{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.528492183560262,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691978168487548,
      "epoch": 0.005042864346949067,
      "grad_norm": 12.8125,
      "learning_rate": 2e-06,
      "loss": 10.8351,
      "mean_token_accuracy": 0.0,
      "num_tokens": 11445.0,
      "step": 5
    },
    {
      "entropy": 10.691911983489991,
      "epoch": 0.010085728693898134,
      "grad_norm": 13.1875,
      "learning_rate": 4.5e-06,
      "loss": 10.7658,
      "mean_token_accuracy": 0.0,
      "num_tokens": 22095.0,
      "step": 10
    },
    {
      "entropy": 10.691094779968262,
      "epoch": 0.015128593040847202,
      "grad_norm": 8.875,
      "learning_rate": 7e-06,
      "loss": 10.5389,
      "mean_token_accuracy": 0.01894580915104598,
      "num_tokens": 33847.0,
      "step": 15
    },
    {
      "entropy": 10.681381797790527,
      "epoch": 0.020171457387796268,
      "grad_norm": 5.8125,
      "learning_rate": 9.5e-06,
      "loss": 10.2545,
      "mean_token_accuracy": 0.036951894313097,
      "num_tokens": 44495.0,
      "step": 20
    },
    {
      "entropy": 10.639113235473634,
      "epoch": 0.025214321734745335,
      "grad_norm": 4.0,
      "learning_rate": 1.2e-05,
      "loss": 9.9884,
      "mean_token_accuracy": 0.038632451370358464,
      "num_tokens": 55298.0,
      "step": 25
    },
    {
      "entropy": 10.603860855102539,
      "epoch": 0.030257186081694403,
      "grad_norm": 3.21875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.8047,
      "mean_token_accuracy": 0.04054084848612547,
      "num_tokens": 66391.0,
      "step": 30
    },
    {
      "entropy": 10.591868877410889,
      "epoch": 0.03530005042864347,
      "grad_norm": 2.65625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.698,
      "mean_token_accuracy": 0.04207097329199314,
      "num_tokens": 77624.0,
      "step": 35
    },
    {
      "entropy": 10.584535121917725,
      "epoch": 0.040342914775592535,
      "grad_norm": 2.546875,
      "learning_rate": 1.95e-05,
      "loss": 9.678,
      "mean_token_accuracy": 0.044261034950613974,
      "num_tokens": 89426.0,
      "step": 40
    },
    {
      "entropy": 10.562712478637696,
      "epoch": 0.0453857791225416,
      "grad_norm": 2.4375,
      "learning_rate": 2.2e-05,
      "loss": 9.6315,
      "mean_token_accuracy": 0.04177976287901401,
      "num_tokens": 100252.0,
      "step": 45
    },
    {
      "entropy": 10.594598770141602,
      "epoch": 0.05042864346949067,
      "grad_norm": 2.5,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.6087,
      "mean_token_accuracy": 0.03483491390943527,
      "num_tokens": 110578.0,
      "step": 50
    },
    {
      "entropy": 10.596640396118165,
      "epoch": 0.05547150781643974,
      "grad_norm": 2.3125,
      "learning_rate": 2.7e-05,
      "loss": 9.5875,
      "mean_token_accuracy": 0.03704058732837438,
      "num_tokens": 121631.0,
      "step": 55
    },
    {
      "entropy": 10.56342658996582,
      "epoch": 0.060514372163388806,
      "grad_norm": 2.21875,
      "learning_rate": 2.95e-05,
      "loss": 9.466,
      "mean_token_accuracy": 0.05011895261704922,
      "num_tokens": 134254.0,
      "step": 60
    },
    {
      "entropy": 10.496850776672364,
      "epoch": 0.06555723651033787,
      "grad_norm": 2.0625,
      "learning_rate": 3.2e-05,
      "loss": 9.4511,
      "mean_token_accuracy": 0.04732503369450569,
      "num_tokens": 144928.0,
      "step": 65
    },
    {
      "entropy": 10.411289405822753,
      "epoch": 0.07060010085728693,
      "grad_norm": 2.234375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2913,
