.58385579937304,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002600316817605239,
      "loss": 4.5644,
      "mean_token_accuracy": 0.22047974169254303,
      "num_tokens": 12520097.0,
      "step": 7125
    },
    {
      "entropy": 5.148102951049805,
      "epoch": 5.587774294670846,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002597315707807866,
      "loss": 4.5781,
      "mean_token_accuracy": 0.22125348895788194,
      "num_tokens": 12528526.0,
      "step": 7130
    },
    {
      "entropy": 5.139005470275879,
      "epoch": 5.591692789968652,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025943148708820784,
      "loss": 4.5831,
      "mean_token_accuracy": 0.22155668437480927,
      "num_tokens": 12537888.0,
      "step": 7135
    },
    {
      "entropy": 5.117470073699951,
      "epoch": 5.5956112852664575,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002591314312190861,
      "loss": 4.5661,
      "mean_token_accuracy": 0.21918662190437316,
      "num_tokens": 12546396.0,
      "step": 7140
    },
    {
      "entropy": 5.1071539402008055,
      "epoch": 5.599529780564263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025883140370966975,
      "loss": 4.4997,
      "mean_token_accuracy": 0.22462174892425538,
      "num_tokens": 12554985.0,
      "step": 7145
    },
    {
      "entropy": 5.140321397781372,
      "epoch": 5.603448275862069,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002585314050961565,
      "loss": 4.5098,
      "mean_token_accuracy": 0.23839443325996398,
      "num_tokens": 12564379.0,
      "step": 7150
    },
    {
      "entropy": 5.157658815383911,
      "epoch": 5.6073667711598745,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002582314359146927,
      "loss": 4.5646,
      "mean_token_accuracy": 0.22123173624277115,
      "num_tokens": 12572655.0,
      "step": 7155
    },
    {
      "entropy": 5.140807485580444,
      "epoch": 5.61128526645768,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025793149670137175,
      "loss": 4.5976,
      "mean_token_accuracy": 0.21654587984085083,
      "num_tokens": 12581778.0,
      "step": 7160
    },
    {
      "entropy": 5.078033208847046,
      "epoch": 5.6152037617554855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002576315879922337,
      "loss": 4.472,
      "mean_token_accuracy": 0.22873728573322297,
      "num_tokens": 12590324.0,
      "step": 7165
    },
    {
      "entropy": 5.13161678314209,
      "epoch": 5.6191222570532915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025733171032326404,
      "loss": 4.6067,
      "mean_token_accuracy": 0.21395044326782225,
      "num_tokens": 12599761.0,
      "step": 7170
    },
    {
      "entropy": 5.1698822498321535,
      "epoch": 5.6230407523510975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002570318642303929,
      "loss": 4.6133,
      "mean_token_accuracy": 0.21642404645681382,
      "num_tokens": 12608853.0,
      "step": 7175
    },
    {
      "entropy": 5.116098642349243,
      "epoch": 5.6269592476489025,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025673205024949377,
      "loss": 4.5115,
      "mean_token_accuracy": 0.22463483065366746,
      "num_tokens": 12617434.0,
      "step": 7180
    },
    {
      "entropy": 5.132462120056152,
      "epoch": 5.6308777429467085,
      "grad_norm": 1.375,
      "learning_rate": 0.0002564322689163828,
      "loss": 4.5628,
      "mean_token_accuracy": 0.22396426051855087,
      "num_tokens": 12626021.0,
      "step": 7185
    },
    {
      "entropy": 5.174086809158325,
      "epoch": 5.6347962382445145,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025613252076681786,
      "loss": 4.544,
      "mean_token_accuracy": 0.22122825384140016,
      "num_tokens": 12634952.0,
      "step": 7190
    },
    {
      "entropy": 5.12405915260315,
      "epoch": 5.63871473354232,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002558328063364976,
      "loss": 4.5568,
      "mean_token_accuracy": 0.22180014997720718,
      "num_tokens": 12644094.0,
      "step": 7195
    },
    {
      "entropy": 5.149139356613159,
      "epoch": 5.6426332288401255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002555331261610603,
      "loss": 4.5827,
      "mean_token_accuracy": 0.2180441901087761,
      "num_tokens": 12652687.0,
      "step": 7200
    },
    {
      "entropy": 5.123326921463013,
      "epoch": 5.646551724137931,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025523348077608285,
      "loss": 4.5379,
      "mean_token_accuracy": 0.22558983713388442,
      "num_tokens": 12661437.0,
      "step": 7205
    },
    {
      "entropy": 5.186764240264893,
      "epoch": 5.650470219435737,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025493387071708024,
      "loss": 4.5943,
      "mean_token_accuracy": 0.22045840919017792,
      "num_tokens": 12669893.0,
      "step": 7210
    },
    {
      "entropy": 5.1424720287323,
      "epoch": 5.6543887147335425,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025463429651950426,
      "loss": 4.6101,
      "mean_token_accuracy": 0.21910127103328705,
      "num_tokens": 12679349.0,
      "step": 7215
    },
    {
      "entropy": 5.153482103347779,
      "epoch": 5.658307210031348,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002543347587187425,
      "loss": 4.5427,
      "mean_token_accuracy": 0.21732700765132903,
      "num_tokens": 12687163.0,
      "step": 7220
    },
    {
      "entropy": 5.134524011611939,
      "epoch": 5.662225705329154,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025403525785011767,
      "loss": 4.5545,
      "mean_token_accuracy": 0.2160551980137825,
      "num_tokens": 12695646.0,
      "step": 7225
    },
    {
      "entropy": 5.148728370666504,
      "epoch": 5.66614420062696,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025373579444888634,
      "loss": 4.5107,
      "mean_token_accuracy": 0.23015272319316865,
      "num_tokens": 12705144.0,
      "step": 7230
    },
    {
      "entropy": 5.134817457199096,
      "epoch": 5.670062695924765,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000253436369050238,
      "loss": 4.5276,
      "mean_token_accuracy": 0.21862490475177765,
      "num_tokens": 12713881.0,
      "step": 7235
    },
    {
      "entropy": 5.129478979110718,
      "epoch": 5.673981191222571,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025313698218929474,
      "loss": 4.492,
      "mean_token_accuracy": 0.23121701329946517,
      "num_tokens": 12722452.0,
      "step": 7240
    },
    {
      "entropy": 5.107900857925415,
      "epoch": 5.677899686520377,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025283763440110923,
      "loss": 4.5269,
      "mean_token_accuracy": 0.22020430266857147,
      "num_tokens": 12731105.0,
      "step": 7245
    },
    {
      "entropy": 5.081765365600586,
      "epoch": 5.681818181818182,
      "grad_norm": 1.375,
      "learning_rate": 0.0002525383262206645,
      "loss": 4.5396,
      "mean_token_accuracy": 0.23038032352924348,
      "num_tokens": 12739096.0,
      "step": 7250
    },
    {
      "entropy": 5.123942279815674,
      "epoch": 5.685736677115988,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002522390581828726,
      "loss": 4.5355,
      "mean_token_accuracy": 0.22441412955522538,
      "num_tokens": 12747910.0,
      "step": 7255
    },
    {
      "entropy": 5.088297986984253,
      "epoch": 5.689655172413794,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025193983082257444,
      "loss": 4.4316,
      "mean_token_accuracy": 0.23370273858308793,
      "num_tokens": 12757387.0,
      "step": 7260
    },
    {
      "entropy": 5.101227807998657,
      "epoch": 5.693573667711599,
      "grad_norm": 1.25,
      "learning_rate": 0.0002516406446745376,
      "loss": 4.5572,
      "mean_token_accuracy": 0.22102911472320558,
      "num_tokens": 12766678.0,
      "step": 7265
    },
    {
      "entropy": 5.086721181869507,
      "epoch": 5.697492163009405,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002513415002734562,
      "loss": 4.5241,
      "mean_token_accuracy": 0.22164885699748993,
      "num_tokens": 12775499.0,
      "step": 7270
    },
    {
      "entropy": 5.174330377578736,
      "epoch": 5.70141065830721,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025104239815394995,
      "loss": 4.6078,
      "mean_token_accuracy": 0.21608633399009705,
      "num_tokens": 12784173.0,
      "step": 7275
    },
    {
      "entropy": 5.170020675659179,
      "epoch": 5.705329153605016,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002507433388505627,
      "loss": 4.6345,
      "mean_token_accuracy": 0.22143435180187226,
      "num_tokens": 12792991.0,
      "step": 7280
    },
    {
      "entropy": 5.095520496368408,
      "epoch": 5.709247648902822,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025044432289776197,
      "loss": 4.5366,
      "mean_token_accuracy": 0.22131821811199187,
      "num_tokens": 12801868.0,
      "step": 7285
    },
    {
      "entropy": 5.1434320449829105,
      "epoch": 5.713166144200627,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025014535082993765,
      "loss": 4.5728,
      "mean_token_accuracy": 0.22643867880105972,
      "num_tokens": 12810467.0,
      "step": 7290
    },
    {
      "entropy": 5.112856006622314,
      "epoch": 5.717084639498433,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002498464231814015,
      "loss": 4.5782,
      "mean_token_accuracy": 0.22142210155725478,
      "num_tokens": 12818874.0,
      "step": 7295
    },
    {
      "entropy": 5.121105337142945,
      "epoch": 5.721003134796238,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024954754048638555,
      "loss": 4.4696,
      "mean_token_accuracy": 0.2250881388783455,
      "num_tokens": 12828029.0,
      "step": 7300
    },
    {
      "entropy": 5.118112659454345,
      "epoch": 5.724921630094044,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024924870327904165,
      "loss": 4.5339,
      "mean_token_accuracy": 0.2231732726097107,
      "num_tokens": 12836185.0,
      "step": 7305
    },
    {
      "entropy": 5.218063926696777,
      "epoch": 5.72884012539185,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002489499120934403,
      "loss": 4.6702,
      "mean_token_accuracy": 0.20779013633728027,
      "num_tokens": 12845810.0,
      "step": 7310
    },
    {
      "entropy": 5.09456582069397,
      "epoch": 5.732758620689655,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024865116746357004,
      "loss": 4.4358,
      "mean_token_accuracy": 0.23992856591939926,
      "num_tokens": 12854567.0,
      "step": 7315
    },
    {
      "entropy": 5.058548498153686,
      "epoch": 5.736677115987461,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002483524699233356,
      "loss": 4.5242,
      "mean_token_accuracy": 0.2201599046587944,
      "num_tokens": 12863172.0,
      "step": 7320
    },
    {
      "entropy": 5.128382682800293,
      "epoch": 5.740595611285267,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024805382000655824,
      "loss": 4.6392,
      "mean_token_accuracy": 0.20978759229183197,
      "num_tokens": 12873335.0,
      "step": 7325
    },
    {
      "entropy": 5.177002763748169,
      "epoch": 5.744514106583072,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002477552182469736,
      "loss": 4.5887,
      "mean_token_accuracy": 0.22355477511882782,
      "num_tokens": 12882695.0,
      "step": 7330
    },
    {
      "entropy": 5.093094539642334,
      "epoch": 5.748432601880878,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002474566651782315,
      "loss": 4.5908,
      "mean_token_accuracy": 0.21473755240440368,
      "num_tokens": 12890739.0,
      "step": 7335
    },
    {
      "entropy": 5.043901205062866,
      "epoch": 5.752351097178684,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002471581613338946,
      "loss": 4.552,
      "mean_token_accuracy": 0.2229151114821434,
      "num_tokens": 12899503.0,
      "step": 7340
    },
    {
      "entropy": 5.152907514572144,
      "epoch": 5.756269592476489,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002468597072474378,
      "loss": 4.5932,
      "mean_token_accuracy": 0.2191447764635086,
      "num_tokens": 12908405.0,
      "step": 7345
    },
    {
      "entropy": 5.183080863952637,
      "epoch": 5.760188087774295,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024656130345224685,
      "loss": 4.6379,
      "mean_token_accuracy": 0.20947250127792358,
      "num_tokens": 12917529.0,
      "step": 7350
    },
    {
      "entropy": 5.204360389709473,
      "epoch": 5.764106583072101,
      "grad_norm": 1.375,
      "learning_rate": 0.00024626295048161775,
      "loss": 4.5966,
      "mean_token_accuracy": 0.21756329089403154,
      "num_tokens": 12926183.0,
      "step": 7355
    },
    {
      "entropy": 5.23629093170166,
      "epoch": 5.768025078369906,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002459646488687557,
      "loss": 4.6495,
      "mean_token_accuracy": 0.2134860038757324,
      "num_tokens": 12934598.0,
      "step": 7360
    },
    {
      "entropy": 5.118347215652466,
      "epoch": 5.771943573667712,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002456663991467739,
      "loss": 4.5609,
      "mean_token_accuracy": 0.21560123413801194,
      "num_tokens": 12943101.0,
      "step": 7365
    },
    {
      "entropy": 5.15675630569458,
      "epoch": 5.775862068965517,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024536820184869313,
      "loss": 4.6099,
      "mean_token_accuracy": 0.21783594638109208,
      "num_tokens": 12951717.0,
      "step": 7370
    },
    {
      "entropy": 5.181252670288086,
      "epoch": 5.779780564263323,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002450700575074401,
      "loss": 4.6688,
      "mean_token_accuracy": 0.21597288101911544,
      "num_tokens": 12960761.0,
      "step": 7375
    },
    {
      "entropy": 5.128368282318116,
      "epoch": 5.783699059561129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024477196665584735,
      "loss": 4.4612,
      "mean_token_accuracy": 0.23750704377889634,
      "num_tokens": 12969216.0,
      "step": 7380
    },
    {
      "entropy": 5.158656263351441,
      "epoch": 5.787617554858934,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002444739298266513,
      "loss": 4.5541,
      "mean_token_accuracy": 0.2163679778575897,
      "num_tokens": 12977260.0,
      "step": 7385
    },
    {
      "entropy": 5.099487400054931,
      "epoch": 5.79153605015674,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024417594755249227,
      "loss": 4.5498,
      "mean_token_accuracy": 0.22029636353254317,
      "num_tokens": 12985383.0,
      "step": 7390
    },
    {
      "entropy": 5.128294324874878,
      "epoch": 5.795454545454545,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024387802036591295,
      "loss": 4.5425,
      "mean_token_accuracy": 0.22547662258148193,
      "num_tokens": 12994675.0,
      "step": 7395
    },
    {
      "entropy": 5.1568550109863285,
      "epoch": 5.799373040752351,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002435801487993573,
      "loss": 4.5528,
      "mean_token_accuracy": 0.22022849172353745,
      "num_tokens": 13002979.0,
      "step": 7400
    },
    {
      "entropy": 5.108413362503052,
      "epoch": 5.803291536050157,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024328233338517024,
      "loss": 4.5462,
      "mean_token_accuracy": 0.21552518010139465,
      "num_tokens": 13011343.0,
      "step": 7405
    },
    {
      "entropy": 5.168239545822144,
      "epoch": 5.807210031347962,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024298457465559626,
      "loss": 4.6714,
      "mean_token_accuracy": 0.207929527759552,
      "num_tokens": 13019684.0,
      "step": 7410
    },
    {
      "entropy": 5.155858039855957,
      "epoch": 5.811128526645768,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024268687314277844,
      "loss": 4.5428,
      "mean_token_accuracy": 0.2246020630002022,
      "num_tokens": 13028358.0,
      "step": 7415
    },
    {
      "entropy": 5.068447589874268,
      "epoch": 5.815047021943574,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024238922937875758,
      "loss": 4.5057,
      "mean_token_accuracy": 0.2213594689965248,
      "num_tokens": 13037092.0,
      "step": 7420
    },
    {
      "entropy": 5.148108243942261,
      "epoch": 5.818965517241379,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002420916438954714,
      "loss": 4.6281,
      "mean_token_accuracy": 0.21565960943698884,
      "num_tokens": 13045624.0,
      "step": 7425
    },
    {
      "entropy": 5.169312524795532,
      "epoch": 5.822884012539185,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002417941172247535,
      "loss": 4.5866,
      "mean_token_accuracy": 0.2189791679382324,
      "num_tokens": 13054384.0,
      "step": 7430
    },
    {
      "entropy": 5.135847854614258,
      "epoch": 5.826802507836991,
      "grad_norm": 1.21875,
      "learning_rate": 0.00024149664989833215,
      "loss": 4.6145,
      "mean_token_accuracy": 0.22059242725372313,
      "num_tokens": 13063786.0,
      "step": 7435
    },
    {
      "entropy": 5.198866128921509,
      "epoch": 5.830721003134796,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024119924244782965,
      "loss": 4.6369,
      "mean_token_accuracy": 0.21038111448287963,
      "num_tokens": 13073446.0,
      "step": 7440
    },
    {
      "entropy": 5.094502067565918,
      "epoch": 5.834639498432602,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024090189540476153,
      "loss": 4.5644,
      "mean_token_accuracy": 0.2195345178246498,
      "num_tokens": 13081780.0,
      "step": 7445
    },
    {
      "entropy": 5.177765655517578,
      "epoch": 5.838557993730408,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024060460930053513,
      "loss": 4.5754,
      "mean_token_accuracy": 0.2240541696548462,
      "num_tokens": 13089995.0,
      "step": 7450
    },
    {
      "entropy": 5.129477548599243,
      "epoch": 5.842476489028213,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024030738466644875,
      "loss": 4.5456,
      "mean_token_accuracy": 0.2201928973197937,
      "num_tokens": 13098767.0,
      "step": 7455
    },
    {
      "entropy": 5.131991720199585,
      "epoch": 5.846394984326019,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002400102220336911,
      "loss": 4.6196,
      "mean_token_accuracy": 0.21950864940881729,
      "num_tokens": 13107744.0,
      "step": 7460
    },
    {
      "entropy": 5.110366773605347,
      "epoch": 5.850313479623824,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002397131219333401,
      "loss": 4.5128,
      "mean_token_accuracy": 0.22484900802373886,
      "num_tokens": 13116400.0,
      "step": 7465
    },
    {
      "entropy": 5.087181186676025,
      "epoch": 5.85423197492163,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002394160848963617,
      "loss": 4.5321,
      "mean_token_accuracy": 0.22314715385437012,
      "num_tokens": 13124973.0,
      "step": 7470
    },
    {
      "entropy": 5.1826849460601805,
      "epoch": 5.858150470219436,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023911911145360916,
      "loss": 4.573,
      "mean_token_accuracy": 0.22096654921770095,
      "num_tokens": 13133061.0,
      "step": 7475
    },
    {
      "entropy": 5.2082743644714355,
      "epoch": 5.862068965517241,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023882220213582235,
      "loss": 4.61,
      "mean_token_accuracy": 0.22066780179738998,
      "num_tokens": 13141322.0,
      "step": 7480
    },
    {
      "entropy": 5.137362384796143,
      "epoch": 5.865987460815047,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002385253574736262,
      "loss": 4.5714,
      "mean_token_accuracy": 0.22064700871706008,
      "num_tokens": 13150553.0,
      "step": 7485
    },
    {
      "entropy": 5.240194988250733,
      "epoch": 5.869905956112853,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023822857799753034,
      "loss": 4.665,
      "mean_token_accuracy": 0.21182449162006378,
      "num_tokens": 13160308.0,
      "step": 7490
    },
    {
      "entropy": 5.195801687240601,
      "epoch": 5.873824451410658,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023793186423792784,
      "loss": 4.6383,
      "mean_token_accuracy": 0.22271146029233932,
      "num_tokens": 13168219.0,
      "step": 7495
    },
    {
      "entropy": 5.19296236038208,
      "epoch": 5.877742946708464,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023763521672509432,
      "loss": 4.6112,
      "mean_token_accuracy": 0.21305580884218217,
      "num_tokens": 13176980.0,
      "step": 7500
    },
    {
      "epoch": 5.877742946708464,
      "eval_entropy": 5.061081781867863,
      "eval_loss": 5.636685371398926,
      "eval_mean_token_accuracy": 0.17258776136666767,
      "eval_num_tokens": 13176980.0,
      "eval_runtime": 2.8415,
      "eval_samples_per_second": 1450.651,
      "eval_steps_per_second": 181.595,
      "step": 7500
    },
    {
      "entropy": 5.2002757549285885,
      "epoch": 5.881661442006269,
      "grad_norm": 1.25,
      "learning_rate": 0.00023733863598918688,
      "loss": 4.6396,
      "mean_token_accuracy": 0.20794833302497864,
      "num_tokens": 13186267.0,
      "step": 7505
    },
    {
      "entropy": 5.156612968444824,
      "epoch": 5.885579937304075,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002370421225602433,
      "loss": 4.654,
      "mean_token_accuracy": 0.2179220974445343,
      "num_tokens": 13195457.0,
      "step": 7510
    },
    {
      "entropy": 5.192200231552124,
      "epoch": 5.889498432601881,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023674567696818145,
      "loss": 4.628,
      "mean_token_accuracy": 0.21614690124988556,
      "num_tokens": 13203707.0,
      "step": 7515
    },
    {
      "entropy": 5.169398355484009,
      "epoch": 5.893416927899686,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002364492997427975,
      "loss": 4.5346,
      "mean_token_accuracy": 0.21971122771501542,
      "num_tokens": 13213013.0,
      "step": 7520
    },
    {
      "entropy": 5.117684984207154,
      "epoch": 5.897335423197492,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023615299141376545,
      "loss": 4.576,
      "mean_token_accuracy": 0.22478134632110597,
      "num_tokens": 13222356.0,
      "step": 7525
    },
    {
      "entropy": 5.131736755371094,
      "epoch": 5.901253918495298,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023585675251063665,
      "loss": 4.5885,
      "mean_token_accuracy": 0.212899349629879,
      "num_tokens": 13231791.0,
      "step": 7530
    },
    {
      "entropy": 5.132737445831299,
      "epoch": 5.905172413793103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002355605835628379,
      "loss": 4.5147,
      "mean_token_accuracy": 0.2259177803993225,
      "num_tokens": 13240445.0,
      "step": 7535
    },
    {
      "entropy": 5.138512325286865,
      "epoch": 5.909090909090909,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023526448509967097,
      "loss": 4.5994,
      "mean_token_accuracy": 0.21186794191598893,
      "num_tokens": 13249048.0,
      "step": 7540
    },
    {
      "entropy": 5.096324396133423,
      "epoch": 5.913009404388715,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023496845765031215,
      "loss": 4.4751,
      "mean_token_accuracy": 0.23003213554620744,
      "num_tokens": 13258206.0,
      "step": 7545
    },
    {
      "entropy": 5.184954357147217,
      "epoch": 5.91692789968652,
      "grad_norm": 1.375,
      "learning_rate": 0.00023467250174381021,
      "loss": 4.5823,
      "mean_token_accuracy": 0.21888032853603362,
      "num_tokens": 13266756.0,
      "step": 7550
    },
    {
      "entropy": 5.13728609085083,
      "epoch": 5.920846394984326,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023437661790908648,
      "loss": 4.6337,
      "mean_token_accuracy": 0.21181762218475342,
      "num_tokens": 13276033.0,
      "step": 7555
    },
    {
      "entropy": 5.196960973739624,
      "epoch": 5.924764890282132,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023408080667493332,
      "loss": 4.5817,
      "mean_token_accuracy": 0.21433744430541993,
      "num_tokens": 13285151.0,
      "step": 7560
    },
    {
      "entropy": 5.090860652923584,
      "epoch": 5.928683385579937,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023378506857001335,
      "loss": 4.5086,
      "mean_token_accuracy": 0.22173936069011688,
      "num_tokens": 13293412.0,
      "step": 7565
    },
    {
      "entropy": 5.156076669692993,
      "epoch": 5.932601880877743,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023348940412285858,
      "loss": 4.597,
      "mean_token_accuracy": 0.2162150412797928,
      "num_tokens": 13301982.0,
      "step": 7570
    },
    {
      "entropy": 5.160928201675415,
      "epoch": 5.936520376175548,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023319381386186923,
      "loss": 4.6061,
      "mean_token_accuracy": 0.2180646315217018,
      "num_tokens": 13309766.0,
      "step": 7575
    },
    {
      "entropy": 5.140422725677491,
      "epoch": 5.940438871473354,
      "grad_norm": 1.34375,
      "learning_rate": 0.000232898298315313,
      "loss": 4.5453,
      "mean_token_accuracy": 0.22937876284122466,
      "num_tokens": 13317960.0,
      "step": 7580
    },
    {
      "entropy": 5.1957480907440186,
      "epoch": 5.94435736677116,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002326028580113243,
      "loss": 4.6449,
      "mean_token_accuracy": 0.2120690107345581,
      "num_tokens": 13326909.0,
      "step": 7585
    },
    {
      "entropy": 5.218386220932007,
      "epoch": 5.948275862068965,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023230749347790265,
      "loss": 4.6578,
      "mean_token_accuracy": 0.2112424224615097,
      "num_tokens": 13335861.0,
      "step": 7590
    },
    {
      "entropy": 5.218826723098755,
      "epoch": 5.952194357366771,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023201220524291232,
      "loss": 4.6096,
      "mean_token_accuracy": 0.2163745328783989,
      "num_tokens": 13345253.0,
      "step": 7595
    },
    {
      "entropy": 5.157000923156739,
      "epoch": 5.956112852664576,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023171699383408152,
      "loss": 4.5736,
      "mean_token_accuracy": 0.22393871396780013,
      "num_tokens": 13354353.0,
      "step": 7600
    },
    {
      "entropy": 5.141835355758667,
      "epoch": 5.960031347962382,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023142185977900065,
      "loss": 4.6172,
      "mean_token_accuracy": 0.22018101811408997,
      "num_tokens": 13363030.0,
      "step": 7605
    },
    {
      "entropy": 5.141004276275635,
      "epoch": 5.963949843260188,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023112680360512218,
      "loss": 4.5783,