      "mean_token_accuracy": 0.04968489818274975,
      "num_tokens": 155518.0,
      "step": 70
    },
    {
      "entropy": 10.424620437622071,
      "epoch": 0.07564296520423601,
      "grad_norm": 2.359375,
      "learning_rate": 3.7e-05,
      "loss": 9.1933,
      "mean_token_accuracy": 0.05173242129385471,
      "num_tokens": 165850.0,
      "step": 75
    },
    {
      "entropy": 10.444954109191894,
      "epoch": 0.08068582955118507,
      "grad_norm": 2.265625,
      "learning_rate": 3.95e-05,
      "loss": 9.1766,
      "mean_token_accuracy": 0.04967899359762669,
      "num_tokens": 176787.0,
      "step": 80
    },
    {
      "entropy": 10.427998161315918,
      "epoch": 0.08572869389813415,
      "grad_norm": 2.015625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.0647,
      "mean_token_accuracy": 0.04877705723047256,
      "num_tokens": 188345.0,
      "step": 85
    },
    {
      "entropy": 10.456082916259765,
      "epoch": 0.0907715582450832,
      "grad_norm": 2.25,
      "learning_rate": 4.45e-05,
      "loss": 9.0037,
      "mean_token_accuracy": 0.04692690074443817,
      "num_tokens": 199315.0,
      "step": 90
    },
    {
      "entropy": 10.339052677154541,
      "epoch": 0.09581442259203228,
      "grad_norm": 2.15625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.8109,
      "mean_token_accuracy": 0.05837383419275284,
      "num_tokens": 209627.0,
      "step": 95
    },
    {
      "entropy": 10.241901874542236,
      "epoch": 0.10085728693898134,
      "grad_norm": 2.0625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.7101,
      "mean_token_accuracy": 0.0651729203760624,
      "num_tokens": 220872.0,
      "step": 100
    },
    {
      "entropy": 10.15722131729126,
      "epoch": 0.1059001512859304,
      "grad_norm": 2.1875,
      "learning_rate": 5.2e-05,
      "loss": 8.5657,
      "mean_token_accuracy": 0.06282355673611165,
      "num_tokens": 233272.0,
      "step": 105
    },
    {
      "entropy": 10.10981206893921,
      "epoch": 0.11094301563287948,
      "grad_norm": 2.140625,
      "learning_rate": 5.45e-05,
      "loss": 8.4933,
      "mean_token_accuracy": 0.06420289240777492,
      "num_tokens": 243529.0,
      "step": 110
    },
    {
      "entropy": 10.059538745880127,
      "epoch": 0.11598587997982854,
      "grad_norm": 1.734375,
      "learning_rate": 5.7e-05,
      "loss": 8.3948,
      "mean_token_accuracy": 0.05781678333878517,
      "num_tokens": 254942.0,
      "step": 115
    },
    {
      "entropy": 9.973904132843018,
      "epoch": 0.12102874432677761,
      "grad_norm": 1.7734375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.2972,
      "mean_token_accuracy": 0.06580851301550865,
      "num_tokens": 266552.0,
      "step": 120
    },
    {
      "entropy": 9.73906192779541,
      "epoch": 0.1260716086737267,
      "grad_norm": 1.5625,
      "learning_rate": 6.2e-05,
      "loss": 8.1497,
      "mean_token_accuracy": 0.06753928288817405,
      "num_tokens": 277906.0,
      "step": 125
    },
    {
      "entropy": 9.731296634674072,
      "epoch": 0.13111447302067575,
      "grad_norm": 1.53125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.0309,
      "mean_token_accuracy": 0.06970244199037552,
      "num_tokens": 289079.0,
      "step": 130
    },
    {
      "entropy": 9.517922592163085,
      "epoch": 0.1361573373676248,
      "grad_norm": 1.4140625,
      "learning_rate": 6.7e-05,
      "loss": 7.9818,
      "mean_token_accuracy": 0.06449979990720749,
      "num_tokens": 300107.0,
      "step": 135
    },
    {
      "entropy": 9.309736728668213,
      "epoch": 0.14120020171457387,
      "grad_norm": 1.640625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8019,
      "mean_token_accuracy": 0.06583718433976174,
      "num_tokens": 310779.0,
      "step": 140
    },
    {
      "entropy": 9.122577667236328,
      "epoch": 0.14624306606152296,
      "grad_norm": 1.359375,
      "learning_rate": 7.2e-05,
      "loss": 7.7857,
      "mean_token_accuracy": 0.0686340156942606,
      "num_tokens": 323478.0,
      "step": 145
    },
    {
      "entropy": 8.875781536102295,
      "epoch": 0.15128593040847202,
      "grad_norm": 1.3515625,
      "learning_rate": 7.45e-05,
      "loss": 7.662,
      "mean_token_accuracy": 0.07218315117061139,
      "num_tokens": 335255.0,
      "step": 150
    },
    {
      "entropy": 8.774222087860107,
      "epoch": 0.15632879475542108,
      "grad_norm": 1.40625,
      "learning_rate": 7.7e-05,
      "loss": 7.5823,
      "mean_token_accuracy": 0.07090610340237617,
      "num_tokens": 344824.0,
      "step": 155
    },
    {
      "entropy": 8.562890911102295,
      "epoch": 0.16137165910237014,
      "grad_norm": 1.34375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.5779,
      "mean_token_accuracy": 0.06917917579412461,
      "num_tokens": 356283.0,
      "step": 160
    },
    {
      "entropy": 8.52785816192627,
      "epoch": 0.1664145234493192,
      "grad_norm": 1.2265625,
      "learning_rate": 8.2e-05,
      "loss": 7.5443,
      "mean_token_accuracy": 0.07162331864237785,
      "num_tokens": 368098.0,
      "step": 165
    },
    {
      "entropy": 8.293674755096436,
      "epoch": 0.1714573877962683,
      "grad_norm": 1.3828125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.4885,
      "mean_token_accuracy": 0.07210201621055604,
      "num_tokens": 380948.0,
      "step": 170
    },
    {
      "entropy": 8.206411838531494,
      "epoch": 0.17650025214321735,
      "grad_norm": 1.6953125,
      "learning_rate": 8.7e-05,
      "loss": 7.4311,
      "mean_token_accuracy": 0.07011085003614426,
      "num_tokens": 390392.0,
      "step": 175
    },
    {
      "entropy": 8.1397864818573,
      "epoch": 0.1815431164901664,
      "grad_norm": 1.3125,
      "learning_rate": 8.95e-05,
      "loss": 7.4686,
      "mean_token_accuracy": 0.07223408669233322,
      "num_tokens": 400328.0,
      "step": 180
    },
    {
      "entropy": 8.180228519439698,
      "epoch": 0.18658598083711547,
      "grad_norm": 1.375,
      "learning_rate": 9.2e-05,
      "loss": 7.4529,
      "mean_token_accuracy": 0.07169274277985097,
      "num_tokens": 410909.0,
      "step": 185
    },
    {
      "entropy": 8.057634353637695,
      "epoch": 0.19162884518406456,
      "grad_norm": 1.2421875,
      "learning_rate": 9.45e-05,
      "loss": 7.4423,
      "mean_token_accuracy": 0.07097707688808441,
      "num_tokens": 422726.0,
      "step": 190
    },
    {
      "entropy": 7.975717973709107,
      "epoch": 0.19667170953101362,
      "grad_norm": 1.46875,
      "learning_rate": 9.7e-05,
      "loss": 7.3252,
      "mean_token_accuracy": 0.07745771333575249,
      "num_tokens": 433376.0,
      "step": 195
    },
    {
      "entropy": 7.932044219970703,
      "epoch": 0.20171457387796268,
      "grad_norm": 1.4375,
      "learning_rate": 9.95e-05,
      "loss": 7.4045,
      "mean_token_accuracy": 0.06969214789569378,
      "num_tokens": 443940.0,
      "step": 200
    },
    {
      "entropy": 7.979291915893555,
      "epoch": 0.20675743822491174,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000102,
      "loss": 7.2735,
      "mean_token_accuracy": 0.07643608674407006,
      "num_tokens": 455016.0,
      "step": 205
    },
    {