      "mean_token_accuracy": 0.22098631262779236,
      "num_tokens": 13372797.0,
      "step": 7610
    },
    {
      "entropy": 5.172771644592285,
      "epoch": 5.967868338557993,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023083182583975942,
      "loss": 4.5132,
      "mean_token_accuracy": 0.22066895216703414,
      "num_tokens": 13381425.0,
      "step": 7615
    },
    {
      "entropy": 5.1031066417694095,
      "epoch": 5.971786833855799,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002305369270100855,
      "loss": 4.4952,
      "mean_token_accuracy": 0.22404034286737443,
      "num_tokens": 13390246.0,
      "step": 7620
    },
    {
      "entropy": 5.141392183303833,
      "epoch": 5.975705329153605,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002302421076431322,
      "loss": 4.5376,
      "mean_token_accuracy": 0.22173997908830642,
      "num_tokens": 13398851.0,
      "step": 7625
    },
    {
      "entropy": 5.161487483978272,
      "epoch": 5.97962382445141,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022994736826578976,
      "loss": 4.6214,
      "mean_token_accuracy": 0.21318777352571489,
      "num_tokens": 13407870.0,
      "step": 7630
    },
    {
      "entropy": 5.14744815826416,
      "epoch": 5.983542319749216,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002296527094048052,
      "loss": 4.6013,
      "mean_token_accuracy": 0.21870656162500382,
      "num_tokens": 13416311.0,
      "step": 7635
    },
    {
      "entropy": 5.15994930267334,
      "epoch": 5.987460815047022,
      "grad_norm": 1.25,
      "learning_rate": 0.0002293581315867817,
      "loss": 4.6578,
      "mean_token_accuracy": 0.20971277356147766,
      "num_tokens": 13425669.0,
      "step": 7640
    },
    {
      "entropy": 5.074539852142334,
      "epoch": 5.991379310344827,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002290636353381774,
      "loss": 4.5385,
      "mean_token_accuracy": 0.22208404988050462,
      "num_tokens": 13434448.0,
      "step": 7645
    },
    {
      "entropy": 5.173510360717773,
      "epoch": 5.995297805642633,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022876922118530508,
      "loss": 4.6076,
      "mean_token_accuracy": 0.21574989706277847,
      "num_tokens": 13443712.0,
      "step": 7650
    },
    {
      "entropy": 5.130525827407837,
      "epoch": 5.999216300940439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002284748896543305,
      "loss": 4.3967,
      "mean_token_accuracy": 0.2324090838432312,
      "num_tokens": 13452511.0,
      "step": 7655
    },
    {
      "entropy": 5.095558500289917,
      "epoch": 6.003134796238244,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022818064127127163,
      "loss": 4.3525,
      "mean_token_accuracy": 0.2394998326897621,
      "num_tokens": 13460990.0,
      "step": 7660
    },
    {
      "entropy": 5.108092355728149,
      "epoch": 6.00705329153605,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022788647656199834,
      "loss": 4.4741,
      "mean_token_accuracy": 0.23613135069608687,
      "num_tokens": 13470293.0,
      "step": 7665
    },
    {
      "entropy": 5.151597261428833,
      "epoch": 6.0109717868338555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022759239605223052,
      "loss": 4.3946,
      "mean_token_accuracy": 0.23197797536849976,
      "num_tokens": 13479508.0,
      "step": 7670
    },
    {
      "entropy": 5.098331165313721,
      "epoch": 6.014890282131661,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022729840026753777,
      "loss": 4.3156,
      "mean_token_accuracy": 0.24126329272985458,
      "num_tokens": 13488163.0,
      "step": 7675
    },
    {
      "entropy": 5.19809627532959,
      "epoch": 6.018808777429467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022700448973333808,
      "loss": 4.4622,
      "mean_token_accuracy": 0.2316286087036133,
      "num_tokens": 13497404.0,
      "step": 7680
    },
    {
      "entropy": 5.133011102676392,
      "epoch": 6.0227272727272725,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022671066497489744,
      "loss": 4.3927,
      "mean_token_accuracy": 0.2370530515909195,
      "num_tokens": 13506142.0,
      "step": 7685
    },
    {
      "entropy": 5.113889169692993,
      "epoch": 6.0266457680250785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002264169265173282,
      "loss": 4.3028,
      "mean_token_accuracy": 0.24973479956388472,
      "num_tokens": 13514525.0,
      "step": 7690
    },
    {
      "entropy": 5.097209548950195,
      "epoch": 6.030564263322884,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022612327488558866,
      "loss": 4.3063,
      "mean_token_accuracy": 0.24466561377048493,
      "num_tokens": 13523732.0,
      "step": 7695
    },
    {
      "entropy": 5.063730669021607,
      "epoch": 6.0344827586206895,
      "grad_norm": 1.25,
      "learning_rate": 0.00022582971060448187,
      "loss": 4.3318,
      "mean_token_accuracy": 0.24371203631162644,
      "num_tokens": 13532904.0,
      "step": 7700
    },
    {
      "entropy": 5.078183603286743,
      "epoch": 6.0384012539184955,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022553623419865488,
      "loss": 4.3602,
      "mean_token_accuracy": 0.23258696794509887,
      "num_tokens": 13542174.0,
      "step": 7705
    },
    {
      "entropy": 5.109649896621704,
      "epoch": 6.0423197492163006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022524284619259745,
      "loss": 4.3912,
      "mean_token_accuracy": 0.2351180925965309,
      "num_tokens": 13550591.0,
      "step": 7710
    },
    {
      "entropy": 5.090293502807617,
      "epoch": 6.0462382445141065,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002249495471106417,
      "loss": 4.3544,
      "mean_token_accuracy": 0.24451110363006592,
      "num_tokens": 13559088.0,
      "step": 7715
    },
    {
      "entropy": 5.054904651641846,
      "epoch": 6.0501567398119125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022465633747696052,
      "loss": 4.2735,
      "mean_token_accuracy": 0.24230126589536666,
      "num_tokens": 13568511.0,
      "step": 7720
    },
    {
      "entropy": 5.135289144515991,
      "epoch": 6.054075235109718,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022436321781556712,
      "loss": 4.4459,
      "mean_token_accuracy": 0.22325546741485597,
      "num_tokens": 13577750.0,
      "step": 7725
    },
    {
      "entropy": 5.138867044448853,
      "epoch": 6.0579937304075235,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002240701886503138,
      "loss": 4.3742,
      "mean_token_accuracy": 0.23529665321111679,
      "num_tokens": 13587138.0,
      "step": 7730
    },
    {
      "entropy": 5.168648099899292,
      "epoch": 6.0619122257053295,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022377725050489128,
      "loss": 4.3839,
      "mean_token_accuracy": 0.2386893227696419,
      "num_tokens": 13595459.0,
      "step": 7735
    },
    {
      "entropy": 5.168027019500732,
      "epoch": 6.065830721003135,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022348440390282753,
      "loss": 4.4516,
      "mean_token_accuracy": 0.23202283233404158,
      "num_tokens": 13605227.0,
      "step": 7740
    },
    {
      "entropy": 5.1937644481658936,
      "epoch": 6.0697492163009406,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022319164936748676,
      "loss": 4.4571,
      "mean_token_accuracy": 0.22701892703771592,
      "num_tokens": 13614139.0,
      "step": 7745
    },
    {
      "entropy": 5.168273305892944,
      "epoch": 6.0736677115987465,
      "grad_norm": 1.375,
      "learning_rate": 0.00022289898742206904,
      "loss": 4.3643,
      "mean_token_accuracy": 0.23764010071754454,
      "num_tokens": 13623420.0,
      "step": 7750
    },
    {
      "entropy": 5.131233882904053,
      "epoch": 6.077586206896552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002226064185896085,
      "loss": 4.4027,
      "mean_token_accuracy": 0.23727071583271026,
      "num_tokens": 13632655.0,
      "step": 7755
    },
    {
      "entropy": 5.127702426910401,
      "epoch": 6.081504702194358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022231394339297322,
      "loss": 4.3244,
      "mean_token_accuracy": 0.23854795098304749,
      "num_tokens": 13641782.0,
      "step": 7760
    },
    {
      "entropy": 5.115083312988281,
      "epoch": 6.085423197492163,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022202156235486358,
      "loss": 4.3384,
      "mean_token_accuracy": 0.2438800886273384,
      "num_tokens": 13650023.0,
      "step": 7765
    },
    {
      "entropy": 5.06524691581726,
      "epoch": 6.089341692789969,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022172927599781224,
      "loss": 4.3442,
      "mean_token_accuracy": 0.23896822929382325,
      "num_tokens": 13659025.0,
      "step": 7770
    },
    {
      "entropy": 5.184850692749023,
      "epoch": 6.093260188087775,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022143708484418206,
      "loss": 4.3991,
      "mean_token_accuracy": 0.23489006757736205,
      "num_tokens": 13668133.0,
      "step": 7775
    },
    {
      "entropy": 5.120969676971436,
      "epoch": 6.09717868338558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022114498941616607,
      "loss": 4.4092,
      "mean_token_accuracy": 0.2260152667760849,
      "num_tokens": 13677661.0,
      "step": 7780
    },
    {
      "entropy": 5.069052362442017,
      "epoch": 6.101097178683386,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022085299023578614,
      "loss": 4.383,
      "mean_token_accuracy": 0.23395193219184876,
      "num_tokens": 13686117.0,
      "step": 7785
    },
    {
      "entropy": 5.077052354812622,
      "epoch": 6.105015673981192,
      "grad_norm": 1.375,
      "learning_rate": 0.00022056108782489225,
      "loss": 4.2956,
      "mean_token_accuracy": 0.23888327330350875,
      "num_tokens": 13694792.0,
      "step": 7790
    },
    {
      "entropy": 5.129629468917846,
      "epoch": 6.108934169278997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022026928270516122,
      "loss": 4.4148,
      "mean_token_accuracy": 0.23177953362464904,
      "num_tokens": 13704038.0,
      "step": 7795
    },
    {
      "entropy": 5.148499250411987,
      "epoch": 6.112852664576803,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021997757539809622,
      "loss": 4.4002,
      "mean_token_accuracy": 0.23162806779146194,
      "num_tokens": 13713266.0,
      "step": 7800
    },
    {
      "entropy": 5.082339715957642,
      "epoch": 6.116771159874608,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021968596642502546,
      "loss": 4.3369,
      "mean_token_accuracy": 0.24328903406858443,
      "num_tokens": 13721847.0,
      "step": 7805
    },
    {
      "entropy": 5.069874620437622,
      "epoch": 6.120689655172414,
      "grad_norm": 1.203125,
      "learning_rate": 0.00021939445630710138,
      "loss": 4.3526,
      "mean_token_accuracy": 0.2382796362042427,
      "num_tokens": 13730925.0,
      "step": 7810
    },
    {
      "entropy": 5.104957628250122,
      "epoch": 6.12460815047022,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021910304556529986,
      "loss": 4.4562,
      "mean_token_accuracy": 0.23077407628297805,
      "num_tokens": 13739814.0,
      "step": 7815
    },
    {
      "entropy": 5.135254573822022,
      "epoch": 6.128526645768025,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021881173472041926,
      "loss": 4.3284,
      "mean_token_accuracy": 0.24153746366500856,
      "num_tokens": 13748859.0,
      "step": 7820
    },
    {
      "entropy": 5.109288501739502,
      "epoch": 6.132445141065831,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021852052429307924,
      "loss": 4.4259,
      "mean_token_accuracy": 0.2351668357849121,
      "num_tokens": 13757249.0,
      "step": 7825
    },
    {
      "entropy": 5.110639476776123,
      "epoch": 6.136363636363637,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021822941480371995,
      "loss": 4.3873,
      "mean_token_accuracy": 0.236005300283432,
      "num_tokens": 13765838.0,
      "step": 7830
    },
    {
      "entropy": 5.0953569412231445,
      "epoch": 6.140282131661442,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00021793840677260135,
      "loss": 4.3308,
      "mean_token_accuracy": 0.25028188079595565,
      "num_tokens": 13775465.0,
      "step": 7835
    },
    {
      "entropy": 5.159471225738526,
      "epoch": 6.144200626959248,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021764750071980195,
      "loss": 4.4664,
      "mean_token_accuracy": 0.22607876360416412,
      "num_tokens": 13785373.0,
      "step": 7840
    },
    {
      "entropy": 5.131366109848022,
      "epoch": 6.148119122257054,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00021735669716521806,
      "loss": 4.3874,
      "mean_token_accuracy": 0.24585026353597642,
      "num_tokens": 13794812.0,
      "step": 7845
    },
    {
      "entropy": 5.115707683563232,
      "epoch": 6.152037617554859,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021706599662856265,
      "loss": 4.3635,
      "mean_token_accuracy": 0.23479970246553422,
      "num_tokens": 13803681.0,
      "step": 7850
    },
    {
      "entropy": 5.088390970230103,
      "epoch": 6.155956112852665,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021677539962936494,
      "loss": 4.4488,
      "mean_token_accuracy": 0.22844769656658173,
      "num_tokens": 13812772.0,
      "step": 7855
    },
    {
      "entropy": 5.139404916763306,
      "epoch": 6.15987460815047,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021648490668696875,
      "loss": 4.3843,
      "mean_token_accuracy": 0.23509930968284606,
      "num_tokens": 13821280.0,
      "step": 7860
    },
    {
      "entropy": 5.072980737686157,
      "epoch": 6.163793103448276,
      "grad_norm": 1.265625,
      "learning_rate": 0.00021619451832053199,
      "loss": 4.3428,
      "mean_token_accuracy": 0.24063576608896256,
      "num_tokens": 13829717.0,
      "step": 7865
    },
    {
      "entropy": 5.153535223007202,
      "epoch": 6.167711598746082,
      "grad_norm": 1.328125,
      "learning_rate": 0.000215904235049026,
      "loss": 4.4447,
      "mean_token_accuracy": 0.2325928509235382,
      "num_tokens": 13838723.0,
      "step": 7870
    },
    {
      "entropy": 5.116384029388428,
      "epoch": 6.171630094043887,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002156140573912338,
      "loss": 4.4371,
      "mean_token_accuracy": 0.22927117645740508,
      "num_tokens": 13847081.0,
      "step": 7875
    },
    {
      "entropy": 5.142874097824096,
      "epoch": 6.175548589341693,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021532398586574997,
      "loss": 4.3516,
      "mean_token_accuracy": 0.24075478613376616,
      "num_tokens": 13855142.0,
      "step": 7880
    },
    {
      "entropy": 5.179494428634643,
      "epoch": 6.179467084639499,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021503402099097959,
      "loss": 4.4431,
      "mean_token_accuracy": 0.23235059380531312,
      "num_tokens": 13863586.0,
      "step": 7885
    },
    {
      "entropy": 5.1496415615081785,
      "epoch": 6.183385579937304,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021474416328513664,
      "loss": 4.4512,
      "mean_token_accuracy": 0.23158558458089828,
      "num_tokens": 13872286.0,
      "step": 7890
    },
    {
      "entropy": 5.092637825012207,
      "epoch": 6.18730407523511,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002144544132662438,
      "loss": 4.3992,
      "mean_token_accuracy": 0.2320816695690155,
      "num_tokens": 13880866.0,
      "step": 7895
    },
    {
      "entropy": 5.079590320587158,
      "epoch": 6.191222570532915,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002141647714521316,
      "loss": 4.3432,
      "mean_token_accuracy": 0.23558954149484634,
      "num_tokens": 13889015.0,
      "step": 7900
    },
    {
      "entropy": 5.146996259689331,
      "epoch": 6.195141065830721,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00021387523836043676,
      "loss": 4.4264,
      "mean_token_accuracy": 0.23673945814371108,
      "num_tokens": 13898457.0,
      "step": 7905
    },
    {
      "entropy": 5.1425903797149655,
      "epoch": 6.199059561128527,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021358581450860186,
      "loss": 4.3677,
      "mean_token_accuracy": 0.24180345088243485,
      "num_tokens": 13906722.0,
      "step": 7910
    },
    {
      "entropy": 5.099170875549317,
      "epoch": 6.202978056426332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002132965004138742,
      "loss": 4.3287,
      "mean_token_accuracy": 0.24374767243862153,
      "num_tokens": 13914714.0,
      "step": 7915
    },
    {
      "entropy": 5.13126130104065,
      "epoch": 6.206896551724138,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002130072965933052,
      "loss": 4.4026,
      "mean_token_accuracy": 0.2355264738202095,
      "num_tokens": 13923378.0,
      "step": 7920
    },
    {
      "entropy": 5.097534418106079,
      "epoch": 6.210815047021944,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021271820356374876,
      "loss": 4.3936,
      "mean_token_accuracy": 0.24014796018600465,
      "num_tokens": 13931947.0,
      "step": 7925
    },
    {
      "entropy": 5.0743529319763185,
      "epoch": 6.214733542319749,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021242922184186108,
      "loss": 4.304,
      "mean_token_accuracy": 0.23562802225351334,
      "num_tokens": 13940949.0,
      "step": 7930
    },
    {
      "entropy": 5.106280374526977,
      "epoch": 6.218652037617555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002121403519440993,
      "loss": 4.4334,
      "mean_token_accuracy": 0.2348330795764923,
      "num_tokens": 13949944.0,
      "step": 7935
    },
    {
      "entropy": 5.130092144012451,
      "epoch": 6.222570532915361,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00021185159438672085,
      "loss": 4.3816,
      "mean_token_accuracy": 0.23104889839887618,
      "num_tokens": 13958938.0,
      "step": 7940
    },
    {
      "entropy": 5.109782457351685,
      "epoch": 6.226489028213166,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021156294968578232,
      "loss": 4.3255,
      "mean_token_accuracy": 0.2553104802966118,
      "num_tokens": 13967875.0,
      "step": 7945
    },
    {
      "entropy": 5.065059947967529,
      "epoch": 6.230407523510972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021127441835713838,
      "loss": 4.316,
      "mean_token_accuracy": 0.24843193590641022,
      "num_tokens": 13975972.0,
      "step": 7950
    },
    {
      "entropy": 5.085794591903687,
      "epoch": 6.234326018808777,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002109860009164416,
      "loss": 4.3026,
      "mean_token_accuracy": 0.24494756609201432,
      "num_tokens": 13985081.0,
      "step": 7955
    },
    {
      "entropy": 5.07928614616394,
      "epoch": 6.238244514106583,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002106976978791405,
      "loss": 4.4173,
      "mean_token_accuracy": 0.23387945741415023,
      "num_tokens": 13992426.0,
      "step": 7960
    },
    {
      "entropy": 5.056031942367554,
      "epoch": 6.242163009404389,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021040950976047935,
      "loss": 4.3432,
      "mean_token_accuracy": 0.2359875798225403,
      "num_tokens": 14001328.0,
      "step": 7965
    },
    {
      "entropy": 5.050757884979248,
      "epoch": 6.246081504702194,
      "grad_norm": 1.375,
      "learning_rate": 0.00021012143707549726,
      "loss": 4.3279,
      "mean_token_accuracy": 0.2401108354330063,
      "num_tokens": 14010340.0,
      "step": 7970
    },
    {
      "entropy": 5.15937180519104,
      "epoch": 6.25,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020983348033902662,
      "loss": 4.4955,
      "mean_token_accuracy": 0.22596255987882613,
      "num_tokens": 14019213.0,
      "step": 7975
    },
    {
      "entropy": 5.116296625137329,
      "epoch": 6.253918495297806,
      "grad_norm": 1.375,
      "learning_rate": 0.00020954564006569294,
      "loss": 4.4163,
      "mean_token_accuracy": 0.23068010658025742,
      "num_tokens": 14028501.0,
      "step": 7980
    },
    {
      "entropy": 5.080770874023438,
      "epoch": 6.257836990595611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020925791676991318,
      "loss": 4.3376,
      "mean_token_accuracy": 0.23825369477272035,
      "num_tokens": 14037969.0,
      "step": 7985
    },
    {
      "entropy": 5.140538835525513,
      "epoch": 6.261755485893417,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020897031096589586,
      "loss": 4.3536,
      "mean_token_accuracy": 0.24682832658290862,
      "num_tokens": 14046987.0,
      "step": 7990
    },
    {
      "entropy": 5.103993082046509,
      "epoch": 6.265673981191223,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020868282316763902,
      "loss": 4.4159,
      "mean_token_accuracy": 0.22679491192102433,
      "num_tokens": 14055894.0,
      "step": 7995
    },
    {
      "entropy": 5.048171138763427,
      "epoch": 6.269592476489028,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020839545388893,
      "loss": 4.3864,
      "mean_token_accuracy": 0.23758617043495178,
      "num_tokens": 14065119.0,
      "step": 8000
    },
    {
      "epoch": 6.269592476489028,
      "eval_entropy": 5.043674329454585,
      "eval_loss": 5.644860744476318,
      "eval_mean_token_accuracy": 0.17243941113814826,
      "eval_num_tokens": 14065119.0,
      "eval_runtime": 3.0308,
      "eval_samples_per_second": 1360.029,
      "eval_steps_per_second": 170.251,
      "step": 8000
    },
    {
      "entropy": 5.11525068283081,
      "epoch": 6.273510971786834,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020810820364334438,
      "loss": 4.3539,
      "mean_token_accuracy": 0.24338852614164352,
      "num_tokens": 14073320.0,
      "step": 8005
    },
    {
      "entropy": 5.1058127880096436,
      "epoch": 6.277429467084639,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020782107294424484,
      "loss": 4.3873,
      "mean_token_accuracy": 0.23356623202562332,
      "num_tokens": 14081730.0,
      "step": 8010
    },
    {
      "entropy": 5.036461210250854,
      "epoch": 6.281347962382445,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002075340623047804,
      "loss": 4.3543,
      "mean_token_accuracy": 0.23577024191617965,
      "num_tokens": 14090738.0,
      "step": 8015
    },
    {
      "entropy": 5.124470901489258,
      "epoch": 6.285266457680251,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020724717223788574,
      "loss": 4.4399,
      "mean_token_accuracy": 0.2300645425915718,
      "num_tokens": 14099412.0,
      "step": 8020
    },
    {
      "entropy": 5.119640159606933,
      "epoch": 6.289184952978056,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020696040325627991,
      "loss": 4.4333,
      "mean_token_accuracy": 0.23672974854707718,
      "num_tokens": 14108334.0,
      "step": 8025
    },
    {
      "entropy": 5.077015495300293,
      "epoch": 6.293103448275862,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020667375587246563,
      "loss": 4.426,
      "mean_token_accuracy": 0.23580227494239808,
      "num_tokens": 14116853.0,
      "step": 8030
    },
    {
      "entropy": 5.177434825897217,
      "epoch": 6.297021943573668,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020638723059872806,
      "loss": 4.5116,
      "mean_token_accuracy": 0.22475309520959855,
      "num_tokens": 14126735.0,
      "step": 8035
    },
    {
      "entropy": 5.100980997085571,
      "epoch": 6.300940438871473,
      "grad_norm": 1.203125,
      "learning_rate": 0.00020610082794713454,
      "loss": 4.4617,
      "mean_token_accuracy": 0.23313294798135759,
      "num_tokens": 14136749.0,
      "step": 8040
    },
    {
      "entropy": 5.114330053329468,
      "epoch": 6.304858934169279,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002058145484295329,
      "loss": 4.4248,
      "mean_token_accuracy": 0.23660560846328735,
      "num_tokens": 14145204.0,
      "step": 8045
    },
    {
      "entropy": 5.212796974182129,
      "epoch": 6.308777429467085,
      "grad_norm": 1.25,
      "learning_rate": 0.00020552839255755107,
      "loss": 4.5046,
      "mean_token_accuracy": 0.22693784534931183,
      "num_tokens": 14154347.0,
      "step": 8050
    },
    {
      "entropy": 5.137429285049438,
      "epoch": 6.31269592476489,
      "grad_norm": 1.25,
      "learning_rate": 0.00020524236084259613,
      "loss": 4.3923,
      "mean_token_accuracy": 0.2355109304189682,
      "num_tokens": 14163039.0,
      "step": 8055
    },
    {
      "entropy": 5.056146764755249,
      "epoch": 6.316614420062696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00020495645379585303,
      "loss": 4.3386,
      "mean_token_accuracy": 0.24215589165687562,
      "num_tokens": 14172228.0,
      "step": 8060
    },
    {
      "entropy": 5.110491609573364,
      "epoch": 6.320532915360501,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020467067192828404,
      "loss": 4.4693,
      "mean_token_accuracy": 0.2315017268061638,
      "num_tokens": 14181514.0,
      "step": 8065
    },
    {
      "entropy": 5.150668716430664,
      "epoch": 6.324451410658307,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020438501575062752,
      "loss": 4.47,
      "mean_token_accuracy": 0.22673523426055908,
      "num_tokens": 14190654.0,
      "step": 8070
    },
    {
      "entropy": 5.097331523895264,
      "epoch": 6.328369905956113,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020409948577339765,
      "loss": 4.3441,
      "mean_token_accuracy": 0.24160103499889374,
      "num_tokens": 14199177.0,
      "step": 8075
    },
    {
      "entropy": 5.088422632217407,
      "epoch": 6.332288401253918,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020381408250688267,
      "loss": 4.3618,
      "mean_token_accuracy": 0.24132620096206664,
      "num_tokens": 14207643.0,
      "step": 8080
    },
    {
      "entropy": 5.141850852966309,
      "epoch": 6.336206896551724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002035288064611444,
      "loss": 4.4595,
      "mean_token_accuracy": 0.22680164724588395,
      "num_tokens": 14216884.0,
      "step": 8085
    },
    {
      "entropy": 5.141024923324585,
      "epoch": 6.34012539184953,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002032436581460177,
      "loss": 4.4116,
      "mean_token_accuracy": 0.23201296031475066,
      "num_tokens": 14225324.0,
      "step": 8090
    },
    {
      "entropy": 5.128659105300903,
      "epoch": 6.344043887147335,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020295863807110853,
      "loss": 4.3503,
      "mean_token_accuracy": 0.2460414320230484,
      "num_tokens": 14233732.0,
      "step": 8095
    },
    {
      "entropy": 5.200534152984619,
      "epoch": 6.347962382445141,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020267374674579408,
      "loss": 4.4381,
      "mean_token_accuracy": 0.22906262874603273,
      "num_tokens": 14242080.0,
      "step": 8100
    },
    {
      "entropy": 5.089348649978637,
      "epoch": 6.351880877742946,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020238898467922129,
      "loss": 4.4332,
      "mean_token_accuracy": 0.2433960407972336,
      "num_tokens": 14251140.0,
      "step": 8105
    },
    {
      "entropy": 5.115765380859375,
      "epoch": 6.355799373040752,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020210435238030626,
      "loss": 4.473,
      "mean_token_accuracy": 0.2299148291349411,
      "num_tokens": 14259900.0,
      "step": 8110
    },
    {
      "entropy": 5.10223650932312,
      "epoch": 6.359717868338558,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020181985035773304,
      "loss": 4.4128,
      "mean_token_accuracy": 0.23627158850431443,
      "num_tokens": 14268414.0,
      "step": 8115
    },
    {
      "entropy": 5.094730234146118,
      "epoch": 6.363636363636363,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020153547911995272,
      "loss": 4.4454,
      "mean_token_accuracy": 0.2351929947733879,
      "num_tokens": 14277339.0,
      "step": 8120
    },
    {
      "entropy": 5.115366840362549,
      "epoch": 6.367554858934169,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020125123917518308,
      "loss": 4.3656,
      "mean_token_accuracy": 0.24017657935619355,
      "num_tokens": 14285710.0,
      "step": 8125
    },
    {
      "entropy": 5.128287744522095,
      "epoch": 6.371473354231975,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020096713103140673,
      "loss": 4.4044,
      "mean_token_accuracy": 0.23350877910852433,
      "num_tokens": 14294625.0,
      "step": 8130
    },
    {
      "entropy": 5.023825979232788,
      "epoch": 6.37539184952978,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020068315519637103,
      "loss": 4.25,
      "mean_token_accuracy": 0.25359842032194135,
      "num_tokens": 14303030.0,
      "step": 8135
    },
    {
      "entropy": 5.033320665359497,
      "epoch": 6.379310344827586,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000200399312177587,
      "loss": 4.3595,
      "mean_token_accuracy": 0.23693870455026628,
      "num_tokens": 14311826.0,
      "step": 8140
    },
    {
      "entropy": 5.156088018417359,
      "epoch": 6.383228840125392,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020011560248232803,
      "loss": 4.4313,
      "mean_token_accuracy": 0.23777896463871,
      "num_tokens": 14319600.0,
      "step": 8145
    },
    {
      "entropy": 5.074273538589478,
      "epoch": 6.387147335423197,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001998320266176294,
      "loss": 4.3623,
      "mean_token_accuracy": 0.232668237388134,
      "num_tokens": 14328768.0,
      "step": 8150
    },
    {
      "entropy": 5.094604253768921,
      "epoch": 6.391065830721003,
      "grad_norm": 1.25,
      "learning_rate": 0.000199548585090287,
      "loss": 4.4293,
      "mean_token_accuracy": 0.227104988694191,
      "num_tokens": 14338236.0,
      "step": 8155
    },
    {
      "entropy": 5.1627120018005375,
      "epoch": 6.394984326018808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000199265278406857,
      "loss": 4.4986,
      "mean_token_accuracy": 0.22481361031532288,
      "num_tokens": 14347190.0,
      "step": 8160
    },
    {
      "entropy": 5.145858907699585,
      "epoch": 6.398902821316614,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019898210707365428,
      "loss": 4.4392,
      "mean_token_accuracy": 0.23659879267215728,
      "num_tokens": 14355847.0,
      "step": 8165
    },
    {
      "entropy": 5.068314266204834,
      "epoch": 6.40282131661442,
      "grad_norm": 1.28125,
      "learning_rate": 0.00019869907159675187,
      "loss": 4.3711,
      "mean_token_accuracy": 0.2320930004119873,
      "num_tokens": 14364280.0,
      "step": 8170
    },
    {
      "entropy": 5.111385011672974,
      "epoch": 6.406739811912225,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019841617248198017,
      "loss": 4.4006,
      "mean_token_accuracy": 0.23496142774820328,
      "num_tokens": 14372857.0,
      "step": 8175
    },
    {
      "entropy": 5.086836147308349,
      "epoch": 6.410658307210031,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001981334102349258,
      "loss": 4.3515,
      "mean_token_accuracy": 0.2394037589430809,
      "num_tokens": 14381157.0,
      "step": 8180
    },
    {
      "entropy": 5.138228559494019,
      "epoch": 6.414576802507837,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019785078536093065,
      "loss": 4.4631,
      "mean_token_accuracy": 0.2266973525285721,
      "num_tokens": 14390404.0,
      "step": 8185
    },
    {
      "entropy": 5.134257745742798,
      "epoch": 6.418495297805642,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019756829836509117,
      "loss": 4.4405,
      "mean_token_accuracy": 0.22731270492076874,
      "num_tokens": 14399426.0,
      "step": 8190
    },
    {
      "entropy": 5.088931608200073,
      "epoch": 6.422413793103448,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001972859497522575,
      "loss": 4.3334,
      "mean_token_accuracy": 0.23154934048652648,
      "num_tokens": 14408233.0,
      "step": 8195
    },
    {
      "entropy": 5.115785455703735,
      "epoch": 6.4263322884012535,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001970037400270323,
      "loss": 4.4056,
      "mean_token_accuracy": 0.23317165076732635,
      "num_tokens": 14416828.0,
      "step": 8200
    },
    {
      "entropy": 5.134962368011474,
      "epoch": 6.4302507836990594,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001967216696937702,
      "loss": 4.4363,
      "mean_token_accuracy": 0.23197238743305207,
      "num_tokens": 14425590.0,
      "step": 8205
    },
    {
      "entropy": 5.176138591766358,
      "epoch": 6.434169278996865,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019643973925657666,
      "loss": 4.5574,
      "mean_token_accuracy": 0.21958873867988588,
      "num_tokens": 14434646.0,
      "step": 8210
    },
    {
      "entropy": 5.174251079559326,
      "epoch": 6.4380877742946705,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019615794921930697,
      "loss": 4.448,
      "mean_token_accuracy": 0.23234059065580367,
      "num_tokens": 14443862.0,
      "step": 8215
    },
    {
      "entropy": 5.1017499446868895,
      "epoch": 6.4420062695924765,
      "grad_norm": 1.375,
      "learning_rate": 0.00019587630008556563,
      "loss": 4.3606,
      "mean_token_accuracy": 0.24099775701761245,
      "num_tokens": 14452813.0,
      "step": 8220
    },
    {
      "entropy": 5.132601261138916,
      "epoch": 6.445924764890282,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019559479235870537,
      "loss": 4.4499,
      "mean_token_accuracy": 0.22990341782569884,
      "num_tokens": 14461912.0,
      "step": 8225
    },
    {
      "entropy": 5.171674489974976,
      "epoch": 6.4498432601880875,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001953134265418261,
      "loss": 4.3952,
      "mean_token_accuracy": 0.2316601112484932,
      "num_tokens": 14470460.0,
      "step": 8230
    },
    {
      "entropy": 5.134877920150757,
      "epoch": 6.4537617554858935,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001950322031377742,
      "loss": 4.3967,
      "mean_token_accuracy": 0.2399676948785782,
      "num_tokens": 14479258.0,
      "step": 8235
    },
    {
      "entropy": 5.11114730834961,
      "epoch": 6.4576802507836994,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019475112264914142,
      "loss": 4.4084,
      "mean_token_accuracy": 0.24252817332744597,
      "num_tokens": 14488209.0,
      "step": 8240
    },
    {
      "entropy": 5.027699899673462,
      "epoch": 6.4615987460815045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019447018557826424,
      "loss": 4.3638,
      "mean_token_accuracy": 0.23228035122156143,
      "num_tokens": 14497234.0,
      "step": 8245
    },
    {
      "entropy": 5.0298449993133545,
      "epoch": 6.4655172413793105,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001941893924272227,
      "loss": 4.2817,
      "mean_token_accuracy": 0.24712865501642228,
      "num_tokens": 14506374.0,
      "step": 8250
    },
    {
      "entropy": 5.114639520645142,
      "epoch": 6.469435736677116,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019390874369783974,
      "loss": 4.4293,
      "mean_token_accuracy": 0.2308681681752205,
      "num_tokens": 14514796.0,
      "step": 8255
    },
    {
      "entropy": 5.116225814819336,
      "epoch": 6.4733542319749215,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019362823989168006,
      "loss": 4.404,
      "mean_token_accuracy": 0.23778246194124222,
      "num_tokens": 14523771.0,
      "step": 8260
    },
    {
      "entropy": 5.073237609863281,
      "epoch": 6.4772727272727275,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019334788151004962,
      "loss": 4.3111,
      "mean_token_accuracy": 0.24790607690811156,
      "num_tokens": 14532130.0,
      "step": 8265
    },
    {
      "entropy": 4.994515800476075,
      "epoch": 6.481191222570533,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019306766905399415,
      "loss": 4.3416,
      "mean_token_accuracy": 0.23662912249565124,
      "num_tokens": 14540278.0,
      "step": 8270
    },
    {
      "entropy": 5.104821014404297,
      "epoch": 6.485109717868339,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019278760302429872,
      "loss": 4.421,
      "mean_token_accuracy": 0.22528692036867143,
      "num_tokens": 14549519.0,
      "step": 8275
    },
    {
      "entropy": 5.102330350875855,
      "epoch": 6.4890282131661445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019250768392148682,
      "loss": 4.4702,
      "mean_token_accuracy": 0.23372383862733842,
      "num_tokens": 14558576.0,
      "step": 8280
    },
    {
      "entropy": 5.073139190673828,
      "epoch": 6.49294670846395,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019222791224581925,
      "loss": 4.3705,
      "mean_token_accuracy": 0.2378414660692215,
      "num_tokens": 14567544.0,
      "step": 8285
    },
    {
      "entropy": 5.067892599105835,
      "epoch": 6.496865203761756,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019194828849729336,
      "loss": 4.3635,
      "mean_token_accuracy": 0.22996239811182023,
      "num_tokens": 14576662.0,
      "step": 8290
    },
    {
      "entropy": 5.05437479019165,
      "epoch": 6.500783699059561,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019166881317564217,
      "loss": 4.3616,
      "mean_token_accuracy": 0.23195963054895402,
      "num_tokens": 14585163.0,
      "step": 8295
    },
    {
      "entropy": 5.036677360534668,
      "epoch": 6.504702194357367,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019138948678033336,
      "loss": 4.3153,
      "mean_token_accuracy": 0.24747663587331772,
      "num_tokens": 14593213.0,
      "step": 8300
    },
    {
      "entropy": 5.073464202880859,
      "epoch": 6.508620689655173,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019111030981056832,
      "loss": 4.3867,
      "mean_token_accuracy": 0.23868631869554519,
      "num_tokens": 14601479.0,
      "step": 8305
    },
    {
      "entropy": 5.13361873626709,
      "epoch": 6.512539184952978,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001908312827652819,
      "loss": 4.3922,
      "mean_token_accuracy": 0.23098443299531937,
      "num_tokens": 14610418.0,
      "step": 8310
    },
    {
      "entropy": 5.082857894897461,
      "epoch": 6.516457680250784,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019055240614314047,
      "loss": 4.4118,
      "mean_token_accuracy": 0.23798256516456603,
      "num_tokens": 14618387.0,
      "step": 8315
    },
    {
      "entropy": 5.071300601959228,
      "epoch": 6.52037617554859,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019027368044254177,
      "loss": 4.3344,
      "mean_token_accuracy": 0.2414350464940071,
      "num_tokens": 14627003.0,
      "step": 8320
    },
    {
      "entropy": 5.107181739807129,
      "epoch": 6.524294670846395,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001899951061616138,
      "loss": 4.4097,
      "mean_token_accuracy": 0.23566458225250245,
      "num_tokens": 14635825.0,
      "step": 8325
    },
    {
      "entropy": 5.069835758209228,
      "epoch": 6.528213166144201,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018971668379821405,
      "loss": 4.3564,
      "mean_token_accuracy": 0.236721171438694,
      "num_tokens": 14644830.0,
      "step": 8330
    },
    {
      "entropy": 5.043557167053223,
      "epoch": 6.532131661442007,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001894384138499284,
      "loss": 4.4081,
      "mean_token_accuracy": 0.233300119638443,
      "num_tokens": 14653633.0,
      "step": 8335
    },
    {
      "entropy": 5.040063905715942,
      "epoch": 6.536050156739812,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018916029681407032,
      "loss": 4.2594,
      "mean_token_accuracy": 0.2562505915760994,
      "num_tokens": 14662212.0,
      "step": 8340
    },
    {
      "entropy": 5.086932945251465,
      "epoch": 6.539968652037618,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001888823331876801,
      "loss": 4.4034,
      "mean_token_accuracy": 0.23681754469871522,
      "num_tokens": 14670942.0,
      "step": 8345
    },
    {
      "entropy": 5.133782434463501,
      "epoch": 6.543887147335424,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018860452346752382,
      "loss": 4.3504,
      "mean_token_accuracy": 0.2368352547287941,
      "num_tokens": 14679167.0,
      "step": 8350
    },
    {
      "entropy": 5.0393805503845215,
      "epoch": 6.547805642633229,
      "grad_norm": 1.25,
      "learning_rate": 0.00018832686815009242,
      "loss": 4.3984,
      "mean_token_accuracy": 0.23250824213027954,
      "num_tokens": 14688188.0,
      "step": 8355
    },
    {
      "entropy": 5.134057998657227,
      "epoch": 6.551724137931035,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018804936773160108,
      "loss": 4.4707,
      "mean_token_accuracy": 0.22669147402048112,
      "num_tokens": 14696958.0,
      "step": 8360
    },
    {
      "entropy": 5.073854684829712,
      "epoch": 6.55564263322884,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000187772022707988,
      "loss": 4.3976,
      "mean_token_accuracy": 0.2285067230463028,
      "num_tokens": 14705524.0,
      "step": 8365
    },
    {
      "entropy": 5.091714668273926,
      "epoch": 6.559561128526646,
      "grad_norm": 1.328125,
      "learning_rate": 0.00018749483357491365,
      "loss": 4.4532,
      "mean_token_accuracy": 0.22631947249174117,
      "num_tokens": 14714500.0,
      "step": 8370
    },
    {
      "entropy": 5.122143173217774,
      "epoch": 6.563479623824452,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018721780082775997,
      "loss": 4.4142,
      "mean_token_accuracy": 0.23164458870887755,
      "num_tokens": 14722816.0,
      "step": 8375
    },
    {
      "entropy": 5.125632190704346,
      "epoch": 6.567398119122257,
      "grad_norm": 1.34375,
      "learning_rate": 0.00018694092496162945,
      "loss": 4.4223,
      "mean_token_accuracy": 0.24054589718580247,
      "num_tokens": 14731453.0,
      "step": 8380
    },
    {
      "entropy": 5.150288867950439,
      "epoch": 6.571316614420063,
      "grad_norm": 1.421875,
      "learning_rate": 0.00018666420647134419,
      "loss": 4.4106,
      "mean_token_accuracy": 0.2437850534915924,
      "num_tokens": 14739924.0,
      "step": 8385
    },
    {
      "entropy": 4.99471526145935,
      "epoch": 6.575235109717868,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001863876458514448,
      "loss": 4.2815,
      "mean_token_accuracy": 0.24719616323709487,
      "num_tokens": 14748356.0,
      "step": 8390
    },
    {
      "entropy": 5.156424474716187,
      "epoch": 6.579153605015674,
      "grad_norm": 1.171875,
      "learning_rate": 0.00018611124359619024,
      "loss": 4.4689,
      "mean_token_accuracy": 0.23101072013378143,
      "num_tokens": 14758846.0,
      "step": 8395
    },
    {
      "entropy": 5.016350317001343,
      "epoch": 6.58307210031348,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018583500019955595,
      "loss": 4.287,
      "mean_token_accuracy": 0.24313644468784332,
      "num_tokens": 14767030.0,
      "step": 8400
    },
    {
      "entropy": 5.104779243469238,
      "epoch": 6.586990595611285,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018555891615523374,
      "loss": 4.3652,
      "mean_token_accuracy": 0.2391281932592392,
      "num_tokens": 14776774.0,
      "step": 8405
    },
    {
      "entropy": 5.07826452255249,
      "epoch": 6.590909090909091,
      "grad_norm": 1.359375,
      "learning_rate": 0.00018528299195663046,
      "loss": 4.3499,
      "mean_token_accuracy": 0.24164195656776427,
      "num_tokens": 14785662.0,
      "step": 8410
    },
    {
      "entropy": 5.14148268699646,
      "epoch": 6.594827586206897,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001850072280968676,
      "loss": 4.4857,
      "mean_token_accuracy": 0.2214024320244789,
      "num_tokens": 14795070.0,
      "step": 8415
    },
    {
      "entropy": 5.114009380340576,
      "epoch": 6.598746081504702,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00018473162506877977,
      "loss": 4.3963,
      "mean_token_accuracy": 0.23532192409038544,
      "num_tokens": 14805452.0,
      "step": 8420
    },
    {
      "entropy": 5.088995552062988,
      "epoch": 6.602664576802508,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018445618336491415,
      "loss": 4.3818,
      "mean_token_accuracy": 0.24365420043468475,
      "num_tokens": 14815632.0,
      "step": 8425
    },
    {
      "entropy": 5.056402492523193,
      "epoch": 6.606583072100314,
      "grad_norm": 1.28125,
      "learning_rate": 0.00018418090347752997,
      "loss": 4.338,
      "mean_token_accuracy": 0.23936591893434525,
      "num_tokens": 14824228.0,
      "step": 8430
    },
    {
      "entropy": 5.08762936592102,
      "epoch": 6.610501567398119,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001839057858985968,
      "loss": 4.4546,
      "mean_token_accuracy": 0.23178823292255402,
      "num_tokens": 14833609.0,
      "step": 8435
    },
    {
      "entropy": 5.085925006866455,
      "epoch": 6.614420062695925,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018363083111979452,
      "loss": 4.3874,
      "mean_token_accuracy": 0.23928953111171722,
      "num_tokens": 14841967.0,
      "step": 8440
    },
    {
      "entropy": 5.121152114868164,
      "epoch": 6.618338557993731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00018335603963251184,
      "loss": 4.4355,
      "mean_token_accuracy": 0.2274794414639473,
      "num_tokens": 14851472.0,
      "step": 8445
    },
    {
      "entropy": 5.165384769439697,
      "epoch": 6.622257053291536,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00018308141192784573,
      "loss": 4.4049,
      "mean_token_accuracy": 0.2388998255133629,
      "num_tokens": 14859672.0,
      "step": 8450
    },
    {
      "entropy": 5.117183065414428,
      "epoch": 6.626175548589342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018280694849660034,
      "loss": 4.3261,
      "mean_token_accuracy": 0.24858736544847487,
      "num_tokens": 14867919.0,
      "step": 8455
    },
    {
      "entropy": 5.094011068344116,
      "epoch": 6.630094043887147,
      "grad_norm": 1.40625,
      "learning_rate": 0.00018253264982928636,
      "loss": 4.322,
      "mean_token_accuracy": 0.24692283868789672,
      "num_tokens": 14875670.0,
      "step": 8460
    },
    {
      "entropy": 5.101661348342896,
      "epoch": 6.634012539184953,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00018225851641612007,
      "loss": 4.3659,
      "mean_token_accuracy": 0.23462289571762085,
      "num_tokens": 14883835.0,
      "step": 8465
    },
    {
      "entropy": 5.037274312973023,
      "epoch": 6.637931034482759,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001819845487470223,
      "loss": 4.3155,
      "mean_token_accuracy": 0.23845092356204986,
      "num_tokens": 14892296.0,
      "step": 8470
    },
    {
      "entropy": 5.038182067871094,
      "epoch": 6.641849529780564,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00018171074731161758,
      "loss": 4.3653,
      "mean_token_accuracy": 0.24707158356904985,
      "num_tokens": 14901334.0,
      "step": 8475
    },
    {
      "entropy": 5.040551280975341,
      "epoch": 6.64576802507837,
      "grad_norm": 1.25,
      "learning_rate": 0.00018143711259923363,
      "loss": 4.4056,
      "mean_token_accuracy": 0.23082399219274521,
      "num_tokens": 14910327.0,
      "step": 8480
    },
    {
      "entropy": 5.117019510269165,
      "epoch": 6.649686520376176,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001811636450989,
      "loss": 4.4849,
      "mean_token_accuracy": 0.22831751108169557,
      "num_tokens": 14919513.0,
      "step": 8485
    },
    {
      "entropy": 5.022596979141236,
      "epoch": 6.653605015673981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001808903452993474,
      "loss": 4.2525,
      "mean_token_accuracy": 0.24327975064516066,
      "num_tokens": 14928127.0,
      "step": 8490
    },
    {
      "entropy": 5.080018043518066,
      "epoch": 6.657523510971787,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001806172136890068,
      "loss": 4.3565,
      "mean_token_accuracy": 0.2419793576002121,
      "num_tokens": 14936771.0,
      "step": 8495
    },
    {
      "entropy": 5.091134548187256,
      "epoch": 6.661442006269592,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001803442507560089,
      "loss": 4.3447,
      "mean_token_accuracy": 0.23821248561143876,
      "num_tokens": 14945423.0,
      "step": 8500
    },
    {
      "epoch": 6.661442006269592,
      "eval_entropy": 4.991707536601281,
      "eval_loss": 5.630890369415283,
      "eval_mean_token_accuracy": 0.17434056197083736,
      "eval_num_tokens": 14945423.0,
      "eval_runtime": 2.8493,
      "eval_samples_per_second": 1446.646,
      "eval_steps_per_second": 181.094,
      "step": 8500
    },
    {
      "entropy": 5.032188892364502,
      "epoch": 6.665360501567398,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001800714569881825,
      "loss": 4.3719,
      "mean_token_accuracy": 0.24152286499738693,
      "num_tokens": 14953859.0,
      "step": 8505
    },
    {
      "entropy": 5.023387336730957,
      "epoch": 6.669278996865204,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017979883287305454,
      "loss": 4.3615,
      "mean_token_accuracy": 0.24197774827480317,
      "num_tokens": 14962598.0,
      "step": 8510
    },
    {
      "entropy": 5.099363517761231,
      "epoch": 6.673197492163009,
      "grad_norm": 1.265625,
      "learning_rate": 0.00017952637889784824,
      "loss": 4.3568,
      "mean_token_accuracy": 0.24211505204439163,
      "num_tokens": 14971606.0,
      "step": 8515
    },
    {
      "entropy": 5.040920829772949,
      "epoch": 6.677115987460815,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017925409554948329,
      "loss": 4.3828,
      "mean_token_accuracy": 0.24452775418758393,
      "num_tokens": 14979686.0,
      "step": 8520
    },
    {
      "entropy": 5.105930233001709,
      "epoch": 6.681034482758621,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00017898198331457393,
      "loss": 4.4701,
      "mean_token_accuracy": 0.22297748327255248,
      "num_tokens": 14988977.0,
      "step": 8525
    },
    {
      "entropy": 5.121250486373901,
      "epoch": 6.684952978056426,
      "grad_norm": 1.375,
      "learning_rate": 0.00017871004267942908,
      "loss": 4.4738,
      "mean_token_accuracy": 0.22734981328248977,
      "num_tokens": 14996871.0,
      "step": 8530
    },
    {
      "entropy": 5.129722356796265,
      "epoch": 6.688871473354232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00017843827413005064,
      "loss": 4.4053,
      "mean_token_accuracy": 0.24080052673816682,
      "num_tokens": 15006022.0,
      "step": 8535
    },
    {
      "entropy": 5.062245559692383,
      "epoch": 6.692789968652038,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001781666781521331,
      "loss": 4.433,
      "mean_token_accuracy": 0.23650194555521012,
      "num_tokens": 15014827.0,
      "step": 8540
    },
    {
      "entropy": 5.058799600601196,
      "epoch": 6.696708463949843,
      "grad_norm": 1.46875,
      "learning_rate": 0.00017789525523106243,
      "loss": 4.3265,
      "mean_token_accuracy": 0.24040284454822541,
      "num_tokens": 15023157.0,
      "step": 8545
    },
    {
      "entropy": 5.111304140090942,
      "epoch": 6.700626959247649,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00017762400585191545,
      "loss": 4.3822,
      "mean_token_accuracy": 0.23542432636022567,
      "num_tokens": 15032539.0,
      "step": 8550
    },
    {
      "entropy": 5.1514040470123295,
      "epoch": 6.704545454545455,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017735293049945888,
      "loss": 4.5414,
      "mean_token_accuracy": 0.22490991055965423,
      "num_tokens": 15041527.0,
      "step": 8555
    },
    {
      "entropy": 5.086416482925415,
      "epoch": 6.70846394984326,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00017708202965814807,
      "loss": 4.3906,
      "mean_token_accuracy": 0.235493566095829,
      "num_tokens": 15050794.0,
      "step": 8560
    },
    {
      "entropy": 5.074027681350708,
      "epoch": 6.712382445141066,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00017681130381212707,
      "loss": 4.3609,
      "mean_token_accuracy": 0.2452659547328949,
      "num_tokens": 15059357.0,
      "step": 8565
    },
    {
      "entropy": 5.142292165756226,
      "epoch": 6.716300940438871,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00017654075344522652,
      "loss": 4.4765,
      "mean_token_accuracy": 0.2307179495692253,
      "num_tokens": 15067668.0,
      "step": 8570
    },
    {
      "entropy": 5.12802734375,
      "epoch": 6.720219435736677,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017627037904096416,
      "loss": 4.3659,
      "mean_token_accuracy": 0.2371155232191086,
      "num_tokens": 15076463.0,
      "step": 8575
    },
    {
      "entropy": 5.040061521530151,
      "epoch": 6.724137931034483,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017600018108254257,
      "loss": 4.3575,
      "mean_token_accuracy": 0.23154978454113007,
      "num_tokens": 15085024.0,
      "step": 8580
    },
    {
      "entropy": 5.045459794998169,
      "epoch": 6.728056426332288,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017573016005284958,
      "loss": 4.3549,
      "mean_token_accuracy": 0.2397107720375061,
      "num_tokens": 15093735.0,
      "step": 8585
    },
    {
      "entropy": 5.094174575805664,
      "epoch": 6.731974921630094,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017546031643445636,
      "loss": 4.3731,
      "mean_token_accuracy": 0.2348713234066963,
      "num_tokens": 15102274.0,
      "step": 8590
    },
    {
      "entropy": 5.112586450576782,
      "epoch": 6.735893416927899,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001751906507096172,
      "loss": 4.3853,
      "mean_token_accuracy": 0.2394071877002716,
      "num_tokens": 15110762.0,
      "step": 8595
    },
    {
      "entropy": 5.028608655929565,
      "epoch": 6.739811912225705,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001749211633602687,
      "loss": 4.3676,
      "mean_token_accuracy": 0.2381587103009224,
      "num_tokens": 15119104.0,
      "step": 8600
    },
    {
      "entropy": 5.055168867111206,
      "epoch": 6.743730407523511,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017465185486802826,
      "loss": 4.3449,
      "mean_token_accuracy": 0.24613660722970962,
      "num_tokens": 15127351.0,
      "step": 8605
    },
    {
      "entropy": 5.160193300247192,
      "epoch": 6.747648902821316,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00017438272571419373,
      "loss": 4.4883,
      "mean_token_accuracy": 0.22871676236391067,
      "num_tokens": 15136246.0,
      "step": 8610
    },
    {
      "entropy": 5.129608011245727,
      "epoch": 6.751567398119122,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001741137763797428,
      "loss": 4.4467,
      "mean_token_accuracy": 0.2313600718975067,
      "num_tokens": 15144633.0,
      "step": 8615
    },
    {
      "entropy": 5.092439985275268,
      "epoch": 6.755485893416928,
      "grad_norm": 1.25,
      "learning_rate": 0.00017384500734533137,
      "loss": 4.3969,
      "mean_token_accuracy": 0.23402444124221802,
      "num_tokens": 15153496.0,
      "step": 8620
    },
    {
      "entropy": 5.065684413909912,
      "epoch": 6.759404388714733,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00017357641909129324,
      "loss": 4.3837,
      "mean_token_accuracy": 0.23546793460845947,
      "num_tokens": 15162242.0,
      "step": 8625
    },
    {
      "entropy": 5.131422662734986,
      "epoch": 6.763322884012539,
      "grad_norm": 1.3125,
      "learning_rate": 0.00017330801209763938,
      "loss": 4.4413,
      "mean_token_accuracy": 0.23425622284412384,
      "num_tokens": 15170001.0,
      "step": 8630
    },
    {
      "entropy": 5.093270206451416,
      "epoch": 6.767241379310345,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017303978684405635,
      "loss": 4.3698,
      "mean_token_accuracy": 0.2367086172103882,
      "num_tokens": 15178086.0,
      "step": 8635
    },
    {
      "entropy": 5.1080069065094,
      "epoch": 6.77115987460815,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017277174380990647,
      "loss": 4.4417,
      "mean_token_accuracy": 0.2273123010993004,
      "num_tokens": 15187087.0,
      "step": 8640
    },
    {
      "entropy": 5.142852449417115,
      "epoch": 6.775078369905956,
      "grad_norm": 1.40625,
      "learning_rate": 0.00017250388347422578,
      "loss": 4.4002,
      "mean_token_accuracy": 0.2393433541059494,
      "num_tokens": 15195850.0,
      "step": 8645
    },
    {
      "entropy": 5.13836259841919,
      "epoch": 6.778996865203762,
      "grad_norm": 1.328125,
      "learning_rate": 0.00017223620631572433,
      "loss": 4.435,
      "mean_token_accuracy": 0.23817990571260453,
      "num_tokens": 15204174.0,
      "step": 8650
    },
    {
      "entropy": 5.115660572052002,
      "epoch": 6.782915360501567,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017196871281278465,
      "loss": 4.3118,
      "mean_token_accuracy": 0.24596981555223466,
      "num_tokens": 15212640.0,
      "step": 8655
    },
    {
      "entropy": 5.1152660846710205,
      "epoch": 6.786833855799373,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00017170140344346086,
      "loss": 4.3683,
      "mean_token_accuracy": 0.23078155815601348,
      "num_tokens": 15220996.0,
      "step": 8660
    },
    {
      "entropy": 5.077930927276611,
      "epoch": 6.790752351097178,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001714342786854781,
      "loss": 4.4191,
      "mean_token_accuracy": 0.2325994551181793,
      "num_tokens": 15229511.0,
      "step": 8665
    },
    {
      "entropy": 5.105262088775635,
      "epoch": 6.794670846394984,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017116733901623182,
      "loss": 4.4635,
      "mean_token_accuracy": 0.2317218914628029,
      "num_tokens": 15238673.0,
      "step": 8670
    },
    {
      "entropy": 5.113886737823487,
      "epoch": 6.79858934169279,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00017090058491278637,
      "loss": 4.4099,
      "mean_token_accuracy": 0.22661271393299104,
      "num_tokens": 15246717.0,
      "step": 8675
    },
    {
      "entropy": 5.112394618988037,
      "epoch": 6.802507836990595,
      "grad_norm": 1.34375,
      "learning_rate": 0.00017063401685187457,
      "loss": 4.3802,
      "mean_token_accuracy": 0.23107732385396956,
      "num_tokens": 15255543.0,
      "step": 8680
    },
    {
      "entropy": 5.035405158996582,
      "epoch": 6.806426332288401,
      "grad_norm": 1.296875,
      "learning_rate": 0.00017036763530989688,
      "loss": 4.3055,
      "mean_token_accuracy": 0.24495591521263121,
      "num_tokens": 15264275.0,
      "step": 8685
    },
    {
      "entropy": 5.10296368598938,
      "epoch": 6.810344827586206,
      "grad_norm": 1.4375,
      "learning_rate": 0.00017010144076292012,
      "loss": 4.3984,
      "mean_token_accuracy": 0.23774681240320206,
      "num_tokens": 15272376.0,
      "step": 8690
    },
    {
      "entropy": 5.142447662353516,
      "epoch": 6.814263322884012,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016983543368667722,
      "loss": 4.4584,
      "mean_token_accuracy": 0.22442379146814345,
      "num_tokens": 15282362.0,
      "step": 8695
    },
    {
      "entropy": 5.087182807922363,
      "epoch": 6.818181818181818,
      "grad_norm": 1.3125,
      "learning_rate": 0.00016956961455656607,
      "loss": 4.4288,
      "mean_token_accuracy": 0.2258739560842514,
      "num_tokens": 15290934.0,
      "step": 8700
    },
    {
      "entropy": 5.057226037979126,
      "epoch": 6.822100313479623,
      "grad_norm": 1.359375,
      "learning_rate": 0.00016930398384764836,
      "loss": 4.3256,
      "mean_token_accuracy": 0.25071902126073836,
      "num_tokens": 15299743.0,
      "step": 8705
    },
    {
      "entropy": 5.10224666595459,
      "epoch": 6.826018808777429,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016903854203464939,
      "loss": 4.3943,
      "mean_token_accuracy": 0.2365369215607643,
      "num_tokens": 15307858.0,
      "step": 8710
    },
    {
      "entropy": 5.084202003479004,
      "epoch": 6.829937304075235,
      "grad_norm": 1.40625,
      "learning_rate": 0.00016877328959195653,
      "loss": 4.3738,
      "mean_token_accuracy": 0.23816307485103608,
      "num_tokens": 15316829.0,
      "step": 8715
    },
    {
      "entropy": 5.152422380447388,
      "epoch": 6.83385579937304,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00016850822699361913,
      "loss": 4.4862,
      "mean_token_accuracy": 0.22851022183895112,
      "num_tokens": 15325394.0,
      "step": 8720
    },
    {
      "entropy": 5.077306127548217,
      "epoch": 6.837774294670846,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001682433547133469,
      "loss": 4.3799,
      "mean_token_accuracy": 0.23447758406400682,
      "num_tokens": 15333988.0,
      "step": 8725
    },
    {
      "entropy": 5.130360794067383,
      "epoch": 6.841692789968652,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016797867322450948,
      "loss": 4.4301,
      "mean_token_accuracy": 0.23289658278226852,
      "num_tokens": 15343107.0,
      "step": 8730
    },
    {
      "entropy": 5.143571853637695,
      "epoch": 6.8456112852664575,
      "grad_norm": 1.5,
      "learning_rate": 0.0001677141830001358,
      "loss": 4.3674,
      "mean_token_accuracy": 0.23583883047103882,
      "num_tokens": 15351440.0,
      "step": 8735
    },
    {
      "entropy": 5.047386837005615,
      "epoch": 6.849529780564263,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016744988451291266,
      "loss": 4.3366,
      "mean_token_accuracy": 0.23386137932538986,
      "num_tokens": 15360426.0,
      "step": 8740
    },
    {
      "entropy": 5.075678920745849,
      "epoch": 6.853448275862069,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016718577823518426,
      "loss": 4.4025,
      "mean_token_accuracy": 0.23798677027225495,
      "num_tokens": 15369252.0,
      "step": 8745
    },
    {
      "entropy": 5.109200477600098,
      "epoch": 6.8573667711598745,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00016692186463895136,
      "loss": 4.3985,
      "mean_token_accuracy": 0.22947316914796828,
      "num_tokens": 15377672.0,
      "step": 8750
    },
    {
      "entropy": 5.077537393569946,
      "epoch": 6.86128526645768,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016665814419587055,
      "loss": 4.3254,
      "mean_token_accuracy": 0.2436169058084488,
      "num_tokens": 15385585.0,
      "step": 8755
    },
    {
      "entropy": 5.112135314941407,
      "epoch": 6.8652037617554855,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001663946173772528,
      "loss": 4.4041,
      "mean_token_accuracy": 0.23270722031593322,
      "num_tokens": 15394193.0,
      "step": 8760
    },
    {
      "entropy": 5.1495672225952145,
      "epoch": 6.8691222570532915,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016613128465406329,
      "loss": 4.4356,
      "mean_token_accuracy": 0.22918337732553482,
      "num_tokens": 15402498.0,
      "step": 8765
    },
    {
      "entropy": 5.0264787673950195,
      "epoch": 6.8730407523510975,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001658681464969204,
      "loss": 4.3037,
      "mean_token_accuracy": 0.247980298101902,
      "num_tokens": 15410893.0,
      "step": 8770
    },
    {
      "entropy": 5.083421039581299,
      "epoch": 6.8769592476489025,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016560520337609465,
      "loss": 4.4704,
      "mean_token_accuracy": 0.2233293116092682,
      "num_tokens": 15420024.0,
      "step": 8775
    },
    {
      "entropy": 5.018947172164917,
      "epoch": 6.8808777429467085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00016534245576150787,
      "loss": 4.3838,
      "mean_token_accuracy": 0.23147960305213927,
      "num_tokens": 15428474.0,
      "step": 8780
    },
    {
      "entropy": 5.06837797164917,
      "epoch": 6.8847962382445145,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001650799041227329,
      "loss": 4.3049,
      "mean_token_accuracy": 0.24483208656311034,
      "num_tokens": 15437019.0,
      "step": 8785
    },
    {
      "entropy": 5.142296361923218,
      "epoch": 6.88871473354232,
      "grad_norm": 1.515625,
      "learning_rate": 0.00016481754892899193,
      "loss": 4.4224,
      "mean_token_accuracy": 0.24000488370656967,
      "num_tokens": 15445838.0,
      "step": 8790
    },
    {
      "entropy": 5.124936199188232,
      "epoch": 6.8926332288401255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001645553906491563,
      "loss": 4.4221,
      "mean_token_accuracy": 0.2369311198592186,
      "num_tokens": 15454378.0,
      "step": 8795
    },
    {
      "entropy": 5.098961114883423,
      "epoch": 6.896551724137931,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001642934297517452,
      "loss": 4.4053,
      "mean_token_accuracy": 0.23084117323160172,
      "num_tokens": 15462447.0,
      "step": 8800
    },
    {
      "entropy": 5.06881799697876,
      "epoch": 6.900470219435737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001640316667049254,
      "loss": 4.3954,
      "mean_token_accuracy": 0.2295443370938301,
      "num_tokens": 15471646.0,
      "step": 8805
    },
    {
      "entropy": 5.107182121276855,
      "epoch": 6.9043887147335425,
      "grad_norm": 1.140625,
      "learning_rate": 0.00016377010197650993,
      "loss": 4.3875,
      "mean_token_accuracy": 0.2352862611413002,
      "num_tokens": 15481945.0,
      "step": 8810
    },
    {
      "entropy": 5.113076066970825,
      "epoch": 6.908307210031348,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001635087360339572,
      "loss": 4.3431,
      "mean_token_accuracy": 0.24779405891895295,
      "num_tokens": 15490558.0,
      "step": 8815
    },
    {
      "entropy": 5.100098180770874,
      "epoch": 6.912225705329154,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016324756934437074,
      "loss": 4.3651,
      "mean_token_accuracy": 0.23864475637674332,
      "num_tokens": 15499150.0,
      "step": 8820
    },
    {
      "entropy": 5.046548223495483,
      "epoch": 6.91614420062696,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001629866023744977,
      "loss": 4.3521,
      "mean_token_accuracy": 0.23591502904891967,
      "num_tokens": 15508736.0,
      "step": 8825
    },
    {
      "entropy": 5.15428204536438,
      "epoch": 6.920062695924765,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00016272583559072832,
      "loss": 4.4211,
      "mean_token_accuracy": 0.23936883062124253,
      "num_tokens": 15517445.0,
      "step": 8830
    },
    {
      "entropy": 5.1209944725036625,
      "epoch": 6.923981191222571,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016246526945909505,
      "loss": 4.4252,
      "mean_token_accuracy": 0.23772648423910142,
      "num_tokens": 15526282.0,
      "step": 8835
    },
    {
      "entropy": 5.14006290435791,
      "epoch": 6.927899686520377,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00016220490444527198,
      "loss": 4.4037,
      "mean_token_accuracy": 0.23961792886257172,
      "num_tokens": 15534776.0,
      "step": 8840
    },
    {
      "entropy": 5.087289142608642,
      "epoch": 6.931818181818182,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001619447410145735,
      "loss": 4.3784,
      "mean_token_accuracy": 0.2384671151638031,
      "num_tokens": 15543173.0,
      "step": 8845
    },
    {
      "entropy": 5.04956021308899,
      "epoch": 6.935736677115988,
      "grad_norm": 1.296875,
      "learning_rate": 0.00016168477963195382,
      "loss": 4.3222,
      "mean_token_accuracy": 0.23554679602384568,
      "num_tokens": 15551603.0,
      "step": 8850
    },
    {
      "entropy": 5.061412382125854,
      "epoch": 6.939655172413794,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00016142502076200622,
      "loss": 4.3722,
      "mean_token_accuracy": 0.23767588585615157,
      "num_tokens": 15560412.0,
      "step": 8855
    },
    {
      "entropy": 5.015926790237427,
      "epoch": 6.943573667711599,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016116546486896183,
      "loss": 4.3574,
      "mean_token_accuracy": 0.2327837884426117,
      "num_tokens": 15569781.0,
      "step": 8860
    },
    {
      "entropy": 5.102956914901734,
      "epoch": 6.947492163009405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016090611241668904,
      "loss": 4.3499,
      "mean_token_accuracy": 0.24055170118808747,
      "num_tokens": 15577640.0,
      "step": 8865
    },
    {
      "entropy": 5.147253322601318,
      "epoch": 6.95141065830721,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001606469638686928,
      "loss": 4.4637,
      "mean_token_accuracy": 0.231366865336895,
      "num_tokens": 15586950.0,
      "step": 8870
    },
    {
      "entropy": 5.102028512954712,
      "epoch": 6.955329153605016,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001603880196881137,
      "loss": 4.4322,
      "mean_token_accuracy": 0.2317335397005081,
      "num_tokens": 15595554.0,
      "step": 8875
    },
    {
      "entropy": 5.101431512832642,
      "epoch": 6.959247648902822,
      "grad_norm": 1.515625,
      "learning_rate": 0.0001601292803377269,
      "loss": 4.3786,
      "mean_token_accuracy": 0.2349329635500908,
      "num_tokens": 15603998.0,
      "step": 8880
    },
    {
      "entropy": 5.155918407440185,
      "epoch": 6.963166144200627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001598707462799414,
      "loss": 4.4466,
      "mean_token_accuracy": 0.22963792979717254,
      "num_tokens": 15612977.0,
      "step": 8885
    },
    {
      "entropy": 5.119667196273804,
      "epoch": 6.967084639498433,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015961241797679977,
      "loss": 4.4286,
      "mean_token_accuracy": 0.23766496926546096,
      "num_tokens": 15622914.0,
      "step": 8890
    },
    {
      "entropy": 5.099386501312256,
      "epoch": 6.971003134796238,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015935429588997637,
      "loss": 4.3977,
      "mean_token_accuracy": 0.2358198955655098,
      "num_tokens": 15631929.0,
      "step": 8895
    },
    {
      "entropy": 5.033832311630249,
      "epoch": 6.974921630094044,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015909638048077724,
      "loss": 4.2792,
      "mean_token_accuracy": 0.24322413355112077,
      "num_tokens": 15640400.0,
      "step": 8900
    },
    {
      "entropy": 5.189240169525147,
      "epoch": 6.97884012539185,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015883867221013915,
      "loss": 4.4654,
      "mean_token_accuracy": 0.23470586389303208,
      "num_tokens": 15648934.0,
      "step": 8905
    },
    {
      "entropy": 5.132853746414185,
      "epoch": 6.982758620689655,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001585811715386285,
      "loss": 4.4066,
      "mean_token_accuracy": 0.2261979326605797,
      "num_tokens": 15658461.0,
      "step": 8910
    },
    {
      "entropy": 5.029687213897705,
      "epoch": 6.986677115987461,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015832387892644072,
      "loss": 4.33,
      "mean_token_accuracy": 0.24004824459552765,
      "num_tokens": 15667182.0,
      "step": 8915
    },
    {
      "entropy": 5.094253253936768,
      "epoch": 6.990595611285267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001580667948333994,
      "loss": 4.4358,
      "mean_token_accuracy": 0.23152667433023452,
      "num_tokens": 15675951.0,
      "step": 8920
    },
    {
      "entropy": 5.122194433212281,
      "epoch": 6.994514106583072,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001578099197189555,
      "loss": 4.3894,
      "mean_token_accuracy": 0.23407917320728303,
      "num_tokens": 15684171.0,
      "step": 8925
    },
    {
      "entropy": 5.053464126586914,
      "epoch": 6.998432601880878,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015755325404218662,
      "loss": 4.392,
      "mean_token_accuracy": 0.2403782680630684,
      "num_tokens": 15692986.0,
      "step": 8930
    },
    {
      "entropy": 5.122729730606079,
      "epoch": 7.002351097178684,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015729679826179577,
      "loss": 4.3585,
      "mean_token_accuracy": 0.24322807937860488,
      "num_tokens": 15702591.0,
      "step": 8935
    },
    {
      "entropy": 5.092265987396241,
      "epoch": 7.006269592476489,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00015704055283611123,
      "loss": 4.2627,
      "mean_token_accuracy": 0.24351628124713898,
      "num_tokens": 15711186.0,
      "step": 8940
    },
    {
      "entropy": 5.0817255020141605,
      "epoch": 7.010188087774295,
      "grad_norm": 1.359375,
      "learning_rate": 0.00015678451822308495,
      "loss": 4.2918,
      "mean_token_accuracy": 0.25220857709646227,
      "num_tokens": 15719831.0,
      "step": 8945
    },
    {
      "entropy": 5.164159536361694,
      "epoch": 7.0141065830721,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001565286948802923,
      "loss": 4.3403,
      "mean_token_accuracy": 0.24039508998394013,
      "num_tokens": 15727926.0,
      "step": 8950
    },
    {
      "entropy": 5.066989755630493,
      "epoch": 7.018025078369906,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00015627308326493118,
      "loss": 4.1988,
      "mean_token_accuracy": 0.25775658488273623,
      "num_tokens": 15735934.0,
      "step": 8955
    },
    {
      "entropy": 5.029991912841797,
      "epoch": 7.021943573667712,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001560176838338209,
      "loss": 4.2825,
      "mean_token_accuracy": 0.2447723150253296,
      "num_tokens": 15745014.0,
      "step": 8960
    },
    {
      "entropy": 5.012461757659912,
      "epoch": 7.025862068965517,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015576249704340173,
      "loss": 4.1975,
      "mean_token_accuracy": 0.25160479098558425,
      "num_tokens": 15753360.0,
      "step": 8965
    },
    {
      "entropy": 5.073371171951294,
      "epoch": 7.029780564263323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00015550752334973368,
      "loss": 4.2608,
      "mean_token_accuracy": 0.24949465095996856,
      "num_tokens": 15762395.0,
      "step": 8970
    },
    {
      "entropy": 5.073869228363037,
      "epoch": 7.033699059561129,
      "grad_norm": 1.328125,
      "learning_rate": 0.00015525276320849628,
      "loss": 4.1712,
      "mean_token_accuracy": 0.2608379527926445,
      "num_tokens": 15771292.0,
      "step": 8975
    },
    {
      "entropy": 5.131912279129028,
      "epoch": 7.037617554858934,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015499821707498702,
      "loss": 4.2642,
      "mean_token_accuracy": 0.2488010287284851,
      "num_tokens": 15779879.0,
      "step": 8980
    },
    {
      "entropy": 5.062150239944458,
      "epoch": 7.04153605015674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00015474388540412138,
      "loss": 4.2639,
      "mean_token_accuracy": 0.25646011978387834,
      "num_tokens": 15788960.0,
      "step": 8985
    },
    {
      "entropy": 5.087431383132935,
      "epoch": 7.045454545454546,
      "grad_norm": 1.28125,
      "learning_rate": 0.000154489768650431,
      "loss": 4.2887,
      "mean_token_accuracy": 0.24196789562702178,
      "num_tokens": 15798260.0,
      "step": 8990
    },
    {
      "entropy": 5.069931507110596,
      "epoch": 7.049373040752351,
      "grad_norm": 1.421875,
      "learning_rate": 0.00015423586726806394,
      "loss": 4.2823,
      "mean_token_accuracy": 0.24971585869789123,
      "num_tokens": 15807480.0,
      "step": 8995
    },
    {
      "entropy": 5.142841863632202,
      "epoch": 7.053291536050157,
      "grad_norm": 1.265625,
      "learning_rate": 0.00015398218171078305,
      "loss": 4.3741,
      "mean_token_accuracy": 0.24170057773590087,
      "num_tokens": 15816330.0,
      "step": 9000
    },
    {
      "epoch": 7.053291536050157,
      "eval_entropy": 5.032915454502254,
      "eval_loss": 5.628068447113037,
      "eval_mean_token_accuracy": 0.17528377539774245,
      "eval_num_tokens": 15816330.0,
      "eval_runtime": 3.0398,
      "eval_samples_per_second": 1356.003,
      "eval_steps_per_second": 169.747,
      "step": 9000
    },
    {
      "entropy": 5.09982533454895,
      "epoch": 7.057210031347962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00015372871243196545,
      "loss": 4.2707,
      "mean_token_accuracy": 0.24948549270629883,
      "num_tokens": 15824600.0,
      "step": 9005
    },
    {
      "entropy": 5.106684017181396,
      "epoch": 7.061128526645768,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001534754598846021,
      "loss": 4.2847,
      "mean_token_accuracy": 0.24972220361232758,
      "num_tokens": 15834170.0,
      "step": 9010
    },
    {
      "entropy": 5.140288782119751,
      "epoch": 7.065047021943574,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00015322242452129616,
      "loss": 4.3062,
      "mean_token_accuracy": 0.24342776089906693,
      "num_tokens": 15843061.0,
      "step": 9015
    },
    {
      "entropy": 5.1076727390289305,
      "epoch": 7.068965517241379,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00015296960679426285,
      "loss": 4.2695,
      "mean_token_accuracy": 0.24518343061208725,
      "num_tokens": 15851788.0,
      "step": 9020
    },
    {
      "entropy": 5.107475900650025,
      "epoch": 7.072884012539185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015271700715532854,
      "loss": 4.3084,
      "mean_token_accuracy": 0.2406119018793106,
      "num_tokens": 15860216.0,
      "step": 9025
    },
    {
      "entropy": 5.071004009246826,
      "epoch": 7.076802507836991,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015246462605592972,
      "loss": 4.1572,
      "mean_token_accuracy": 0.25923371464014056,
      "num_tokens": 15869498.0,
      "step": 9030
    },
    {
      "entropy": 5.109587144851685,
      "epoch": 7.080721003134796,
      "grad_norm": 1.34375,
      "learning_rate": 0.00015221246394711224,
      "loss": 4.2357,
      "mean_token_accuracy": 0.2539739340543747,
      "num_tokens": 15877981.0,
      "step": 9035
    },
    {
      "entropy": 5.108943939208984,
      "epoch": 7.084639498432602,
      "grad_norm": 1.40625,
      "learning_rate": 0.00015196052127953085,
      "loss": 4.2626,
      "mean_token_accuracy": 0.24461860358715057,
      "num_tokens": 15886454.0,
      "step": 9040
    },
    {
      "entropy": 5.067775821685791,
      "epoch": 7.088557993730408,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001517087985034477,
      "loss": 4.216,
      "mean_token_accuracy": 0.259482216835022,
      "num_tokens": 15895025.0,
      "step": 9045
    },
    {
      "entropy": 5.102508401870727,
      "epoch": 7.092476489028213,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001514572960687325,
      "loss": 4.3415,
      "mean_token_accuracy": 0.2411337375640869,
      "num_tokens": 15903611.0,
      "step": 9050
    },
    {
      "entropy": 5.12601752281189,
      "epoch": 7.096394984326019,
      "grad_norm": 1.296875,
      "learning_rate": 0.00015120601442486061,
      "loss": 4.3032,
      "mean_token_accuracy": 0.24273794889450073,
      "num_tokens": 15912078.0,
      "step": 9055
    },
    {
      "entropy": 5.020508480072022,
      "epoch": 7.100313479623824,
      "grad_norm": 1.546875,
      "learning_rate": 0.00015095495402091329,
      "loss": 4.2264,
      "mean_token_accuracy": 0.25681900084018705,
      "num_tokens": 15920914.0,
      "step": 9060
    },
    {
      "entropy": 5.106277179718018,
      "epoch": 7.10423197492163,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001507041153055761,
      "loss": 4.2171,
      "mean_token_accuracy": 0.24838048666715623,
      "num_tokens": 15929199.0,
      "step": 9065
    },
    {
      "entropy": 5.183600950241089,
      "epoch": 7.108150470219436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00015045349872713838,
      "loss": 4.361,
      "mean_token_accuracy": 0.24611321985721588,
      "num_tokens": 15937817.0,
      "step": 9070
    },
    {
      "entropy": 5.077832508087158,
      "epoch": 7.112068965517241,
      "grad_norm": 1.390625,
      "learning_rate": 0.00015020310473349275,
      "loss": 4.2623,
      "mean_token_accuracy": 0.2565354689955711,
      "num_tokens": 15945997.0,
      "step": 9075
    },
    {
      "entropy": 5.051411581039429,
      "epoch": 7.115987460815047,
      "grad_norm": 1.375,
      "learning_rate": 0.00014995293377213388,
      "loss": 4.2309,
      "mean_token_accuracy": 0.24460477083921434,
      "num_tokens": 15954215.0,
      "step": 9080
    },
    {
      "entropy": 5.0879675388336185,
      "epoch": 7.119905956112853,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001497029862901578,
      "loss": 4.2723,
      "mean_token_accuracy": 0.25756254494190217,
      "num_tokens": 15962222.0,
      "step": 9085
    },
    {
      "entropy": 5.1210075378417965,
      "epoch": 7.123824451410658,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014945326273426102,
      "loss": 4.211,
      "mean_token_accuracy": 0.24838227331638335,
      "num_tokens": 15970712.0,
      "step": 9090
    },
    {
      "entropy": 5.118081140518188,
      "epoch": 7.127742946708464,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014920376355074034,
      "loss": 4.3006,
      "mean_token_accuracy": 0.24373193979263305,
      "num_tokens": 15979049.0,
      "step": 9095
    },
    {
      "entropy": 5.081648111343384,
      "epoch": 7.131661442006269,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014895448918549097,
      "loss": 4.2734,
      "mean_token_accuracy": 0.24722461998462678,
      "num_tokens": 15988221.0,
      "step": 9100
    },
    {
      "entropy": 5.058177804946899,
      "epoch": 7.135579937304075,
      "grad_norm": 1.34375,
      "learning_rate": 0.00014870544008400681,
      "loss": 4.2363,
      "mean_token_accuracy": 0.25172221958637236,
      "num_tokens": 15996579.0,
      "step": 9105
    },
    {
      "entropy": 5.076094150543213,
      "epoch": 7.139498432601881,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001484566166913789,
      "loss": 4.2562,
      "mean_token_accuracy": 0.250088669359684,
      "num_tokens": 16005489.0,
      "step": 9110
    },
    {
      "entropy": 5.029453325271606,
      "epoch": 7.143416927899686,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014820801945229504,
      "loss": 4.2174,
      "mean_token_accuracy": 0.25420559048652647,
      "num_tokens": 16014167.0,
      "step": 9115
    },
    {
      "entropy": 5.0481525421142575,
      "epoch": 7.147335423197492,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001479596488110387,
      "loss": 4.1624,
      "mean_token_accuracy": 0.2595665231347084,
      "num_tokens": 16022184.0,
      "step": 9120
    },
    {
      "entropy": 5.08054895401001,
      "epoch": 7.151253918495298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014771150521148863,
      "loss": 4.3034,
      "mean_token_accuracy": 0.24574496895074843,
      "num_tokens": 16031704.0,
      "step": 9125
    },
    {
      "entropy": 5.117170095443726,
      "epoch": 7.155172413793103,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001474635890971176,
      "loss": 4.3053,
      "mean_token_accuracy": 0.25463479161262514,
      "num_tokens": 16040913.0,
      "step": 9130
    },
    {
      "entropy": 5.083073949813842,
      "epoch": 7.159090909090909,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001472159009109919,
      "loss": 4.2281,
      "mean_token_accuracy": 0.2510302931070328,
      "num_tokens": 16048943.0,
      "step": 9135
    },
    {
      "entropy": 5.068734121322632,
      "epoch": 7.163009404388715,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00014696844109577043,
      "loss": 4.2059,
      "mean_token_accuracy": 0.249106328189373,
      "num_tokens": 16057999.0,
      "step": 9140
    },
    {
      "entropy": 5.088970041275024,
      "epoch": 7.16692789968652,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00014672121009370415,
      "loss": 4.3397,
      "mean_token_accuracy": 0.2389057010412216,
      "num_tokens": 16066920.0,
      "step": 9145
    },
    {
      "entropy": 5.044091987609863,
      "epoch": 7.170846394984326,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00014647420834663482,
      "loss": 4.307,
      "mean_token_accuracy": 0.23958202749490737,
      "num_tokens": 16075128.0,
      "step": 9150
    },
    {
      "entropy": 5.065071249008179,
      "epoch": 7.174764890282131,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001462274362959946,
      "loss": 4.2893,
      "mean_token_accuracy": 0.24642620980739594,
      "num_tokens": 16083949.0,
      "step": 9155
    },
    {
      "entropy": 5.11691575050354,
      "epoch": 7.178683385579937,
      "grad_norm": 1.5,
      "learning_rate": 0.0001459808943828053,
      "loss": 4.2428,
      "mean_token_accuracy": 0.25050083100795745,
      "num_tokens": 16092891.0,
      "step": 9160
    },
    {
      "entropy": 5.156085729598999,
      "epoch": 7.182601880877743,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00014573458304767707,
      "loss": 4.3454,
      "mean_token_accuracy": 0.2319442629814148,
      "num_tokens": 16101602.0,
      "step": 9165
    },
    {
      "entropy": 5.091942214965821,
      "epoch": 7.186520376175548,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00014548850273080839,
      "loss": 4.3045,
      "mean_token_accuracy": 0.24074742048978806,
      "num_tokens": 16110430.0,
      "step": 9170
    },
    {
      "entropy": 5.00772705078125,
      "epoch": 7.190438871473354,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00014524265387198465,
      "loss": 4.243,
      "mean_token_accuracy": 0.2492967888712883,
      "num_tokens": 16119444.0,
      "step": 9175
    },
    {
      "entropy": 5.065815687179565,
      "epoch": 7.19435736677116,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014499703691057766,
      "loss": 4.2339,
      "mean_token_accuracy": 0.2516627207398415,
      "num_tokens": 16128721.0,
      "step": 9180
    },
    {
      "entropy": 5.16632342338562,
      "epoch": 7.198275862068965,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001447516522855447,
      "loss": 4.3226,
      "mean_token_accuracy": 0.2374977484345436,
      "num_tokens": 16137508.0,
      "step": 9185
    },
    {
      "entropy": 5.153549289703369,
      "epoch": 7.202194357366771,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00014450650043542775,
      "loss": 4.2515,
      "mean_token_accuracy": 0.2508171334862709,
      "num_tokens": 16146239.0,
      "step": 9190
    },
    {
      "entropy": 5.110477542877197,
      "epoch": 7.206112852664576,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014426158179835317,
      "loss": 4.2425,
      "mean_token_accuracy": 0.24534540325403215,
      "num_tokens": 16154769.0,
      "step": 9195
    },
    {
      "entropy": 4.9980597496032715,
      "epoch": 7.210031347962382,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001440168968120301,
      "loss": 4.1746,
      "mean_token_accuracy": 0.2632712244987488,
      "num_tokens": 16163307.0,
      "step": 9200
    },
    {
      "entropy": 5.068648386001587,
      "epoch": 7.213949843260188,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00014377244591375033,
      "loss": 4.2545,
      "mean_token_accuracy": 0.24707067161798477,
      "num_tokens": 16171533.0,
      "step": 9205
    },
    {
      "entropy": 5.018707275390625,
      "epoch": 7.217868338557993,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001435282295403873,
      "loss": 4.2348,
      "mean_token_accuracy": 0.2570388659834862,
      "num_tokens": 16180163.0,
      "step": 9210
    },
    {
      "entropy": 5.038453769683838,
      "epoch": 7.221786833855799,
      "grad_norm": 1.40625,
      "learning_rate": 0.00014328424812839525,
      "loss": 4.2056,
      "mean_token_accuracy": 0.2538474455475807,
      "num_tokens": 16188060.0,
      "step": 9215
    },
    {
      "entropy": 5.065629673004151,
      "epoch": 7.225705329153605,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014304050211380836,
      "loss": 4.3066,
      "mean_token_accuracy": 0.24383010864257812,
      "num_tokens": 16197597.0,
      "step": 9220
    },
    {
      "entropy": 5.1221331596374515,
      "epoch": 7.22962382445141,
      "grad_norm": 1.53125,
      "learning_rate": 0.00014279699193224047,
      "loss": 4.3176,
      "mean_token_accuracy": 0.23303416520357131,
      "num_tokens": 16206288.0,
      "step": 9225
    },
    {
      "entropy": 5.087893915176392,
      "epoch": 7.233542319749216,
      "grad_norm": 1.328125,
      "learning_rate": 0.00014255371801888379,
      "loss": 4.2779,
      "mean_token_accuracy": 0.24432917088270187,
      "num_tokens": 16215321.0,
      "step": 9230
    },
    {
      "entropy": 5.0756010055542,
      "epoch": 7.237460815047022,
      "grad_norm": 1.46875,
      "learning_rate": 0.00014231068080850813,
      "loss": 4.1859,
      "mean_token_accuracy": 0.26211777329444885,
      "num_tokens": 16224445.0,
      "step": 9235
    },
    {
      "entropy": 5.075403356552124,
      "epoch": 7.241379310344827,
      "grad_norm": 1.375,
      "learning_rate": 0.00014206788073546035,
      "loss": 4.2566,
      "mean_token_accuracy": 0.254065103828907,
      "num_tokens": 16233189.0,
      "step": 9240
    },
    {
      "entropy": 5.03113956451416,
      "epoch": 7.245297805642633,
      "grad_norm": 1.421875,
      "learning_rate": 0.00014182531823366372,
      "loss": 4.2265,
      "mean_token_accuracy": 0.2560042947530746,
      "num_tokens": 16241664.0,
      "step": 9245
    },
    {
      "entropy": 5.0588172435760494,
      "epoch": 7.2492163009404385,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014158299373661666,
      "loss": 4.3009,
      "mean_token_accuracy": 0.24887900799512863,
      "num_tokens": 16251293.0,
      "step": 9250
    },
    {
      "entropy": 5.100256395339966,
      "epoch": 7.253134796238244,
      "grad_norm": 1.375,
      "learning_rate": 0.00014134090767739223,
      "loss": 4.2518,
      "mean_token_accuracy": 0.25285371094942094,
      "num_tokens": 16260134.0,
      "step": 9255
    },
    {
      "entropy": 5.113372039794922,
      "epoch": 7.25705329153605,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00014109906048863772,
      "loss": 4.2559,
      "mean_token_accuracy": 0.25506080985069274,
      "num_tokens": 16269027.0,
      "step": 9260
    },
    {
      "entropy": 5.099272155761719,
      "epoch": 7.2609717868338555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001408574526025731,
      "loss": 4.2803,
      "mean_token_accuracy": 0.24962224066257477,
      "num_tokens": 16277692.0,
      "step": 9265
    },
    {
      "entropy": 5.072432136535644,
      "epoch": 7.264890282131661,
      "grad_norm": 1.359375,
      "learning_rate": 0.00014061608445099083,
      "loss": 4.291,
      "mean_token_accuracy": 0.24009734094142915,
      "num_tokens": 16287059.0,
      "step": 9270
    },
    {
      "entropy": 5.044191455841064,
      "epoch": 7.268808777429467,
      "grad_norm": 1.5625,
      "learning_rate": 0.00014037495646525495,
      "loss": 4.2572,
      "mean_token_accuracy": 0.2508635461330414,
      "num_tokens": 16294880.0,
      "step": 9275
    },
    {
      "entropy": 5.073480033874512,
      "epoch": 7.2727272727272725,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001401340690763003,
      "loss": 4.2042,
      "mean_token_accuracy": 0.2542799040675163,
      "num_tokens": 16304101.0,
      "step": 9280
    },
    {
      "entropy": 5.10424017906189,
      "epoch": 7.2766457680250785,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013989342271463173,
      "loss": 4.3667,
      "mean_token_accuracy": 0.24016123116016388,
      "num_tokens": 16312887.0,
      "step": 9285
    },
    {
      "entropy": 5.043989753723144,
      "epoch": 7.280564263322884,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001396530178103232,
      "loss": 4.2523,
      "mean_token_accuracy": 0.24772798269987106,
      "num_tokens": 16322256.0,
      "step": 9290
    },
    {
      "entropy": 5.092056560516357,
      "epoch": 7.2844827586206895,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001394128547930174,
      "loss": 4.2846,
      "mean_token_accuracy": 0.24836019575595855,
      "num_tokens": 16331810.0,
      "step": 9295
    },
    {
      "entropy": 5.122595500946045,
      "epoch": 7.2884012539184955,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001391729340919245,
      "loss": 4.291,
      "mean_token_accuracy": 0.24694954752922058,
      "num_tokens": 16341385.0,
      "step": 9300
    },
    {
      "entropy": 5.075524854660034,
      "epoch": 7.2923197492163006,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013893325613582172,
      "loss": 4.2348,
      "mean_token_accuracy": 0.2535801962018013,
      "num_tokens": 16350231.0,
      "step": 9305
    },
    {
      "entropy": 5.0988812923431395,
      "epoch": 7.2962382445141065,
      "grad_norm": 1.375,
      "learning_rate": 0.0001386938213530523,
      "loss": 4.3255,
      "mean_token_accuracy": 0.24663783460855485,
      "num_tokens": 16359659.0,
      "step": 9310
    },
    {
      "entropy": 5.089129638671875,
      "epoch": 7.3001567398119125,
      "grad_norm": 1.328125,
      "learning_rate": 0.00013845463017152527,
      "loss": 4.2477,
      "mean_token_accuracy": 0.25432735979557036,
      "num_tokens": 16368606.0,
      "step": 9315
    },
    {
      "entropy": 5.133270502090454,
      "epoch": 7.304075235109718,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00013821568301871384,
      "loss": 4.3202,
      "mean_token_accuracy": 0.24321169555187225,
      "num_tokens": 16376428.0,
      "step": 9320
    },
    {
      "entropy": 5.058445310592651,
      "epoch": 7.3079937304075235,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013797698032165534,
      "loss": 4.2601,
      "mean_token_accuracy": 0.2511559143662453,
      "num_tokens": 16384983.0,
      "step": 9325
    },
    {
      "entropy": 5.093526411056518,
      "epoch": 7.3119122257053295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00013773852250695035,
      "loss": 4.2919,
      "mean_token_accuracy": 0.24474958032369615,
      "num_tokens": 16393747.0,
      "step": 9330
    },
    {
      "entropy": 5.0855528831481935,
      "epoch": 7.315830721003135,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013750031000076145,
      "loss": 4.2611,
      "mean_token_accuracy": 0.24635968059301377,
      "num_tokens": 16401899.0,
      "step": 9335
    },
    {
      "entropy": 5.089543008804322,
      "epoch": 7.3197492163009406,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001372623432288131,
      "loss": 4.2679,
      "mean_token_accuracy": 0.244497050344944,
      "num_tokens": 16410787.0,
      "step": 9340
    },
    {
      "entropy": 5.139873456954956,
      "epoch": 7.3236677115987465,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013702462261639054,
      "loss": 4.3196,
      "mean_token_accuracy": 0.24348561614751815,
      "num_tokens": 16420023.0,
      "step": 9345
    },
    {
      "entropy": 5.101294136047363,
      "epoch": 7.327586206896552,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000136787148588339,
      "loss": 4.2168,
      "mean_token_accuracy": 0.25356807708740237,
      "num_tokens": 16428394.0,
      "step": 9350
    },
    {
      "entropy": 5.091585683822632,
      "epoch": 7.331504702194358,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001365499215690632,
      "loss": 4.25,
      "mean_token_accuracy": 0.24908274561166763,
      "num_tokens": 16436429.0,
      "step": 9355
    },
    {
      "entropy": 4.975730609893799,
      "epoch": 7.335423197492163,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00013631294198252603,
      "loss": 4.2009,
      "mean_token_accuracy": 0.2556259110569954,
      "num_tokens": 16445140.0,
      "step": 9360
    },
    {
      "entropy": 5.0450225353240965,
      "epoch": 7.339341692789969,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00013607621025224864,
      "loss": 4.2394,
      "mean_token_accuracy": 0.24843940287828445,
      "num_tokens": 16453932.0,
      "step": 9365
    },
    {
      "entropy": 5.037712621688843,
      "epoch": 7.343260188087775,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000135839726801309,
      "loss": 4.2272,
      "mean_token_accuracy": 0.24661013782024382,
      "num_tokens": 16462539.0,
      "step": 9370
    },
    {
      "entropy": 5.054360294342041,
      "epoch": 7.34717868338558,
      "grad_norm": 1.34375,
      "learning_rate": 0.00013560349205234124,
      "loss": 4.3036,
      "mean_token_accuracy": 0.2436927855014801,
      "num_tokens": 16471979.0,
      "step": 9375
    },
    {
      "entropy": 5.117911291122437,
      "epoch": 7.351097178683386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013536750642753538,
      "loss": 4.3789,
      "mean_token_accuracy": 0.23454816937446593,
      "num_tokens": 16481266.0,
      "step": 9380
    },
    {
      "entropy": 5.102658271789551,
      "epoch": 7.355015673981192,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013513177034863588,
      "loss": 4.3187,
      "mean_token_accuracy": 0.24837499111890793,
      "num_tokens": 16490360.0,
      "step": 9385
    },
    {
      "entropy": 5.130635690689087,
      "epoch": 7.358934169278997,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001348962842369413,
      "loss": 4.3505,
      "mean_token_accuracy": 0.23551813513040543,
      "num_tokens": 16498043.0,
      "step": 9390
    },
    {
      "entropy": 5.069915151596069,
      "epoch": 7.362852664576803,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013466104851330346,
      "loss": 4.2955,
      "mean_token_accuracy": 0.24699692726135253,
      "num_tokens": 16506343.0,
      "step": 9395
    },
    {
      "entropy": 5.072348546981812,
      "epoch": 7.366771159874608,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001344260635981268,
      "loss": 4.2524,
      "mean_token_accuracy": 0.2523703664541245,
      "num_tokens": 16515308.0,
      "step": 9400
    },
    {
      "entropy": 5.098485612869263,
      "epoch": 7.370689655172414,
      "grad_norm": 1.296875,
      "learning_rate": 0.00013419132991136762,
      "loss": 4.2696,
      "mean_token_accuracy": 0.2432551607489586,
      "num_tokens": 16524379.0,
      "step": 9405
    },
    {
      "entropy": 5.105703592300415,
      "epoch": 7.37460815047022,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013395684787253278,
      "loss": 4.2792,
      "mean_token_accuracy": 0.25310360342264177,
      "num_tokens": 16532797.0,
      "step": 9410
    },
    {
      "entropy": 5.10154709815979,
      "epoch": 7.378526645768025,
      "grad_norm": 1.53125,
      "learning_rate": 0.00013372261790068,
      "loss": 4.2716,
      "mean_token_accuracy": 0.2446860373020172,
      "num_tokens": 16540984.0,
      "step": 9415
    },
    {
      "entropy": 5.065394115447998,
      "epoch": 7.382445141065831,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013348864041441604,
      "loss": 4.1715,
      "mean_token_accuracy": 0.25545849800109866,
      "num_tokens": 16549386.0,
      "step": 9420
    },
    {
      "entropy": 5.042044258117675,
      "epoch": 7.386363636363637,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013325491583189664,
      "loss": 4.2779,
      "mean_token_accuracy": 0.24895893037319183,
      "num_tokens": 16557673.0,
      "step": 9425
    },
    {
      "entropy": 5.081043672561646,
      "epoch": 7.390282131661442,
      "grad_norm": 1.375,
      "learning_rate": 0.00013302144457082553,
      "loss": 4.3167,
      "mean_token_accuracy": 0.2440224200487137,
      "num_tokens": 16566383.0,
      "step": 9430
    },
    {
      "entropy": 5.100075960159302,
      "epoch": 7.394200626959248,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013278822704845372,
      "loss": 4.2457,
      "mean_token_accuracy": 0.2566226437687874,
      "num_tokens": 16574819.0,
      "step": 9435
    },
    {
      "entropy": 5.083183622360229,
      "epoch": 7.398119122257054,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00013255526368157876,
      "loss": 4.2567,
      "mean_token_accuracy": 0.2490334615111351,
      "num_tokens": 16582571.0,
      "step": 9440
    },
    {
      "entropy": 5.125161457061767,
      "epoch": 7.402037617554859,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013232255488654387,
      "loss": 4.3253,
      "mean_token_accuracy": 0.2441723421216011,
      "num_tokens": 16592045.0,
      "step": 9445
    },
    {
      "entropy": 5.09113564491272,
      "epoch": 7.405956112852665,
      "grad_norm": 1.375,
      "learning_rate": 0.00013209010107923747,
      "loss": 4.2065,
      "mean_token_accuracy": 0.2556499809026718,
      "num_tokens": 16600864.0,
      "step": 9450
    },
    {
      "entropy": 5.135703372955322,
      "epoch": 7.40987460815047,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013185790267509218,
      "loss": 4.3793,
      "mean_token_accuracy": 0.24783104211091994,
      "num_tokens": 16610193.0,
      "step": 9455
    },
    {
      "entropy": 5.073501396179199,
      "epoch": 7.413793103448276,
      "grad_norm": 1.421875,
      "learning_rate": 0.00013162596008908423,
      "loss": 4.3205,
      "mean_token_accuracy": 0.24213551580905915,
      "num_tokens": 16618663.0,
      "step": 9460
    },
    {
      "entropy": 5.065561723709107,
      "epoch": 7.417711598746082,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00013139427373573255,
      "loss": 4.2524,
      "mean_token_accuracy": 0.2518284901976585,
      "num_tokens": 16626885.0,
      "step": 9465
    },
    {
      "entropy": 5.115686511993408,
      "epoch": 7.421630094043887,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013116284402909834,
      "loss": 4.3317,
      "mean_token_accuracy": 0.24362442642450333,
      "num_tokens": 16636592.0,
      "step": 9470
    },
    {
      "entropy": 5.122469758987426,
      "epoch": 7.425548589341693,
      "grad_norm": 1.40625,
      "learning_rate": 0.000130931671382784,
      "loss": 4.3405,
      "mean_token_accuracy": 0.24566463977098466,
      "num_tokens": 16645897.0,
      "step": 9475
    },
    {
      "entropy": 5.119991779327393,
      "epoch": 7.429467084639499,
      "grad_norm": 1.375,
      "learning_rate": 0.00013070075620993233,
      "loss": 4.2676,
      "mean_token_accuracy": 0.2477321669459343,
      "num_tokens": 16654034.0,
      "step": 9480
    },
    {
      "entropy": 5.058962011337281,
      "epoch": 7.433385579937304,
      "grad_norm": 1.390625,
      "learning_rate": 0.00013047009892322642,
      "loss": 4.2825,
      "mean_token_accuracy": 0.24879611134529114,
      "num_tokens": 16662510.0,
      "step": 9485
    },
    {
      "entropy": 5.087330150604248,
      "epoch": 7.43730407523511,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001302396999348882,
      "loss": 4.2415,
      "mean_token_accuracy": 0.2583288297057152,
      "num_tokens": 16671014.0,
      "step": 9490
    },
    {
      "entropy": 5.095565605163574,
      "epoch": 7.441222570532915,
      "grad_norm": 1.484375,
      "learning_rate": 0.00013000955965667788,
      "loss": 4.3128,
      "mean_token_accuracy": 0.24696000367403032,
      "num_tokens": 16680068.0,
      "step": 9495
    },
    {
      "entropy": 5.1004359245300295,
      "epoch": 7.445141065830721,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001297796784998936,
      "loss": 4.3405,
      "mean_token_accuracy": 0.24907656610012055,
      "num_tokens": 16688478.0,
      "step": 9500
    },
    {
      "epoch": 7.445141065830721,
      "eval_entropy": 4.987293692522271,
      "eval_loss": 5.623910427093506,
      "eval_mean_token_accuracy": 0.17512743024863014,
      "eval_num_tokens": 16688478.0,
      "eval_runtime": 2.8368,
      "eval_samples_per_second": 1453.067,
      "eval_steps_per_second": 181.898,
      "step": 9500
    },
    {
      "entropy": 5.108563947677612,
      "epoch": 7.449059561128527,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012955005687537025,
      "loss": 4.3853,
      "mean_token_accuracy": 0.24170157462358474,
      "num_tokens": 16698913.0,
      "step": 9505
    },
    {
      "entropy": 5.035240983963012,
      "epoch": 7.452978056426332,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012932069519347873,
      "loss": 4.2321,
      "mean_token_accuracy": 0.25370187163352964,
      "num_tokens": 16708306.0,
      "step": 9510
    },
    {
      "entropy": 5.10676064491272,
      "epoch": 7.456896551724138,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012909159386412578,
      "loss": 4.2945,
      "mean_token_accuracy": 0.2524004325270653,
      "num_tokens": 16716345.0,
      "step": 9515
    },
    {
      "entropy": 5.066718006134034,
      "epoch": 7.460815047021944,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012886275329675246,
      "loss": 4.2672,
      "mean_token_accuracy": 0.25064661651849746,
      "num_tokens": 16725405.0,
      "step": 9520
    },
    {
      "entropy": 5.053071641921997,
      "epoch": 7.464733542319749,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001286341739003341,
      "loss": 4.3066,
      "mean_token_accuracy": 0.24711575210094452,
      "num_tokens": 16734771.0,
      "step": 9525
    },
    {
      "entropy": 5.050468778610229,
      "epoch": 7.468652037617555,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00012840585608337907,
      "loss": 4.3115,
      "mean_token_accuracy": 0.24018340557813644,
      "num_tokens": 16742403.0,
      "step": 9530
    },
    {
      "entropy": 5.052300119400025,
      "epoch": 7.472570532915361,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012817780025392833,
      "loss": 4.2656,
      "mean_token_accuracy": 0.2513988420367241,
      "num_tokens": 16751450.0,
      "step": 9535
    },
    {
      "entropy": 5.051341581344604,
      "epoch": 7.476489028213166,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012795000681955468,
      "loss": 4.2525,
      "mean_token_accuracy": 0.2461089387536049,
      "num_tokens": 16760216.0,
      "step": 9540
    },
    {
      "entropy": 5.054883193969727,
      "epoch": 7.480407523510972,
      "grad_norm": 1.46875,
      "learning_rate": 0.00012772247618736182,
      "loss": 4.3319,
      "mean_token_accuracy": 0.24416444450616837,
      "num_tokens": 16768648.0,
      "step": 9545
    },
    {
      "entropy": 5.018112468719482,
      "epoch": 7.484326018808777,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012749520876398395,
      "loss": 4.2053,
      "mean_token_accuracy": 0.2566802754998207,
      "num_tokens": 16776587.0,
      "step": 9550
    },
    {
      "entropy": 5.060021543502808,
      "epoch": 7.488244514106583,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012726820495558483,
      "loss": 4.267,
      "mean_token_accuracy": 0.25382149368524554,
      "num_tokens": 16785937.0,
      "step": 9555
    },
    {
      "entropy": 5.101051568984985,
      "epoch": 7.492163009404389,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012704146516785705,
      "loss": 4.2279,
      "mean_token_accuracy": 0.25448717921972275,
      "num_tokens": 16794359.0,
      "step": 9560
    },
    {
      "entropy": 5.137592506408692,
      "epoch": 7.496081504702194,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001268149898060213,
      "loss": 4.3119,
      "mean_token_accuracy": 0.24513939917087554,
      "num_tokens": 16803685.0,
      "step": 9565
    },
    {
      "entropy": 5.019044208526611,
      "epoch": 7.5,
      "grad_norm": 1.484375,
      "learning_rate": 0.00012658877927482586,
      "loss": 4.2378,
      "mean_token_accuracy": 0.25344776213169096,
      "num_tokens": 16812267.0,
      "step": 9570
    },
    {
      "entropy": 5.099522399902344,
      "epoch": 7.503918495297806,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001263628339785456,
      "loss": 4.3332,
      "mean_token_accuracy": 0.2435259625315666,
      "num_tokens": 16820334.0,
      "step": 9575
    },
    {
      "entropy": 5.102103567123413,
      "epoch": 7.507836990595611,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012613715432098138,
      "loss": 4.2904,
      "mean_token_accuracy": 0.2523706257343292,
      "num_tokens": 16828908.0,
      "step": 9580
    },
    {
      "entropy": 5.125962257385254,
      "epoch": 7.511755485893417,
      "grad_norm": 1.453125,
      "learning_rate": 0.00012591174070545932,
      "loss": 4.3132,
      "mean_token_accuracy": 0.2429458037018776,
      "num_tokens": 16837772.0,
      "step": 9585
    },
    {
      "entropy": 5.1114874362945555,
      "epoch": 7.515673981191222,
      "grad_norm": 1.4375,
      "learning_rate": 0.00012568659353483024,
      "loss": 4.316,
      "mean_token_accuracy": 0.2424798622727394,
      "num_tokens": 16846714.0,
      "step": 9590
    },
    {
      "entropy": 5.071550416946411,
      "epoch": 7.519592476489028,
      "grad_norm": 1.390625,
      "learning_rate": 0.00012546171321146848,
      "loss": 4.2559,
      "mean_token_accuracy": 0.25270114541053773,
      "num_tokens": 16854745.0,
      "step": 9595
    },
    {
      "entropy": 5.096583366394043,
      "epoch": 7.523510971786834,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012523710013727173,
      "loss": 4.2777,
      "mean_token_accuracy": 0.2557577133178711,
      "num_tokens": 16864181.0,
      "step": 9600
    },
    {
      "entropy": 5.099077558517456,
      "epoch": 7.527429467084639,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00012501275471365998,
      "loss": 4.3222,
      "mean_token_accuracy": 0.2465886428952217,
      "num_tokens": 16873551.0,
      "step": 9605
    },
    {
      "entropy": 5.1373780250549315,
      "epoch": 7.531347962382445,
      "grad_norm": 1.375,
      "learning_rate": 0.00012478867734157486,
      "loss": 4.3711,
      "mean_token_accuracy": 0.2463907018303871,
      "num_tokens": 16883108.0,
      "step": 9610
    },
    {
      "entropy": 5.040398979187012,
      "epoch": 7.535266457680251,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00012456486842147885,
      "loss": 4.2629,
      "mean_token_accuracy": 0.2586983233690262,
      "num_tokens": 16891772.0,
      "step": 9615
    },
    {
      "entropy": 5.067703819274902,
      "epoch": 7.539184952978056,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00012434132835335496,
      "loss": 4.2659,
      "mean_token_accuracy": 0.2456803634762764,
      "num_tokens": 16900519.0,
      "step": 9620
    },
    {
      "entropy": 5.091511011123657,
      "epoch": 7.543103448275862,
      "grad_norm": 1.359375,
      "learning_rate": 0.00012411805753670543,
      "loss": 4.3094,
      "mean_token_accuracy": 0.2528110682964325,
      "num_tokens": 16909228.0,
      "step": 9625
    },
    {
      "entropy": 5.0807935237884525,
      "epoch": 7.547021943573668,
      "grad_norm": 1.453125,
      "learning_rate": 0.0001238950563705514,
      "loss": 4.3166,
      "mean_token_accuracy": 0.24798122495412828,
      "num_tokens": 16918100.0,
      "step": 9630
    },
    {
      "entropy": 5.107285642623902,
      "epoch": 7.550940438871473,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001236723252534322,
      "loss": 4.279,
      "mean_token_accuracy": 0.2478616029024124,
      "num_tokens": 16926911.0,
      "step": 9635
    },
    {
      "entropy": 5.1000786304473875,
      "epoch": 7.554858934169279,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00012344986458340433,
      "loss": 4.3125,
      "mean_token_accuracy": 0.24336836636066436,
      "num_tokens": 16935638.0,
      "step": 9640
    },
    {
      "entropy": 5.051099061965942,
      "epoch": 7.558777429467085,
      "grad_norm": 1.34375,
      "learning_rate": 0.00012322767475804122,
      "loss": 4.1689,
      "mean_token_accuracy": 0.264642222225666,
      "num_tokens": 16944351.0,
      "step": 9645
    },
    {
      "entropy": 5.122267389297486,
      "epoch": 7.56269592476489,
      "grad_norm": 1.515625,
      "learning_rate": 0.00012300575617443194,
      "loss": 4.2769,
      "mean_token_accuracy": 0.24902917444705963,
      "num_tokens": 16952486.0,
      "step": 9650
    },
    {
      "entropy": 5.130142688751221,
      "epoch": 7.566614420062696,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001227841092291812,
      "loss": 4.3355,
      "mean_token_accuracy": 0.24250130504369735,
      "num_tokens": 16960877.0,
      "step": 9655
    },
    {
      "entropy": 5.070782613754273,
      "epoch": 7.570532915360501,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012256273431840792,
      "loss": 4.3114,
      "mean_token_accuracy": 0.2480763763189316,
      "num_tokens": 16970868.0,
      "step": 9660
    },
    {
      "entropy": 5.0874535083770756,
      "epoch": 7.574451410658307,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00012234163183774482,
      "loss": 4.3145,
      "mean_token_accuracy": 0.24221487045288087,
      "num_tokens": 16980339.0,
      "step": 9665
    },
    {
      "entropy": 5.066333961486817,
      "epoch": 7.578369905956113,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012212080218233814,
      "loss": 4.2307,
      "mean_token_accuracy": 0.25241571217775344,
      "num_tokens": 16988786.0,
      "step": 9670
    },
    {
      "entropy": 5.08683500289917,
      "epoch": 7.582288401253918,
      "grad_norm": 1.328125,
      "learning_rate": 0.00012190024574684614,
      "loss": 4.2869,
      "mean_token_accuracy": 0.24819004088640212,
      "num_tokens": 16998454.0,
      "step": 9675
    },
    {
      "entropy": 5.098667430877685,
      "epoch": 7.586206896551724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012167996292543889,
      "loss": 4.2658,
      "mean_token_accuracy": 0.2500756174325943,
      "num_tokens": 17006951.0,
      "step": 9680
    },
    {
      "entropy": 5.146718502044678,
      "epoch": 7.59012539184953,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001214599541117976,
      "loss": 4.2809,
      "mean_token_accuracy": 0.2505610540509224,
      "num_tokens": 17015991.0,
      "step": 9685
    },
    {
      "entropy": 5.102245616912842,
      "epoch": 7.594043887147335,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00012124021969911366,
      "loss": 4.3137,
      "mean_token_accuracy": 0.24318447411060334,
      "num_tokens": 17024493.0,
      "step": 9690
    },
    {
      "entropy": 5.106430101394653,
      "epoch": 7.597962382445141,
      "grad_norm": 1.265625,
      "learning_rate": 0.00012102076008008804,
      "loss": 4.2857,
      "mean_token_accuracy": 0.25179476886987684,
      "num_tokens": 17033402.0,
      "step": 9695
    },
    {
      "entropy": 5.034291839599609,
      "epoch": 7.601880877742946,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00012080157564693073,
      "loss": 4.2238,
      "mean_token_accuracy": 0.2504350423812866,
      "num_tokens": 17042417.0,
      "step": 9700
    },
    {
      "entropy": 5.125713348388672,
      "epoch": 7.605799373040752,
      "grad_norm": 1.421875,
      "learning_rate": 0.00012058266679135987,
      "loss": 4.3999,
      "mean_token_accuracy": 0.23663944751024246,
      "num_tokens": 17050395.0,
      "step": 9705
    },
    {
      "entropy": 5.126303243637085,
      "epoch": 7.609717868338558,
      "grad_norm": 1.40625,
      "learning_rate": 0.00012036403390460107,
      "loss": 4.3486,
      "mean_token_accuracy": 0.2411719709634781,
      "num_tokens": 17060077.0,
      "step": 9710
    },
    {
      "entropy": 5.1405191898345945,
      "epoch": 7.613636363636363,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00012014567737738667,
      "loss": 4.2927,
      "mean_token_accuracy": 0.24038466960191726,
      "num_tokens": 17068099.0,
      "step": 9715
    },
    {
      "entropy": 5.082380151748657,
      "epoch": 7.617554858934169,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011992759759995529,
      "loss": 4.2362,
      "mean_token_accuracy": 0.24687227308750154,
      "num_tokens": 17076846.0,
      "step": 9720
    },
    {
      "entropy": 5.083771419525147,
      "epoch": 7.621473354231975,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011970979496205078,
      "loss": 4.2614,
      "mean_token_accuracy": 0.24829350858926774,
      "num_tokens": 17085960.0,
      "step": 9725
    },
    {
      "entropy": 5.0590283393859865,
      "epoch": 7.62539184952978,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00011949226985292183,
      "loss": 4.2383,
      "mean_token_accuracy": 0.25571011006832123,
      "num_tokens": 17095327.0,
      "step": 9730
    },
    {
      "entropy": 5.045916509628296,
      "epoch": 7.629310344827586,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011927502266132098,
      "loss": 4.263,
      "mean_token_accuracy": 0.24840926826000215,
      "num_tokens": 17103961.0,
      "step": 9735
    },
    {
      "entropy": 5.027958869934082,
      "epoch": 7.633228840125392,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011905805377550428,
      "loss": 4.2664,
      "mean_token_accuracy": 0.24665691554546357,
      "num_tokens": 17112427.0,
      "step": 9740
    },
    {
      "entropy": 5.054262971878051,
      "epoch": 7.637147335423197,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001188413635832303,
      "loss": 4.3204,
      "mean_token_accuracy": 0.24157961159944535,
      "num_tokens": 17120890.0,
      "step": 9745
    },
    {
      "entropy": 5.0367903232574465,
      "epoch": 7.641065830721003,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011862495247175947,
      "loss": 4.18,
      "mean_token_accuracy": 0.2579312786459923,
      "num_tokens": 17129881.0,
      "step": 9750
    },
    {
      "entropy": 5.147681427001953,
      "epoch": 7.644984326018808,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011840882082785357,
      "loss": 4.3292,
      "mean_token_accuracy": 0.2516689747571945,
      "num_tokens": 17138900.0,
      "step": 9755
    },
    {
      "entropy": 5.0864499568939205,
      "epoch": 7.648902821316614,
      "grad_norm": 1.421875,
      "learning_rate": 0.000118192969037775,
      "loss": 4.2892,
      "mean_token_accuracy": 0.2457812413573265,
      "num_tokens": 17147023.0,
      "step": 9760
    },
    {
      "entropy": 5.14745774269104,
      "epoch": 7.65282131661442,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011797739748728575,
      "loss": 4.3376,
      "mean_token_accuracy": 0.23754289746284485,
      "num_tokens": 17156926.0,
      "step": 9765
    },
    {
      "entropy": 5.112680149078369,
      "epoch": 7.656739811912225,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001177621065616473,
      "loss": 4.3177,
      "mean_token_accuracy": 0.24664452373981477,
      "num_tokens": 17165079.0,
      "step": 9770
    },
    {
      "entropy": 5.038353538513183,
      "epoch": 7.660658307210031,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011754709664561931,
      "loss": 4.2408,
      "mean_token_accuracy": 0.2532151311635971,
      "num_tokens": 17173509.0,
      "step": 9775
    },
    {
      "entropy": 5.121630907058716,
      "epoch": 7.664576802507837,
      "grad_norm": 1.296875,
      "learning_rate": 0.00011733236812345943,
      "loss": 4.2955,
      "mean_token_accuracy": 0.24775527268648148,
      "num_tokens": 17182399.0,
      "step": 9780
    },
    {
      "entropy": 5.063233947753906,
      "epoch": 7.668495297805642,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011711792137892221,
      "loss": 4.2726,
      "mean_token_accuracy": 0.24862852394580842,
      "num_tokens": 17190733.0,
      "step": 9785
    },
    {
      "entropy": 5.03784294128418,
      "epoch": 7.672413793103448,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011690375679525896,
      "loss": 4.2934,
      "mean_token_accuracy": 0.2425881266593933,
      "num_tokens": 17200078.0,
      "step": 9790
    },
    {
      "entropy": 5.092141199111938,
      "epoch": 7.6763322884012535,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011668987475521642,
      "loss": 4.3085,
      "mean_token_accuracy": 0.2519854992628098,
      "num_tokens": 17208430.0,
      "step": 9795
    },
    {
      "entropy": 5.107367467880249,
      "epoch": 7.6802507836990594,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00011647627564103636,
      "loss": 4.3612,
      "mean_token_accuracy": 0.24524620920419693,
      "num_tokens": 17217309.0,
      "step": 9800
    },
    {
      "entropy": 5.086805629730224,
      "epoch": 7.684169278996865,
      "grad_norm": 1.390625,
      "learning_rate": 0.00011626295983445525,
      "loss": 4.2191,
      "mean_token_accuracy": 0.2587997242808342,
      "num_tokens": 17225700.0,
      "step": 9805
    },
    {
      "entropy": 5.026204538345337,
      "epoch": 7.6880877742946705,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011604992771670292,
      "loss": 4.2641,
      "mean_token_accuracy": 0.24543796628713607,
      "num_tokens": 17234582.0,
      "step": 9810
    },
    {
      "entropy": 4.992440795898437,
      "epoch": 7.6920062695924765,
      "grad_norm": 1.421875,
      "learning_rate": 0.00011583717966850219,
      "loss": 4.2606,
      "mean_token_accuracy": 0.25298705101013186,
      "num_tokens": 17243207.0,
      "step": 9815
    },
    {
      "entropy": 5.025320768356323,
      "epoch": 7.695924764890282,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00011562471607006845,
      "loss": 4.2283,
      "mean_token_accuracy": 0.24664120078086854,
      "num_tokens": 17251658.0,
      "step": 9820
    },
    {
      "entropy": 5.1202935695648195,
      "epoch": 7.6998432601880875,
      "grad_norm": 1.4375,
      "learning_rate": 0.00011541253730110866,
      "loss": 4.3733,
      "mean_token_accuracy": 0.24365784972906113,
      "num_tokens": 17261175.0,
      "step": 9825
    },
    {
      "entropy": 5.049628877639771,
      "epoch": 7.7037617554858935,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001152006437408206,
      "loss": 4.1791,
      "mean_token_accuracy": 0.25960619151592257,
      "num_tokens": 17269302.0,
      "step": 9830
    },
    {
      "entropy": 5.0942957401275635,
      "epoch": 7.7076802507836994,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011498903576789233,
      "loss": 4.3416,
      "mean_token_accuracy": 0.24350888878107071,
      "num_tokens": 17277911.0,
      "step": 9835
    },
    {
      "entropy": 5.0810051441192625,
      "epoch": 7.7115987460815045,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00011477771376050177,
      "loss": 4.2719,
      "mean_token_accuracy": 0.2456583008170128,
      "num_tokens": 17287219.0,
      "step": 9840
    },
    {
      "entropy": 5.0520203590393065,
      "epoch": 7.7155172413793105,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00011456667809631556,
      "loss": 4.2283,
      "mean_token_accuracy": 0.24782970547676086,
      "num_tokens": 17295716.0,
      "step": 9845
    },
    {
      "entropy": 5.065628433227539,
      "epoch": 7.7194357366771165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011435592915248855,
      "loss": 4.312,
      "mean_token_accuracy": 0.24648393839597701,
      "num_tokens": 17305870.0,
      "step": 9850
    },
    {
      "entropy": 5.072757768630981,
      "epoch": 7.7233542319749215,
      "grad_norm": 1.40625,
      "learning_rate": 0.00011414546730566342,
      "loss": 4.2743,
      "mean_token_accuracy": 0.24977494180202484,
      "num_tokens": 17314130.0,
      "step": 9855
    },
    {
      "entropy": 5.076673364639282,
      "epoch": 7.7272727272727275,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011393529293196955,
      "loss": 4.1873,
      "mean_token_accuracy": 0.2621548384428024,
      "num_tokens": 17322320.0,
      "step": 9860
    },
    {
      "entropy": 5.084717178344727,
      "epoch": 7.731191222570533,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011372540640702263,
      "loss": 4.2918,
      "mean_token_accuracy": 0.24879802018404007,
      "num_tokens": 17331162.0,
      "step": 9865
    },
    {
      "entropy": 5.044370222091675,
      "epoch": 7.735109717868339,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00011351580810592382,
      "loss": 4.1991,
      "mean_token_accuracy": 0.25766499936580656,
      "num_tokens": 17339568.0,
      "step": 9870
    },
    {
      "entropy": 5.100082778930664,
      "epoch": 7.7390282131661445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00011330649840325938,
      "loss": 4.286,
      "mean_token_accuracy": 0.24904076159000396,
      "num_tokens": 17349354.0,
      "step": 9875
    },
    {
      "entropy": 5.0497935771942135,
      "epoch": 7.74294670846395,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001130974776730997,
      "loss": 4.3096,
      "mean_token_accuracy": 0.24057269096374512,
      "num_tokens": 17358668.0,
      "step": 9880
    },
    {
      "entropy": 5.046694040298462,
      "epoch": 7.746865203761756,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00011288874628899864,
      "loss": 4.2095,
      "mean_token_accuracy": 0.25295254588127136,
      "num_tokens": 17366792.0,
      "step": 9885
    },
    {
      "entropy": 5.0840808868408205,
      "epoch": 7.750783699059561,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011268030462399309,
      "loss": 4.2839,
      "mean_token_accuracy": 0.24961548745632173,
      "num_tokens": 17375589.0,
      "step": 9890
    },
    {
      "entropy": 5.138284587860108,
      "epoch": 7.754702194357367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00011247215305060213,
      "loss": 4.3351,
      "mean_token_accuracy": 0.24647124856710434,
      "num_tokens": 17385816.0,
      "step": 9895
    },
    {
      "entropy": 5.026936101913452,
      "epoch": 7.758620689655173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011226429194082627,
      "loss": 4.221,
      "mean_token_accuracy": 0.25574178248643875,
      "num_tokens": 17395267.0,
      "step": 9900
    },
    {
      "entropy": 5.126526165008545,
      "epoch": 7.762539184952978,
      "grad_norm": 1.359375,
      "learning_rate": 0.000112056721666147,
      "loss": 4.2935,
      "mean_token_accuracy": 0.248771770298481,
      "num_tokens": 17403544.0,
      "step": 9905
    },
    {
      "entropy": 5.087206029891968,
      "epoch": 7.766457680250784,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00011184944259752622,
      "loss": 4.2732,
      "mean_token_accuracy": 0.25172866731882093,
      "num_tokens": 17412090.0,
      "step": 9910
    },
    {
      "entropy": 5.059094429016113,
      "epoch": 7.77037617554859,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00011164245510540512,
      "loss": 4.2719,
      "mean_token_accuracy": 0.24497840255498887,
      "num_tokens": 17420558.0,
      "step": 9915
    },
    {
      "entropy": 5.063074207305908,
      "epoch": 7.774294670846395,
      "grad_norm": 1.453125,
      "learning_rate": 0.00011143575955970384,
      "loss": 4.2917,
      "mean_token_accuracy": 0.24533278942108155,
      "num_tokens": 17429901.0,
      "step": 9920
    },
    {
      "entropy": 5.113965368270874,
      "epoch": 7.778213166144201,
      "grad_norm": 1.46875,
      "learning_rate": 0.00011122935632982101,
      "loss": 4.3124,
      "mean_token_accuracy": 0.24214827865362168,
      "num_tokens": 17438543.0,
      "step": 9925
    },
    {
      "entropy": 5.1641768455505375,
      "epoch": 7.782131661442007,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001110232457846325,
      "loss": 4.3241,
      "mean_token_accuracy": 0.24485959857702255,
      "num_tokens": 17447234.0,
      "step": 9930
    },
    {
      "entropy": 5.028472518920898,
      "epoch": 7.786050156739812,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011081742829249142,
      "loss": 4.2011,
      "mean_token_accuracy": 0.2535095915198326,
      "num_tokens": 17455542.0,
      "step": 9935
    },
    {
      "entropy": 5.08992190361023,
      "epoch": 7.789968652037618,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011061190422122687,
      "loss": 4.2587,
      "mean_token_accuracy": 0.2569694295525551,
      "num_tokens": 17465035.0,
      "step": 9940
    },
    {
      "entropy": 5.097151422500611,
      "epoch": 7.793887147335424,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011040667393814383,
      "loss": 4.2747,
      "mean_token_accuracy": 0.24987898617982865,
      "num_tokens": 17473748.0,
      "step": 9945
    },
    {
      "entropy": 5.020683431625367,
      "epoch": 7.797805642633229,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00011020173781002199,
      "loss": 4.1613,
      "mean_token_accuracy": 0.260040320456028,
      "num_tokens": 17481525.0,
      "step": 9950
    },
    {
      "entropy": 5.10503306388855,
      "epoch": 7.801724137931035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001099970962031154,
      "loss": 4.3151,
      "mean_token_accuracy": 0.24635845124721528,
      "num_tokens": 17490044.0,
      "step": 9955
    },
    {
      "entropy": 5.1620183944702145,
      "epoch": 7.80564263322884,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00010979274948315191,
      "loss": 4.3569,
      "mean_token_accuracy": 0.24108865559101106,
      "num_tokens": 17498094.0,
      "step": 9960
    },
    {
      "entropy": 5.088546419143677,
      "epoch": 7.809561128526646,
      "grad_norm": 1.390625,
      "learning_rate": 0.00010958869801533222,
      "loss": 4.2944,
      "mean_token_accuracy": 0.24892078340053558,
      "num_tokens": 17507677.0,
      "step": 9965
    },
    {
      "entropy": 5.0620481967926025,
      "epoch": 7.813479623824452,
      "grad_norm": 1.359375,
      "learning_rate": 0.00010938494216432928,
      "loss": 4.2632,
      "mean_token_accuracy": 0.24457866102457046,
      "num_tokens": 17516489.0,
      "step": 9970
    },
    {
      "entropy": 5.076725244522095,
      "epoch": 7.817398119122257,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010918148229428796,
      "loss": 4.3045,
      "mean_token_accuracy": 0.24815995395183563,
      "num_tokens": 17524972.0,
      "step": 9975
    },
    {
      "entropy": 5.067804431915283,
      "epoch": 7.821316614420063,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000108978318768824,
      "loss": 4.2868,
      "mean_token_accuracy": 0.2460113912820816,
      "num_tokens": 17533745.0,
      "step": 9980
    },
    {
      "entropy": 5.0432346820831295,
      "epoch": 7.825235109717868,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010877545195102347,
      "loss": 4.274,
      "mean_token_accuracy": 0.2387847885489464,
      "num_tokens": 17542812.0,
      "step": 9985
    },
    {
      "entropy": 5.072667407989502,
      "epoch": 7.829153605015674,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010857288220344242,
      "loss": 4.2457,
      "mean_token_accuracy": 0.2522255554795265,
      "num_tokens": 17551123.0,
      "step": 9990
    },
    {
      "entropy": 5.0924781322479244,
      "epoch": 7.83307210031348,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010837060988810571,
      "loss": 4.3619,
      "mean_token_accuracy": 0.23770271837711335,
      "num_tokens": 17560729.0,
      "step": 9995
    },
    {
      "entropy": 5.122906684875488,
      "epoch": 7.836990595611285,
      "grad_norm": 1.375,
      "learning_rate": 0.00010816863536650687,
      "loss": 4.3018,
      "mean_token_accuracy": 0.24498500525951386,
      "num_tokens": 17569842.0,
      "step": 10000
    },
    {
      "epoch": 7.836990595611285,
      "eval_entropy": 5.00048799662627,
      "eval_loss": 5.6207146644592285,
      "eval_mean_token_accuracy": 0.17562680871969508,
      "eval_num_tokens": 17569842.0,
      "eval_runtime": 2.8332,
      "eval_samples_per_second": 1454.875,
      "eval_steps_per_second": 182.124,
      "step": 10000
    },
    {
      "entropy": 5.072719764709473,
      "epoch": 7.840909090909091,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010796695899960696,
      "loss": 4.2084,
      "mean_token_accuracy": 0.25993279069662095,
      "num_tokens": 17578887.0,
      "step": 10005
    },
    {
      "entropy": 5.0902135372161865,
      "epoch": 7.844827586206897,
      "grad_norm": 1.40625,
      "learning_rate": 0.00010776558114783455,
      "loss": 4.2403,
      "mean_token_accuracy": 0.2469596803188324,
      "num_tokens": 17587321.0,
      "step": 10010
    },
    {
      "entropy": 5.075358247756958,
      "epoch": 7.848746081504702,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00010756450217108441,
      "loss": 4.2584,
      "mean_token_accuracy": 0.24979417622089387,
      "num_tokens": 17596295.0,
      "step": 10015
    },
    {
      "entropy": 5.074442291259766,
      "epoch": 7.852664576802508,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001073637224287172,
      "loss": 4.2338,
      "mean_token_accuracy": 0.2557582601904869,
      "num_tokens": 17605033.0,
      "step": 10020
    },
    {
      "entropy": 5.00795259475708,
      "epoch": 7.856583072100314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010716324227955904,
      "loss": 4.2573,
      "mean_token_accuracy": 0.25106081962585447,
      "num_tokens": 17613706.0,
      "step": 10025
    },
    {
      "entropy": 5.111376571655273,
      "epoch": 7.860501567398119,
      "grad_norm": 1.296875,
      "learning_rate": 0.00010696306208190034,
      "loss": 4.3156,
      "mean_token_accuracy": 0.24292644560337068,
      "num_tokens": 17622082.0,
      "step": 10030
    },
    {
      "entropy": 4.9746565341949465,
      "epoch": 7.864420062695925,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010676318219349564,
      "loss": 4.1995,
      "mean_token_accuracy": 0.25431394577026367,
      "num_tokens": 17630980.0,
      "step": 10035
    },
    {
      "entropy": 5.032153701782226,
      "epoch": 7.868338557993731,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010656360297156262,
      "loss": 4.2472,
      "mean_token_accuracy": 0.24854181408882142,
      "num_tokens": 17639679.0,
      "step": 10040
    },
    {
      "entropy": 4.998709058761596,
      "epoch": 7.872257053291536,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001063643247727818,
      "loss": 4.1917,
      "mean_token_accuracy": 0.26047023236751554,
      "num_tokens": 17648542.0,
      "step": 10045
    },
    {
      "entropy": 5.028048372268676,
      "epoch": 7.876175548589342,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010616534795329553,
      "loss": 4.1807,
      "mean_token_accuracy": 0.25642394870519636,
      "num_tokens": 17657305.0,
      "step": 10050
    },
    {
      "entropy": 5.089817333221435,
      "epoch": 7.880094043887147,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00010596667286870778,
      "loss": 4.301,
      "mean_token_accuracy": 0.24390340745449066,
      "num_tokens": 17665876.0,
      "step": 10055
    },
    {
      "entropy": 5.059748029708862,
      "epoch": 7.884012539184953,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010576829987408304,
      "loss": 4.2395,
      "mean_token_accuracy": 0.25311962962150575,
      "num_tokens": 17674756.0,
      "step": 10060
    },
    {
      "entropy": 5.044168424606323,
      "epoch": 7.887931034482759,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010557022932394611,
      "loss": 4.2011,
      "mean_token_accuracy": 0.271086984872818,
      "num_tokens": 17683989.0,
      "step": 10065
    },
    {
      "entropy": 5.14534740447998,
      "epoch": 7.891849529780564,
      "grad_norm": 1.328125,
      "learning_rate": 0.00010537246157228104,
      "loss": 4.2956,
      "mean_token_accuracy": 0.2593579351902008,
      "num_tokens": 17692776.0,
      "step": 10070
    },
    {
      "entropy": 5.0667396068573,
      "epoch": 7.89576802507837,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010517499697253103,
      "loss": 4.1731,
      "mean_token_accuracy": 0.25418960154056547,
      "num_tokens": 17701428.0,
      "step": 10075
    },
    {
      "entropy": 5.066579675674438,
      "epoch": 7.899686520376176,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010497783587759724,
      "loss": 4.3023,
      "mean_token_accuracy": 0.24250917583703996,
      "num_tokens": 17710343.0,
      "step": 10080
    },
    {
      "entropy": 5.074942636489868,
      "epoch": 7.903605015673981,
      "grad_norm": 1.46875,
      "learning_rate": 0.00010478097863983857,
      "loss": 4.2965,
      "mean_token_accuracy": 0.24747479259967803,
      "num_tokens": 17719218.0,
      "step": 10085
    },
    {
      "entropy": 5.122753858566284,
      "epoch": 7.907523510971787,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001045844256110707,
      "loss": 4.3434,
      "mean_token_accuracy": 0.23858604729175567,
      "num_tokens": 17727288.0,
      "step": 10090
    },
    {
      "entropy": 5.111692476272583,
      "epoch": 7.911442006269592,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010438817714256589,
      "loss": 4.3255,
      "mean_token_accuracy": 0.2442009598016739,
      "num_tokens": 17736130.0,
      "step": 10095
    },
    {
      "entropy": 5.067315673828125,
      "epoch": 7.915360501567398,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010419223358505196,
      "loss": 4.2115,
      "mean_token_accuracy": 0.2631120473146439,
      "num_tokens": 17745122.0,
      "step": 10100
    },
    {
      "entropy": 5.02108325958252,
      "epoch": 7.919278996865204,
      "grad_norm": 1.484375,
      "learning_rate": 0.00010399659528871169,
      "loss": 4.2257,
      "mean_token_accuracy": 0.2563833400607109,
      "num_tokens": 17753817.0,
      "step": 10105
    },
    {
      "entropy": 5.1118405818939205,
      "epoch": 7.923197492163009,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00010380126260318265,
      "loss": 4.3044,
      "mean_token_accuracy": 0.2472682625055313,
      "num_tokens": 17762793.0,
      "step": 10110
    },
    {
      "entropy": 5.024303150177002,
      "epoch": 7.927115987460815,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010360623587755579,
      "loss": 4.262,
      "mean_token_accuracy": 0.2532124534249306,
      "num_tokens": 17772223.0,
      "step": 10115
    },
    {
      "entropy": 5.045841455459595,
      "epoch": 7.931034482758621,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001034115154603757,
      "loss": 4.2284,
      "mean_token_accuracy": 0.24744421541690825,
      "num_tokens": 17781543.0,
      "step": 10120
    },
    {
      "entropy": 5.045814561843872,
      "epoch": 7.934952978056426,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00010321710169963919,
      "loss": 4.2443,
      "mean_token_accuracy": 0.24423699378967284,
      "num_tokens": 17790258.0,
      "step": 10125
    },
    {
      "entropy": 5.036077833175659,
      "epoch": 7.938871473354232,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010302299494279527,
      "loss": 4.238,
      "mean_token_accuracy": 0.24767802208662032,
      "num_tokens": 17798862.0,
      "step": 10130
    },
    {
      "entropy": 5.071421909332275,
      "epoch": 7.942789968652038,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010282919553674415,
      "loss": 4.2914,
      "mean_token_accuracy": 0.24781841039657593,
      "num_tokens": 17807285.0,
      "step": 10135
    },
    {
      "entropy": 5.073719930648804,
      "epoch": 7.946708463949843,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00010263570382783675,
      "loss": 4.3269,
      "mean_token_accuracy": 0.24135707020759584,
      "num_tokens": 17816891.0,
      "step": 10140
    },
    {
      "entropy": 5.0865143775939945,
      "epoch": 7.950626959247649,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010244252016187419,
      "loss": 4.2846,
      "mean_token_accuracy": 0.25066730082035066,
      "num_tokens": 17825895.0,
      "step": 10145
    },
    {
      "entropy": 5.127944850921631,
      "epoch": 7.954545454545455,
      "grad_norm": 1.453125,
      "learning_rate": 0.00010224964488410699,
      "loss": 4.3493,
      "mean_token_accuracy": 0.2460557997226715,
      "num_tokens": 17834921.0,
      "step": 10150
    },
    {
      "entropy": 5.111541080474853,
      "epoch": 7.95846394984326,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001020570783392345,
      "loss": 4.2531,
      "mean_token_accuracy": 0.25344068706035616,
      "num_tokens": 17844310.0,
      "step": 10155
    },
    {
      "entropy": 5.159103631973267,
      "epoch": 7.962382445141066,
      "grad_norm": 1.375,
      "learning_rate": 0.00010186482087140439,
      "loss": 4.2932,
      "mean_token_accuracy": 0.2503374546766281,
      "num_tokens": 17853561.0,
      "step": 10160
    },
    {
      "entropy": 5.022298955917359,
      "epoch": 7.966300940438871,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000101672872824212,
      "loss": 4.2443,
      "mean_token_accuracy": 0.2432657554745674,
      "num_tokens": 17863077.0,
      "step": 10165
    },
    {
      "entropy": 5.07559871673584,
      "epoch": 7.970219435736677,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00010148123454069943,
      "loss": 4.1899,
      "mean_token_accuracy": 0.2576782926917076,
      "num_tokens": 17871780.0,
      "step": 10170
    },
    {
      "entropy": 5.040722894668579,
      "epoch": 7.974137931034483,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001012899063633555,
      "loss": 4.2261,
      "mean_token_accuracy": 0.25221225172281264,
      "num_tokens": 17881144.0,
      "step": 10175
    },
    {
      "entropy": 5.040059232711792,
      "epoch": 7.978056426332288,
      "grad_norm": 1.3125,
      "learning_rate": 0.00010109888863411473,
      "loss": 4.2893,
      "mean_token_accuracy": 0.24712306559085845,
      "num_tokens": 17890754.0,
      "step": 10180
    },
    {
      "entropy": 5.10335340499878,
      "epoch": 7.981974921630094,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001009081816943567,
      "loss": 4.3526,
      "mean_token_accuracy": 0.24287159889936447,
      "num_tokens": 17899992.0,
      "step": 10185
    },
    {
      "entropy": 5.023664283752441,
      "epoch": 7.985893416927899,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010071778588490558,
      "loss": 4.2238,
      "mean_token_accuracy": 0.2545998588204384,
      "num_tokens": 17907847.0,
      "step": 10190
    },
    {
      "entropy": 5.0852453231811525,
      "epoch": 7.989811912225705,
      "grad_norm": 1.34375,
      "learning_rate": 0.00010052770154602968,
      "loss": 4.2767,
      "mean_token_accuracy": 0.24814445674419403,
      "num_tokens": 17916566.0,
      "step": 10195
    },
    {
      "entropy": 5.042820644378662,
      "epoch": 7.993730407523511,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00010033792901744046,
      "loss": 4.165,
      "mean_token_accuracy": 0.26123504936695097,
      "num_tokens": 17924969.0,
      "step": 10200
    },
    {
      "entropy": 5.0738208293914795,
      "epoch": 7.997648902821316,
      "grad_norm": 1.375,
      "learning_rate": 0.00010014846863829221,
      "loss": 4.1843,
      "mean_token_accuracy": 0.2536504790186882,
      "num_tokens": 17933776.0,
      "step": 10205
    },
    {
      "entropy": 4.994382858276367,
      "epoch": 8.001567398119123,
      "grad_norm": 1.4765625,
      "learning_rate": 9.995932074718131e-05,
      "loss": 4.1655,
      "mean_token_accuracy": 0.25610885918140414,
      "num_tokens": 17941651.0,
      "step": 10210
    },
    {
      "entropy": 5.078711938858032,
      "epoch": 8.005485893416928,
      "grad_norm": 1.453125,
      "learning_rate": 9.977048568214578e-05,
      "loss": 4.2909,
      "mean_token_accuracy": 0.2511640265583992,
      "num_tokens": 17951278.0,
      "step": 10215
    },
    {
      "entropy": 5.042738246917724,
      "epoch": 8.009404388714733,
      "grad_norm": 1.359375,
      "learning_rate": 9.958196378066444e-05,
      "loss": 4.2048,
      "mean_token_accuracy": 0.2576929584145546,
      "num_tokens": 17960182.0,
      "step": 10220
    },
    {
      "entropy": 5.0350720405578615,
      "epoch": 8.013322884012538,
      "grad_norm": 1.5,
      "learning_rate": 9.93937553796565e-05,
      "loss": 4.175,
      "mean_token_accuracy": 0.25795582979917525,
      "num_tokens": 17968221.0,
      "step": 10225
    },
    {
      "entropy": 5.04392442703247,
      "epoch": 8.017241379310345,
      "grad_norm": 1.4375,
      "learning_rate": 9.920586081548089e-05,
      "loss": 4.0718,
      "mean_token_accuracy": 0.2677145302295685,
      "num_tokens": 17975724.0,
      "step": 10230
    },
    {
      "entropy": 5.049734449386596,
      "epoch": 8.02115987460815,
      "grad_norm": 1.4140625,
      "learning_rate": 9.901828042393573e-05,
      "loss": 4.2808,
      "mean_token_accuracy": 0.24945725947618486,
      "num_tokens": 17984425.0,
      "step": 10235
    },
    {
      "entropy": 5.037394428253174,
      "epoch": 8.025078369905955,
      "grad_norm": 1.4765625,
      "learning_rate": 9.88310145402575e-05,
      "loss": 4.1094,
      "mean_token_accuracy": 0.26025208085775375,
      "num_tokens": 17993319.0,
      "step": 10240
    },
    {
      "entropy": 5.041788101196289,
      "epoch": 8.028996865203762,
      "grad_norm": 1.2578125,
      "learning_rate": 9.864406349912081e-05,
      "loss": 4.1211,
      "mean_token_accuracy": 0.26115481108427047,
      "num_tokens": 18002239.0,
      "step": 10245
    },
    {
      "entropy": 5.077745008468628,
      "epoch": 8.032915360501567,
      "grad_norm": 1.3828125,
      "learning_rate": 9.845742763463739e-05,
      "loss": 4.2327,
      "mean_token_accuracy": 0.25663321018218993,
      "num_tokens": 18010745.0,
      "step": 10250
    },
    {
      "entropy": 5.034240293502807,
      "epoch": 8.036833855799372,
      "grad_norm": 1.3671875,
      "learning_rate": 9.827110728035584e-05,
      "loss": 4.2002,
      "mean_token_accuracy": 0.2556937143206596,
      "num_tokens": 18019535.0,
      "step": 10255
    },
    {
      "entropy": 5.071003484725952,
      "epoch": 8.04075235109718,
      "grad_norm": 1.1953125,
      "learning_rate": 9.808510276926075e-05,
      "loss": 4.1979,
      "mean_token_accuracy": 0.2556600034236908,
      "num_tokens": 18029128.0,
      "step": 10260
    },
    {
      "entropy": 5.080536985397339,
      "epoch": 8.044670846394984,
      "grad_norm": 1.234375,
      "learning_rate": 9.789941443377249e-05,
      "loss": 4.2325,
      "mean_token_accuracy": 0.2518587067723274,
      "num_tokens": 18038304.0,
      "step": 10265
    },
    {
      "entropy": 5.062565040588379,
      "epoch": 8.04858934169279,
      "grad_norm": 1.421875,
      "learning_rate": 9.771404260574613e-05,
      "loss": 4.19,
      "mean_token_accuracy": 0.25858530700206755,
      "num_tokens": 18046628.0,
      "step": 10270
    },
    {
      "entropy": 5.123085880279541,
      "epoch": 8.052507836990596,
      "grad_norm": 1.3828125,
      "learning_rate": 9.752898761647113e-05,
      "loss": 4.2668,
      "mean_token_accuracy": 0.2483106255531311,
      "num_tokens": 18055654.0,
      "step": 10275
    },
    {
      "entropy": 5.0364823818206785,
      "epoch": 8.056426332288401,
      "grad_norm": 1.265625,
      "learning_rate": 9.734424979667089e-05,
      "loss": 4.1944,
      "mean_token_accuracy": 0.26146974563598635,
      "num_tokens": 18065844.0,
      "step": 10280
    },
    {
      "entropy": 5.075096082687378,
      "epoch": 8.060344827586206,
      "grad_norm": 1.3515625,
      "learning_rate": 9.715982947650179e-05,
      "loss": 4.2271,
      "mean_token_accuracy": 0.2509082779288292,
      "num_tokens": 18074497.0,
      "step": 10285
    },
    {
      "entropy": 5.1153459548950195,
      "epoch": 8.064263322884013,
      "grad_norm": 1.359375,
      "learning_rate": 9.697572698555278e-05,
      "loss": 4.2336,
      "mean_token_accuracy": 0.2526580452919006,
      "num_tokens": 18083967.0,
      "step": 10290
    },
    {
      "entropy": 5.038693189620972,
      "epoch": 8.068181818181818,
      "grad_norm": 1.46875,
      "learning_rate": 9.679194265284492e-05,
      "loss": 4.2142,
      "mean_token_accuracy": 0.2566305920481682,
      "num_tokens": 18092263.0,
      "step": 10295
    },
    {
      "entropy": 5.1327033042907715,
      "epoch": 8.072100313479623,
      "grad_norm": 1.453125,
      "learning_rate": 9.660847680683069e-05,
      "loss": 4.257,
      "mean_token_accuracy": 0.25165228843688964,
      "num_tokens": 18101559.0,
      "step": 10300
    },
    {
      "entropy": 5.126171112060547,
      "epoch": 8.07601880877743,
      "grad_norm": 1.453125,
      "learning_rate": 9.64253297753932e-05,
      "loss": 4.1911,
      "mean_token_accuracy": 0.26437978744506835,
      "num_tokens": 18109850.0,
      "step": 10305
    },
    {
      "entropy": 5.079541730880737,
      "epoch": 8.079937304075235,
      "grad_norm": 1.375,
      "learning_rate": 9.62425018858459e-05,
      "loss": 4.167,
      "mean_token_accuracy": 0.25853431671857835,
      "num_tokens": 18119205.0,
      "step": 10310
    },
    {
      "entropy": 5.046890926361084,
      "epoch": 8.08385579937304,
      "grad_norm": 1.40625,
      "learning_rate": 9.605999346493194e-05,
      "loss": 4.1579,
      "mean_token_accuracy": 0.25372861325740814,
      "num_tokens": 18128713.0,
      "step": 10315
    },
    {
      "entropy": 5.070139074325562,
      "epoch": 8.087774294670846,
      "grad_norm": 1.5234375,
      "learning_rate": 9.587780483882339e-05,
      "loss": 4.1416,
      "mean_token_accuracy": 0.2588116616010666,
      "num_tokens": 18137080.0,
      "step": 10320
    },
    {
      "entropy": 5.108174562454224,
      "epoch": 8.091692789968652,
      "grad_norm": 1.4140625,
      "learning_rate": 9.569593633312084e-05,
      "loss": 4.2624,
      "mean_token_accuracy": 0.2513464942574501,
      "num_tokens": 18146449.0,
      "step": 10325
    },
    {
      "entropy": 5.118718528747559,
      "epoch": 8.095611285266457,
      "grad_norm": 1.421875,
      "learning_rate": 9.551438827285294e-05,
      "loss": 4.2813,
      "mean_token_accuracy": 0.24313308000564576,
      "num_tokens": 18155394.0,
      "step": 10330
    },
    {
      "entropy": 5.102547597885132,
      "epoch": 8.099529780564263,
      "grad_norm": 1.296875,
      "learning_rate": 9.533316098247539e-05,
      "loss": 4.2748,
      "mean_token_accuracy": 0.24257466942071915,
      "num_tokens": 18164538.0,
      "step": 10335
    },
    {
      "entropy": 5.095793199539185,
      "epoch": 8.10344827586207,
      "grad_norm": 1.34375,
      "learning_rate": 9.515225478587077e-05,
      "loss": 4.2097,
      "mean_token_accuracy": 0.25986034572124483,
      "num_tokens": 18174396.0,
      "step": 10340
    },
    {
      "entropy": 5.110096836090088,
      "epoch": 8.107366771159874,
      "grad_norm": 1.3828125,
      "learning_rate": 9.497167000634773e-05,
      "loss": 4.1708,
      "mean_token_accuracy": 0.2516869157552719,
      "num_tokens": 18183030.0,
      "step": 10345
    },
    {
      "entropy": 5.089336252212524,
      "epoch": 8.11128526645768,
      "grad_norm": 1.390625,
      "learning_rate": 9.47914069666406e-05,
      "loss": 4.2048,
      "mean_token_accuracy": 0.2521478205919266,
      "num_tokens": 18191139.0,
      "step": 10350
    },
    {
      "entropy": 5.014074373245239,
      "epoch": 8.115203761755486,
      "grad_norm": 1.3984375,
      "learning_rate": 9.461146598890873e-05,
      "loss": 4.1602,
      "mean_token_accuracy": 0.2595292657613754,
      "num_tokens": 18199429.0,
      "step": 10355
    },
    {
      "entropy": 5.067401027679443,
      "epoch": 8.119122257053291,
      "grad_norm": 1.484375,
      "learning_rate": 9.443184739473573e-05,
      "loss": 4.2199,
      "mean_token_accuracy": 0.25118866860866546,
      "num_tokens": 18207765.0,
      "step": 10360
    },
    {
      "entropy": 5.0376194477081295,
      "epoch": 8.123040752351097,
      "grad_norm": 1.4765625,
      "learning_rate": 9.425255150512921e-05,
      "loss": 4.2304,
      "mean_token_accuracy": 0.25082095712423325,
      "num_tokens": 18216524.0,
      "step": 10365
    },
    {
      "entropy": 5.0840802669525145,
      "epoch": 8.126959247648903,
      "grad_norm": 1.484375,
      "learning_rate": 9.407357864051996e-05,
      "loss": 4.2049,
      "mean_token_accuracy": 0.2587078869342804,
      "num_tokens": 18224647.0,
      "step": 10370
    },
    {
      "entropy": 5.055821514129638,
      "epoch": 8.130877742946709,
      "grad_norm": 1.4296875,
      "learning_rate": 9.389492912076156e-05,
      "loss": 4.1998,
      "mean_token_accuracy": 0.255357126891613,
      "num_tokens": 18234127.0,
      "step": 10375
    },
    {
      "entropy": 5.054931592941284,
      "epoch": 8.134796238244514,
      "grad_norm": 1.3984375,
      "learning_rate": 9.371660326512952e-05,
      "loss": 4.1334,
      "mean_token_accuracy": 0.26156606078147887,
      "num_tokens": 18242455.0,
      "step": 10380
    },
    {
      "entropy": 5.123177337646484,
      "epoch": 8.13871473354232,
      "grad_norm": 1.2890625,
      "learning_rate": 9.353860139232126e-05,
      "loss": 4.2573,
      "mean_token_accuracy": 0.24804783165454863,
      "num_tokens": 18251201.0,
      "step": 10385
    },
    {
      "entropy": 5.073599338531494,
      "epoch": 8.142633228840126,
      "grad_norm": 1.3515625,
      "learning_rate": 9.336092382045488e-05,
      "loss": 4.1998,
      "mean_token_accuracy": 0.24886968582868577,
      "num_tokens": 18260651.0,
      "step": 10390
    },
    {
      "entropy": 5.1325524806976315,
      "epoch": 8.14655172413793,
      "grad_norm": 1.4609375,
      "learning_rate": 9.318357086706894e-05,
      "loss": 4.2573,
      "mean_token_accuracy": 0.25445543229579926,
      "num_tokens": 18269218.0,
      "step": 10395
    },
    {
      "entropy": 5.0846912384033205,
      "epoch": 8.150470219435737,
      "grad_norm": 1.4453125,
      "learning_rate": 9.300654284912211e-05,
      "loss": 4.2053,
      "mean_token_accuracy": 0.24805504828691483,
      "num_tokens": 18277371.0,
      "step": 10400
    },
    {
      "entropy": 5.051848888397217,
      "epoch": 8.154388714733543,
      "grad_norm": 1.2578125,
      "learning_rate": 9.282984008299196e-05,
      "loss": 4.2277,
      "mean_token_accuracy": 0.24882320463657379,
      "num_tokens": 18287503.0,
      "step": 10405
    },
    {
      "entropy": 5.059165143966675,
      "epoch": 8.158307210031348,
      "grad_norm": 1.5390625,
      "learning_rate": 9.265346288447516e-05,
      "loss": 4.2226,
      "mean_token_accuracy": 0.25478904992341994,
      "num_tokens": 18296753.0,
      "step": 10410
    },
    {
      "entropy": 5.088523578643799,
      "epoch": 8.162225705329154,
      "grad_norm": 1.4140625,
      "learning_rate": 9.24774115687862e-05,
      "loss": 4.2533,
      "mean_token_accuracy": 0.2531737327575684,
      "num_tokens": 18305068.0,
      "step": 10415
    },
    {
      "entropy": 5.084874057769776,
      "epoch": 8.16614420062696,
      "grad_norm": 1.3203125,
      "learning_rate": 9.23016864505575e-05,
      "loss": 4.2381,
      "mean_token_accuracy": 0.2569823622703552,
      "num_tokens": 18313816.0,
      "step": 10420
    },
    {
      "entropy": 5.0716986656188965,
      "epoch": 8.170062695924765,
      "grad_norm": 1.296875,
      "learning_rate": 9.212628784383825e-05,
      "loss": 4.1616,
      "mean_token_accuracy": 0.26139300018548967,
      "num_tokens": 18322630.0,
      "step": 10425
    },
    {
      "entropy": 5.025461912155151,
      "epoch": 8.17398119122257,
      "grad_norm": 1.4453125,
      "learning_rate": 9.195121606209411e-05,
      "loss": 4.1583,
      "mean_token_accuracy": 0.2571008116006851,
      "num_tokens": 18330942.0,
      "step": 10430
    },
    {
      "entropy": 5.028965806961059,
      "epoch": 8.177899686520377,
      "grad_norm": 1.46875,
      "learning_rate": 9.177647141820693e-05,
      "loss": 4.2193,
      "mean_token_accuracy": 0.25611336827278136,
      "num_tokens": 18339375.0,
      "step": 10435
    },
    {
      "entropy": 5.117905998229981,
      "epoch": 8.181818181818182,
      "grad_norm": 1.4453125,
      "learning_rate": 9.160205422447357e-05,
      "loss": 4.2167,
      "mean_token_accuracy": 0.2549737349152565,
      "num_tokens": 18347884.0,
      "step": 10440
    },
    {
      "entropy": 5.063860177993774,
      "epoch": 8.185736677115987,
      "grad_norm": 1.3828125,
      "learning_rate": 9.142796479260588e-05,
      "loss": 4.219,
      "mean_token_accuracy": 0.2589388132095337,
      "num_tokens": 18357201.0,
      "step": 10445
    },
    {
      "entropy": 5.112305641174316,
      "epoch": 8.189655172413794,
      "grad_norm": 1.3828125,
      "learning_rate": 9.125420343372995e-05,
      "loss": 4.277,
      "mean_token_accuracy": 0.24722056686878205,
      "num_tokens": 18365716.0,
      "step": 10450
    },
    {
      "entropy": 5.046955537796021,
      "epoch": 8.193573667711599,
      "grad_norm": 1.2890625,
      "learning_rate": 9.108077045838546e-05,
      "loss": 4.186,
      "mean_token_accuracy": 0.26419543474912643,
      "num_tokens": 18374724.0,
      "step": 10455
    },
    {
      "entropy": 5.078075122833252,
      "epoch": 8.197492163009404,
      "grad_norm": 1.421875,
      "learning_rate": 9.090766617652525e-05,
      "loss": 4.2267,
      "mean_token_accuracy": 0.2558902516961098,
      "num_tokens": 18383653.0,
      "step": 10460
    },
    {
      "entropy": 5.083233880996704,
      "epoch": 8.20141065830721,
      "grad_norm": 1.421875,
      "learning_rate": 9.073489089751468e-05,
      "loss": 4.208,
      "mean_token_accuracy": 0.255522058904171,
      "num_tokens": 18392729.0,
      "step": 10465
    },
    {
      "entropy": 5.1120631217956545,
      "epoch": 8.205329153605016,
      "grad_norm": 1.3359375,
      "learning_rate": 9.056244493013129e-05,
      "loss": 4.2528,
      "mean_token_accuracy": 0.24974115788936616,
      "num_tokens": 18402257.0,
      "step": 10470
    },
    {
      "entropy": 5.095790958404541,
      "epoch": 8.20924764890282,
      "grad_norm": 1.4140625,
      "learning_rate": 9.039032858256401e-05,
      "loss": 4.2489,
      "mean_token_accuracy": 0.25502512902021407,
      "num_tokens": 18410714.0,
      "step": 10475
    },
    {
      "entropy": 5.073963451385498,
      "epoch": 8.213166144200628,
      "grad_norm": 1.2890625,
      "learning_rate": 9.021854216241251e-05,
      "loss": 4.2795,
      "mean_token_accuracy": 0.24827054589986802,
      "num_tokens": 18419256.0,
      "step": 10480
    },
    {
      "entropy": 5.069138860702514,
      "epoch": 8.217084639498433,
      "grad_norm": 1.3828125,
      "learning_rate": 9.004708597668716e-05,
      "loss": 4.2055,
      "mean_token_accuracy": 0.25936128944158554,
      "num_tokens": 18428202.0,
      "step": 10485
    },
    {
      "entropy": 5.098595476150512,
      "epoch": 8.221003134796238,
      "grad_norm": 1.4765625,
      "learning_rate": 8.987596033180784e-05,
      "loss": 4.1905,
      "mean_token_accuracy": 0.2562639206647873,
      "num_tokens": 18437339.0,
      "step": 10490
    },
    {
      "entropy": 5.106858730316162,
      "epoch": 8.224921630094045,
      "grad_norm": 1.4140625,
      "learning_rate": 8.970516553360383e-05,
      "loss": 4.208,
      "mean_token_accuracy": 0.25522884875535967,
      "num_tokens": 18446667.0,
      "step": 10495
    },
    {
      "entropy": 5.083641767501831,
      "epoch": 8.22884012539185,
      "grad_norm": 1.5078125,
      "learning_rate": 8.953470188731318e-05,
      "loss": 4.1405,
      "mean_token_accuracy": 0.2616897150874138,
      "num_tokens": 18455425.0,
      "step": 10500
    },
    {
      "epoch": 8.22884012539185,
      "eval_entropy": 4.9981888426366705,
      "eval_loss": 5.623587131500244,
      "eval_mean_token_accuracy": 0.1758962116644595,
      "eval_num_tokens": 18455425.0,
      "eval_runtime": 3.0233,
      "eval_samples_per_second": 1363.409,
      "eval_steps_per_second": 170.674,
      "step": 10500
    },
    {
      "entropy": 5.043186330795288,
      "epoch": 8.232758620689655,
      "grad_norm": 1.390625,
      "learning_rate": 8.936456969758206e-05,
      "loss": 4.1154,
      "mean_token_accuracy": 0.2624488815665245,
      "num_tokens": 18464976.0,
      "step": 10505
    },
    {
      "entropy": 5.0838254451751705,
      "epoch": 8.236677115987462,
      "grad_norm": 1.34375,
      "learning_rate": 8.919476926846426e-05,
      "loss": 4.1762,
      "mean_token_accuracy": 0.2560855135321617,
      "num_tokens": 18473709.0,
      "step": 10510
    },
    {
      "entropy": 5.089413976669311,
      "epoch": 8.240595611285267,
      "grad_norm": 1.3984375,
      "learning_rate": 8.90253009034206e-05,
      "loss": 4.2047,
      "mean_token_accuracy": 0.2554166316986084,
      "num_tokens": 18483208.0,
      "step": 10515
    },
    {
      "entropy": 5.141398811340332,
      "epoch": 8.244514106583072,
      "grad_norm": 1.4375,
      "learning_rate": 8.88561649053187e-05,
      "loss": 4.2718,
      "mean_token_accuracy": 0.25413823276758196,
      "num_tokens": 18491872.0,
      "step": 10520
    },
    {
      "entropy": 5.12670259475708,
      "epoch": 8.248432601880877,
      "grad_norm": 1.390625,
      "learning_rate": 8.868736157643183e-05,
      "loss": 4.29,
      "mean_token_accuracy": 0.24386305809020997,
      "num_tokens": 18500740.0,
      "step": 10525
    },
    {
      "entropy": 5.070064926147461,
      "epoch": 8.252351097178684,
      "grad_norm": 1.40625,
      "learning_rate": 8.851889121843904e-05,
      "loss": 4.1948,
      "mean_token_accuracy": 0.25742192268371583,
      "num_tokens": 18509553.0,
      "step": 10530
    },
    {
      "entropy": 5.117561101913452,
      "epoch": 8.256269592476489,
      "grad_norm": 1.546875,
      "learning_rate": 8.835075413242408e-05,
      "loss": 4.2786,
      "mean_token_accuracy": 0.2514671429991722,
      "num_tokens": 18518549.0,
      "step": 10535
    },
    {
      "entropy": 5.135479164123535,
      "epoch": 8.260188087774294,
      "grad_norm": 1.4609375,
      "learning_rate": 8.818295061887531e-05,
      "loss": 4.2365,
      "mean_token_accuracy": 0.25053566247224807,
      "num_tokens": 18527264.0,
      "step": 10540
    },
    {
      "entropy": 5.097079277038574,
      "epoch": 8.2641065830721,
      "grad_norm": 1.5,
      "learning_rate": 8.801548097768473e-05,
      "loss": 4.1871,
      "mean_token_accuracy": 0.26063693016767503,
      "num_tokens": 18535193.0,
      "step": 10545
    },
    {
      "entropy": 5.0923402309417725,
      "epoch": 8.268025078369906,
      "grad_norm": 1.3125,
      "learning_rate": 8.784834550814769e-05,
      "loss": 4.226,
      "mean_token_accuracy": 0.25717274844646454,
      "num_tokens": 18544386.0,
      "step": 10550
    },
    {
      "entropy": 5.032943964004517,
      "epoch": 8.271943573667711,
      "grad_norm": 1.40625,
      "learning_rate": 8.768154450896253e-05,
      "loss": 4.0908,
      "mean_token_accuracy": 0.26699895560741427,
      "num_tokens": 18552745.0,
      "step": 10555
    },
    {
      "entropy": 5.09471173286438,
      "epoch": 8.275862068965518,
      "grad_norm": 1.359375,
      "learning_rate": 8.751507827822955e-05,
      "loss": 4.1864,
      "mean_token_accuracy": 0.2578438475728035,
      "num_tokens": 18560953.0,
      "step": 10560
    },
    {
      "entropy": 5.109809350967407,
      "epoch": 8.279780564263323,
      "grad_norm": 1.3515625,
      "learning_rate": 8.73489471134509e-05,
      "loss": 4.1361,
      "mean_token_accuracy": 0.27358477413654325,
      "num_tokens": 18569324.0,
      "step": 10565
    },
    {
      "entropy": 5.136924791336059,
      "epoch": 8.283699059561128,
      "grad_norm": 1.515625,
      "learning_rate": 8.718315131152999e-05,
      "loss": 4.271,
      "mean_token_accuracy": 0.25046689361333846,
      "num_tokens": 18578432.0,
      "step": 10570
    },
    {
      "entropy": 5.061399984359741,
      "epoch": 8.287617554858935,
      "grad_norm": 1.421875,
      "learning_rate": 8.701769116877076e-05,
      "loss": 4.2107,
      "mean_token_accuracy": 0.25385096967220305,
      "num_tokens": 18587147.0,
      "step": 10575
    },
    {
      "entropy": 5.11795597076416,
      "epoch": 8.29153605015674,
      "grad_norm": 1.2421875,
      "learning_rate": 8.685256698087725e-05,
      "loss": 4.2017,
      "mean_token_accuracy": 0.25357308238744736,
      "num_tokens": 18596933.0,
      "step": 10580
    },
    {
      "entropy": 5.156735563278199,
      "epoch": 8.295454545454545,
      "grad_norm": 1.359375,
      "learning_rate": 8.668777904295328e-05,
      "loss": 4.2639,
      "mean_token_accuracy": 0.2509946197271347,
      "num_tokens": 18605666.0,
      "step": 10585
    },
    {
      "entropy": 5.128520727157593,
      "epoch": 8.299373040752352,
      "grad_norm": 1.3671875,
      "learning_rate": 8.652332764950147e-05,
      "loss": 4.3347,
      "mean_token_accuracy": 0.24797298908233642,
      "num_tokens": 18614801.0,
      "step": 10590
    },
    {
      "entropy": 5.07725715637207,
      "epoch": 8.303291536050157,
      "grad_norm": 1.3125,
      "learning_rate": 8.635921309442327e-05,
      "loss": 4.1769,
      "mean_token_accuracy": 0.25820470303297044,
      "num_tokens": 18623489.0,
      "step": 10595
    },
    {
      "entropy": 5.073741483688354,
      "epoch": 8.307210031347962,
      "grad_norm": 1.5078125,
      "learning_rate": 8.619543567101787e-05,
      "loss": 4.186,
      "mean_token_accuracy": 0.2586586534976959,
      "num_tokens": 18631496.0,
      "step": 10600
    },
    {
      "entropy": 5.0894678115844725,
      "epoch": 8.311128526645769,
      "grad_norm": 1.5,
      "learning_rate": 8.603199567198222e-05,
      "loss": 4.233,
      "mean_token_accuracy": 0.2521470010280609,
      "num_tokens": 18640481.0,
      "step": 10605
    },
    {
      "entropy": 5.13355712890625,
      "epoch": 8.315047021943574,
      "grad_norm": 1.359375,
      "learning_rate": 8.586889338941002e-05,
      "loss": 4.3156,
      "mean_token_accuracy": 0.2467433527112007,
      "num_tokens": 18650118.0,
      "step": 10610
    },
    {
      "entropy": 5.05808048248291,
      "epoch": 8.318965517241379,
      "grad_norm": 1.390625,
      "learning_rate": 8.570612911479145e-05,
      "loss": 4.2008,
      "mean_token_accuracy": 0.25706599056720736,
      "num_tokens": 18658981.0,
      "step": 10615
    },
    {
      "entropy": 5.124745512008667,
      "epoch": 8.322884012539184,
      "grad_norm": 1.34375,
      "learning_rate": 8.554370313901273e-05,
      "loss": 4.2548,
      "mean_token_accuracy": 0.25144014656543734,
      "num_tokens": 18668324.0,
      "step": 10620
    },
    {
      "entropy": 5.055501651763916,
      "epoch": 8.326802507836991,
      "grad_norm": 1.40625,
      "learning_rate": 8.538161575235541e-05,
      "loss": 4.1579,
      "mean_token_accuracy": 0.2685968279838562,
      "num_tokens": 18676945.0,
      "step": 10625
    },
    {
      "entropy": 5.129256772994995,
      "epoch": 8.330721003134796,
      "grad_norm": 1.484375,
      "learning_rate": 8.521986724449593e-05,
      "loss": 4.2546,
      "mean_token_accuracy": 0.24906838983297347,
      "num_tokens": 18685980.0,
      "step": 10630
    },
    {
      "entropy": 5.095791339874268,
      "epoch": 8.334639498432601,
      "grad_norm": 1.3828125,
      "learning_rate": 8.5058457904505e-05,
      "loss": 4.2514,
      "mean_token_accuracy": 0.25136866569519045,
      "num_tokens": 18694834.0,
      "step": 10635
    },
    {
      "entropy": 5.04308705329895,
      "epoch": 8.338557993730408,
      "grad_norm": 1.4296875,
      "learning_rate": 8.489738802084742e-05,
      "loss": 4.198,
      "mean_token_accuracy": 0.24855799973011017,
      "num_tokens": 18703747.0,
      "step": 10640
    },
    {
      "entropy": 5.056109952926636,
      "epoch": 8.342476489028213,
      "grad_norm": 1.421875,
      "learning_rate": 8.473665788138112e-05,
      "loss": 4.1794,
      "mean_token_accuracy": 0.25656205266714094,
      "num_tokens": 18712202.0,
      "step": 10645
    },
    {
      "entropy": 5.063709497451782,
      "epoch": 8.346394984326018,
      "grad_norm": 1.40625,
      "learning_rate": 8.457626777335691e-05,
      "loss": 4.2123,
      "mean_token_accuracy": 0.25497734993696214,
      "num_tokens": 18721340.0,
      "step": 10650
    },
    {
      "entropy": 5.064907169342041,
      "epoch": 8.350313479623825,
      "grad_norm": 1.4296875,
      "learning_rate": 8.4416217983418e-05,
      "loss": 4.2629,
      "mean_token_accuracy": 0.2470559298992157,
      "num_tokens": 18730295.0,
      "step": 10655
    },
    {
      "entropy": 5.060745286941528,
      "epoch": 8.35423197492163,
      "grad_norm": 1.3984375,
      "learning_rate": 8.425650879759923e-05,
      "loss": 4.1993,
      "mean_token_accuracy": 0.2572849467396736,
      "num_tokens": 18739040.0,
      "step": 10660
    },
    {
      "entropy": 4.9909460067749025,
      "epoch": 8.358150470219435,
      "grad_norm": 1.359375,
      "learning_rate": 8.409714050132683e-05,
      "loss": 4.1235,
      "mean_token_accuracy": 0.26936892569065096,
      "num_tokens": 18747655.0,
      "step": 10665
    },
    {
      "entropy": 5.074930000305176,
      "epoch": 8.362068965517242,
      "grad_norm": 1.5078125,
      "learning_rate": 8.393811337941781e-05,
      "loss": 4.1751,
      "mean_token_accuracy": 0.2627860069274902,
      "num_tokens": 18755602.0,
      "step": 10670
    },
    {
      "entropy": 5.055441522598267,
      "epoch": 8.365987460815047,
      "grad_norm": 1.4453125,
      "learning_rate": 8.377942771607943e-05,
      "loss": 4.1774,
      "mean_token_accuracy": 0.2644303172826767,
      "num_tokens": 18764332.0,
      "step": 10675
    },
    {
      "entropy": 5.035217189788819,
      "epoch": 8.369905956112852,
      "grad_norm": 1.3046875,
      "learning_rate": 8.362108379490868e-05,
      "loss": 4.2195,
      "mean_token_accuracy": 0.25301694571971894,
      "num_tokens": 18773106.0,
      "step": 10680
    },
    {
      "entropy": 5.06776704788208,
      "epoch": 8.373824451410659,
      "grad_norm": 1.3984375,
      "learning_rate": 8.346308189889175e-05,
      "loss": 4.2437,
      "mean_token_accuracy": 0.24953767359256745,
      "num_tokens": 18781603.0,
      "step": 10685
    },
    {
      "entropy": 5.0347637176513675,
      "epoch": 8.377742946708464,
      "grad_norm": 1.4296875,
      "learning_rate": 8.330542231040379e-05,
      "loss": 4.1665,
      "mean_token_accuracy": 0.260590472817421,
      "num_tokens": 18790152.0,
      "step": 10690
    },
    {
      "entropy": 5.0837737083435055,
      "epoch": 8.38166144200627,
      "grad_norm": 1.4140625,
      "learning_rate": 8.314810531120795e-05,
      "loss": 4.1471,
      "mean_token_accuracy": 0.2661029189825058,
      "num_tokens": 18798397.0,
      "step": 10695
    },
    {
      "entropy": 5.062899971008301,
      "epoch": 8.385579937304076,
      "grad_norm": 1.4453125,
      "learning_rate": 8.299113118245519e-05,
      "loss": 4.1985,
      "mean_token_accuracy": 0.25839450806379316,
      "num_tokens": 18807523.0,
      "step": 10700
    },
    {
      "entropy": 5.069334077835083,
      "epoch": 8.389498432601881,
      "grad_norm": 1.3515625,
      "learning_rate": 8.283450020468377e-05,
      "loss": 4.2187,
      "mean_token_accuracy": 0.2502423897385597,
      "num_tokens": 18816469.0,
      "step": 10705
    },
    {
      "entropy": 5.13216986656189,
      "epoch": 8.393416927899686,
      "grad_norm": 1.453125,
      "learning_rate": 8.267821265781875e-05,
      "loss": 4.2292,
      "mean_token_accuracy": 0.2532289296388626,
      "num_tokens": 18823994.0,
      "step": 10710
    },
    {
      "entropy": 5.025593185424805,
      "epoch": 8.397335423197493,
      "grad_norm": 1.3515625,
      "learning_rate": 8.252226882117118e-05,
      "loss": 4.1366,
      "mean_token_accuracy": 0.26414495408535005,
      "num_tokens": 18833159.0,
      "step": 10715
    },
    {
      "entropy": 5.036218643188477,
      "epoch": 8.401253918495298,
      "grad_norm": 1.375,
      "learning_rate": 8.236666897343797e-05,
      "loss": 4.1212,
      "mean_token_accuracy": 0.26260329484939576,
      "num_tokens": 18841754.0,
      "step": 10720
    },
    {
      "entropy": 5.042536926269531,
      "epoch": 8.405172413793103,
      "grad_norm": 1.3984375,
      "learning_rate": 8.221141339270138e-05,
      "loss": 4.1858,
      "mean_token_accuracy": 0.2603021219372749,
      "num_tokens": 18850493.0,
      "step": 10725
    },
    {
      "entropy": 5.1154132843017575,
      "epoch": 8.409090909090908,
      "grad_norm": 1.4296875,
      "learning_rate": 8.205650235642828e-05,
      "loss": 4.2758,
      "mean_token_accuracy": 0.25167289674282073,
      "num_tokens": 18859765.0,
      "step": 10730
    },
    {
      "entropy": 5.072766590118408,
      "epoch": 8.413009404388715,
      "grad_norm": 1.4609375,
      "learning_rate": 8.190193614146971e-05,
      "loss": 4.248,
      "mean_token_accuracy": 0.2539701029658318,
      "num_tokens": 18868685.0,
      "step": 10735
    },
    {
      "entropy": 5.033316612243652,
      "epoch": 8.41692789968652,
      "grad_norm": 1.3671875,
      "learning_rate": 8.17477150240607e-05,
      "loss": 4.1788,
      "mean_token_accuracy": 0.25887560844421387,
      "num_tokens": 18877496.0,
      "step": 10740
    },
    {
      "entropy": 5.066567707061767,
      "epoch": 8.420846394984325,
      "grad_norm": 1.3359375,
      "learning_rate": 8.159383927981931e-05,
      "loss": 4.1863,
      "mean_token_accuracy": 0.258779676258564,
      "num_tokens": 18885828.0,
      "step": 10745
    },
    {
      "entropy": 5.122095823287964,
      "epoch": 8.424764890282132,
      "grad_norm": 1.4609375,
      "learning_rate": 8.14403091837464e-05,
      "loss": 4.2835,
      "mean_token_accuracy": 0.24313966631889344,
      "num_tokens": 18894498.0,
      "step": 10750
    },
    {
      "entropy": 5.089343786239624,
      "epoch": 8.428683385579937,
      "grad_norm": 1.46875,
      "learning_rate": 8.128712501022526e-05,
      "loss": 4.2293,
      "mean_token_accuracy": 0.2561744347214699,
      "num_tokens": 18902996.0,
      "step": 10755
    },
    {
      "entropy": 5.039628887176514,
      "epoch": 8.432601880877742,
      "grad_norm": 1.4453125,
      "learning_rate": 8.11342870330208e-05,
      "loss": 4.1475,
      "mean_token_accuracy": 0.2654921978712082,
      "num_tokens": 18911731.0,
      "step": 10760
    },
    {
      "entropy": 5.037811231613159,
      "epoch": 8.43652037617555,
      "grad_norm": 1.4765625,
      "learning_rate": 8.098179552527921e-05,
      "loss": 4.1217,
      "mean_token_accuracy": 0.26806780844926836,
      "num_tokens": 18920912.0,
      "step": 10765
    },
    {
      "entropy": 5.03811206817627,
      "epoch": 8.440438871473354,
      "grad_norm": 1.484375,
      "learning_rate": 8.082965075952757e-05,
      "loss": 4.1538,
      "mean_token_accuracy": 0.26308228969573977,
      "num_tokens": 18930012.0,
      "step": 10770
    },
    {
      "entropy": 5.0760599136352536,
      "epoch": 8.44435736677116,
      "grad_norm": 1.515625,
      "learning_rate": 8.067785300767333e-05,
      "loss": 4.2835,
      "mean_token_accuracy": 0.24623428732156755,
      "num_tokens": 18939026.0,
      "step": 10775
    },
    {
      "entropy": 5.088501501083374,
      "epoch": 8.448275862068966,
      "grad_norm": 1.3359375,
      "learning_rate": 8.05264025410036e-05,
      "loss": 4.2232,
      "mean_token_accuracy": 0.2476661190390587,
      "num_tokens": 18948387.0,
      "step": 10780
    },
    {
      "entropy": 5.099605417251587,
      "epoch": 8.452194357366771,
      "grad_norm": 1.390625,
      "learning_rate": 8.037529963018488e-05,
      "loss": 4.2605,
      "mean_token_accuracy": 0.2507855802774429,
      "num_tokens": 18957234.0,
      "step": 10785
    },
    {
      "entropy": 5.077880954742431,
      "epoch": 8.456112852664576,
      "grad_norm": 1.6328125,
      "learning_rate": 8.022454454526272e-05,
      "loss": 4.1964,
      "mean_token_accuracy": 0.2589456051588058,
      "num_tokens": 18965501.0,
      "step": 10790
    },
    {
      "entropy": 5.070105791091919,
      "epoch": 8.460031347962383,
      "grad_norm": 1.4375,
      "learning_rate": 8.00741375556608e-05,
      "loss": 4.203,
      "mean_token_accuracy": 0.25352300703525543,
      "num_tokens": 18973225.0,
      "step": 10795
    },
    {
      "entropy": 5.091630220413208,
      "epoch": 8.463949843260188,
      "grad_norm": 1.484375,
      "learning_rate": 7.99240789301808e-05,
      "loss": 4.1997,
      "mean_token_accuracy": 0.2602038010954857,
      "num_tokens": 18982424.0,
      "step": 10800
    },
    {
      "entropy": 5.075785303115845,
      "epoch": 8.467868338557993,
      "grad_norm": 1.3046875,
      "learning_rate": 7.977436893700177e-05,
      "loss": 4.2528,
      "mean_token_accuracy": 0.25237204879522324,
      "num_tokens": 18991519.0,
      "step": 10805
    },
    {
      "entropy": 5.031460857391357,
      "epoch": 8.4717868338558,
      "grad_norm": 1.328125,
      "learning_rate": 7.962500784367985e-05,
      "loss": 4.1882,
      "mean_token_accuracy": 0.25793329030275347,
      "num_tokens": 19000498.0,
      "step": 10810
    },
    {
      "entropy": 5.114910364151001,
      "epoch": 8.475705329153605,
      "grad_norm": 1.375,
      "learning_rate": 7.947599591714746e-05,
      "loss": 4.278,
      "mean_token_accuracy": 0.25081587433815,
      "num_tokens": 19009640.0,
      "step": 10815
    },
    {
      "entropy": 5.039729166030884,
      "epoch": 8.47962382445141,
      "grad_norm": 1.3984375,
      "learning_rate": 7.932733342371306e-05,
      "loss": 4.254,
      "mean_token_accuracy": 0.24860153794288636,
      "num_tokens": 19018337.0,
      "step": 10820
    },
    {
      "entropy": 5.158927059173584,
      "epoch": 8.483542319749215,
      "grad_norm": 1.375,
      "learning_rate": 7.917902062906066e-05,
      "loss": 4.3038,
      "mean_token_accuracy": 0.2460125967860222,
      "num_tokens": 19027232.0,
      "step": 10825
    },
    {
      "entropy": 5.001966142654419,
      "epoch": 8.487460815047022,
      "grad_norm": 1.3828125,
      "learning_rate": 7.90310577982493e-05,
      "loss": 4.1665,
      "mean_token_accuracy": 0.2620976880192757,
      "num_tokens": 19036661.0,
      "step": 10830
    },
    {
      "entropy": 5.053235149383545,
      "epoch": 8.491379310344827,
      "grad_norm": 1.3984375,
      "learning_rate": 7.888344519571247e-05,
      "loss": 4.1749,
      "mean_token_accuracy": 0.26083250939846037,
      "num_tokens": 19045955.0,
      "step": 10835
    },
    {
      "entropy": 5.123163795471191,
      "epoch": 8.495297805642632,
      "grad_norm": 1.484375,
      "learning_rate": 7.873618308525797e-05,
      "loss": 4.276,
      "mean_token_accuracy": 0.24636686444282532,
      "num_tokens": 19054183.0,
      "step": 10840
    },
    {
      "entropy": 5.15804615020752,
      "epoch": 8.49921630094044,
      "grad_norm": 1.34375,
      "learning_rate": 7.858927173006694e-05,
      "loss": 4.217,
      "mean_token_accuracy": 0.25995326191186907,
      "num_tokens": 19063029.0,
      "step": 10845
    },
    {
      "entropy": 5.060181617736816,
      "epoch": 8.503134796238244,
      "grad_norm": 1.3671875,
      "learning_rate": 7.844271139269384e-05,
      "loss": 4.1637,
      "mean_token_accuracy": 0.26595467031002046,
      "num_tokens": 19071325.0,
      "step": 10850
    },
    {
      "entropy": 5.073323202133179,
      "epoch": 8.50705329153605,
      "grad_norm": 1.3828125,
      "learning_rate": 7.829650233506569e-05,
      "loss": 4.2623,
      "mean_token_accuracy": 0.2454324930906296,
      "num_tokens": 19080082.0,
      "step": 10855
    },
    {
      "entropy": 5.035569858551026,
      "epoch": 8.510971786833856,
      "grad_norm": 1.4375,
      "learning_rate": 7.815064481848188e-05,
      "loss": 4.178,
      "mean_token_accuracy": 0.2517479032278061,
      "num_tokens": 19089286.0,
      "step": 10860
    },
    {
      "entropy": 5.062751770019531,
      "epoch": 8.514890282131661,
      "grad_norm": 1.4375,
      "learning_rate": 7.800513910361343e-05,
      "loss": 4.1284,
      "mean_token_accuracy": 0.27001445442438127,
      "num_tokens": 19098397.0,
      "step": 10865
    },
    {
      "entropy": 5.1510657787323,
      "epoch": 8.518808777429467,
      "grad_norm": 1.40625,
      "learning_rate": 7.785998545050253e-05,
      "loss": 4.2985,
      "mean_token_accuracy": 0.24067392796278,
      "num_tokens": 19107939.0,
      "step": 10870
    },
    {
      "entropy": 5.094768524169922,
      "epoch": 8.522727272727273,
      "grad_norm": 1.3828125,
      "learning_rate": 7.771518411856243e-05,
      "loss": 4.2528,
      "mean_token_accuracy": 0.24678900539875032,
      "num_tokens": 19116541.0,
      "step": 10875
    },
    {
      "entropy": 5.149053430557251,
      "epoch": 8.526645768025078,
      "grad_norm": 1.4375,
      "learning_rate": 7.75707353665765e-05,
      "loss": 4.2543,
      "mean_token_accuracy": 0.25213380306959154,
      "num_tokens": 19124830.0,
      "step": 10880
    },
    {
      "entropy": 4.990253877639771,
      "epoch": 8.530564263322884,
      "grad_norm": 1.3671875,
      "learning_rate": 7.742663945269816e-05,
      "loss": 4.1163,
      "mean_token_accuracy": 0.2628975257277489,
      "num_tokens": 19133518.0,
      "step": 10885
    },
    {
      "entropy": 5.099353218078614,
      "epoch": 8.53448275862069,
      "grad_norm": 1.375,
      "learning_rate": 7.72828966344501e-05,
      "loss": 4.2771,
      "mean_token_accuracy": 0.25573421716690065,
      "num_tokens": 19142989.0,
      "step": 10890
    },
    {
      "entropy": 5.175001573562622,
      "epoch": 8.538401253918495,
      "grad_norm": 1.34375,
      "learning_rate": 7.713950716872411e-05,
      "loss": 4.3295,
      "mean_token_accuracy": 0.24999918788671494,
      "num_tokens": 19151603.0,
      "step": 10895
    },
    {
      "entropy": 5.043934774398804,
      "epoch": 8.5423197492163,
      "grad_norm": 1.6171875,
      "learning_rate": 7.699647131178038e-05,
      "loss": 4.1549,
      "mean_token_accuracy": 0.25867428034543993,
      "num_tokens": 19159949.0,
      "step": 10900
    },
    {
      "entropy": 5.070485305786133,
      "epoch": 8.546238244514107,
      "grad_norm": 1.3828125,
      "learning_rate": 7.685378931924713e-05,
      "loss": 4.2546,
      "mean_token_accuracy": 0.24836337715387344,
      "num_tokens": 19169594.0,
      "step": 10905
    },
    {
      "entropy": 5.081017971038818,
      "epoch": 8.550156739811912,
      "grad_norm": 1.46875,
      "learning_rate": 7.67114614461203e-05,
      "loss": 4.1597,
      "mean_token_accuracy": 0.2625373288989067,
      "num_tokens": 19177660.0,
      "step": 10910
    },
    {
      "entropy": 5.055023431777954,
      "epoch": 8.554075235109718,
      "grad_norm": 1.4609375,
      "learning_rate": 7.656948794676283e-05,
      "loss": 4.2141,
      "mean_token_accuracy": 0.2565560802817345,
      "num_tokens": 19185890.0,
      "step": 10915
    },
    {
      "entropy": 5.002670240402222,
      "epoch": 8.557993730407524,
      "grad_norm": 1.4453125,
      "learning_rate": 7.642786907490437e-05,
      "loss": 4.1604,
      "mean_token_accuracy": 0.258914715051651,
      "num_tokens": 19194820.0,
      "step": 10920
    },
    {
      "entropy": 5.054674291610718,
      "epoch": 8.56191222570533,
      "grad_norm": 1.4765625,
      "learning_rate": 7.628660508364083e-05,
      "loss": 4.2398,
      "mean_token_accuracy": 0.2522713214159012,
      "num_tokens": 19203887.0,
      "step": 10925
    },
    {
      "entropy": 5.1219793319702145,
      "epoch": 8.565830721003135,
      "grad_norm": 1.34375,
      "learning_rate": 7.614569622543391e-05,
      "loss": 4.3006,
      "mean_token_accuracy": 0.2443852663040161,
      "num_tokens": 19212540.0,
      "step": 10930
    },
    {
      "entropy": 5.066019344329834,
      "epoch": 8.56974921630094,
      "grad_norm": 1.4296875,
      "learning_rate": 7.600514275211054e-05,
      "loss": 4.1825,
      "mean_token_accuracy": 0.25951949805021285,
      "num_tokens": 19221838.0,
      "step": 10935
    },
    {
      "entropy": 4.976387071609497,
      "epoch": 8.573667711598747,
      "grad_norm": 1.4609375,
      "learning_rate": 7.58649449148625e-05,
      "loss": 4.1069,
      "mean_token_accuracy": 0.26618069112300874,
      "num_tokens": 19229538.0,
      "step": 10940
    },
    {
      "entropy": 5.007119846343994,
      "epoch": 8.577586206896552,
      "grad_norm": 1.3203125,
      "learning_rate": 7.572510296424615e-05,
      "loss": 4.083,
      "mean_token_accuracy": 0.262335304915905,
      "num_tokens": 19237977.0,
      "step": 10945
    },
    {
      "entropy": 5.05847020149231,
      "epoch": 8.581504702194357,
      "grad_norm": 1.4609375,
      "learning_rate": 7.558561715018174e-05,
      "loss": 4.1984,
      "mean_token_accuracy": 0.25098466873168945,
      "num_tokens": 19246801.0,
      "step": 10950
    },
    {
      "entropy": 5.046053218841553,
      "epoch": 8.585423197492164,
      "grad_norm": 1.3046875,
      "learning_rate": 7.544648772195293e-05,
      "loss": 4.2056,
      "mean_token_accuracy": 0.2600756615400314,
      "num_tokens": 19256271.0,
      "step": 10955
    },
    {
      "entropy": 5.095232057571411,
      "epoch": 8.589341692789969,
      "grad_norm": 1.4453125,
      "learning_rate": 7.530771492820673e-05,
      "loss": 4.1961,
      "mean_token_accuracy": 0.25063690096139907,
      "num_tokens": 19264210.0,
      "step": 10960
    },
    {
      "entropy": 5.024372196197509,
      "epoch": 8.593260188087774,
      "grad_norm": 1.3828125,
      "learning_rate": 7.516929901695249e-05,
      "loss": 4.1314,
      "mean_token_accuracy": 0.25847766250371934,
      "num_tokens": 19272700.0,
      "step": 10965
    },
    {
      "entropy": 5.090874099731446,
      "epoch": 8.59717868338558,
      "grad_norm": 1.4140625,
      "learning_rate": 7.503124023556192e-05,
      "loss": 4.2382,
      "mean_token_accuracy": 0.25347224324941636,
      "num_tokens": 19281674.0,
      "step": 10970
    },
    {
      "entropy": 5.0780009746551515,
      "epoch": 8.601097178683386,
      "grad_norm": 1.3671875,
      "learning_rate": 7.48935388307685e-05,
      "loss": 4.1716,
      "mean_token_accuracy": 0.2616412237286568,
      "num_tokens": 19290748.0,
      "step": 10975
    },
    {
      "entropy": 5.02712779045105,
      "epoch": 8.60501567398119,
      "grad_norm": 1.484375,
      "learning_rate": 7.47561950486669e-05,
      "loss": 4.094,
      "mean_token_accuracy": 0.26732946634292604,
      "num_tokens": 19299018.0,
      "step": 10980
    },
    {
      "entropy": 5.038415193557739,
      "epoch": 8.608934169278998,
      "grad_norm": 1.390625,
      "learning_rate": 7.461920913471276e-05,
      "loss": 4.1037,
      "mean_token_accuracy": 0.26519618928432465,
      "num_tokens": 19307915.0,
      "step": 10985
    },
    {
      "entropy": 5.063757848739624,
      "epoch": 8.612852664576803,
      "grad_norm": 1.3359375,
      "learning_rate": 7.448258133372208e-05,
      "loss": 4.1813,
      "mean_token_accuracy": 0.27096630036830904,
      "num_tokens": 19316358.0,
      "step": 10990
    },
    {
      "entropy": 5.085788869857788,
      "epoch": 8.616771159874608,
      "grad_norm": 1.421875,
      "learning_rate": 7.434631188987095e-05,
      "loss": 4.2763,
      "mean_token_accuracy": 0.250113508105278,
      "num_tokens": 19325329.0,
      "step": 10995
    },
    {
      "entropy": 5.092767953872681,
      "epoch": 8.620689655172415,
      "grad_norm": 1.421875,
      "learning_rate": 7.42104010466949e-05,
      "loss": 4.2273,
      "mean_token_accuracy": 0.25731480568647386,
      "num_tokens": 19333857.0,
      "step": 11000
    },
    {
      "epoch": 8.620689655172415,
      "eval_entropy": 4.992106920988985,
      "eval_loss": 5.620754718780518,
      "eval_mean_token_accuracy": 0.1755531839160032,
      "eval_num_tokens": 19333857.0,
      "eval_runtime": 2.8457,
      "eval_samples_per_second": 1448.483,
      "eval_steps_per_second": 181.324,
      "step": 11000
    }
  ],
  "logging_steps": 5,
  "max_steps": 12750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5991056775808e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}