      "entropy": 7.8824797630310055,
      "epoch": 0.2118003025718608,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.3292,
      "mean_token_accuracy": 0.07622824087738991,
      "num_tokens": 466380.0,
      "step": 210
    },
    {
      "entropy": 7.831804847717285,
      "epoch": 0.2168431669188099,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000107,
      "loss": 7.1783,
      "mean_token_accuracy": 0.08013702630996704,
      "num_tokens": 476915.0,
      "step": 215
    },
    {
      "entropy": 7.80485315322876,
      "epoch": 0.22188603126575895,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001095,
      "loss": 7.2725,
      "mean_token_accuracy": 0.0824611522257328,
      "num_tokens": 488924.0,
      "step": 220
    },
    {
      "entropy": 7.809728908538818,
      "epoch": 0.22692889561270801,
      "grad_norm": 1.203125,
      "learning_rate": 0.000112,
      "loss": 7.2162,
      "mean_token_accuracy": 0.07684922367334365,
      "num_tokens": 498965.0,
      "step": 225
    },
    {
      "entropy": 7.819616842269897,
      "epoch": 0.23197175995965708,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001145,
      "loss": 7.235,
      "mean_token_accuracy": 0.0765653271228075,
      "num_tokens": 509755.0,
      "step": 230
    },
    {
      "entropy": 7.771869659423828,
      "epoch": 0.23701462430660616,
      "grad_norm": 1.34375,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.2042,
      "mean_token_accuracy": 0.0763448379933834,
      "num_tokens": 521221.0,
      "step": 235
    },
    {
      "entropy": 7.7396564960479735,
      "epoch": 0.24205748865355523,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.3172,
      "mean_token_accuracy": 0.07263238392770291,
      "num_tokens": 533595.0,
      "step": 240
    },
    {
      "entropy": 7.770624494552612,
      "epoch": 0.24710035300050429,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000122,
      "loss": 7.2608,
      "mean_token_accuracy": 0.07456091418862343,
      "num_tokens": 543672.0,
      "step": 245
    },
    {
      "entropy": 7.743118524551392,
      "epoch": 0.2521432173474534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001245,
      "loss": 7.127,
      "mean_token_accuracy": 0.0765512753278017,
      "num_tokens": 554906.0,
      "step": 250
    },
    {
      "entropy": 7.675862503051758,
      "epoch": 0.25718608169440244,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000127,
      "loss": 7.2259,
      "mean_token_accuracy": 0.0814048271626234,
      "num_tokens": 566419.0,
      "step": 255
    },
    {
      "entropy": 7.71265172958374,
      "epoch": 0.2622289460413515,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001295,
      "loss": 7.1691,
      "mean_token_accuracy": 0.07766851969063282,
      "num_tokens": 578848.0,
      "step": 260
    },
    {
      "entropy": 7.5628129005432125,
      "epoch": 0.26727181038830056,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000132,
      "loss": 7.1197,
      "mean_token_accuracy": 0.07973449081182479,
      "num_tokens": 590601.0,
      "step": 265
    },
    {
      "entropy": 7.681688404083252,
      "epoch": 0.2723146747352496,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.1774,
      "mean_token_accuracy": 0.07436439543962478,
      "num_tokens": 602021.0,
      "step": 270
    },
    {
      "entropy": 7.616954660415649,
      "epoch": 0.2773575390821987,
      "grad_norm": 1.15625,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.2583,
      "mean_token_accuracy": 0.07453522607684135,
      "num_tokens": 613966.0,
      "step": 275
    },
    {
      "entropy": 7.710917282104492,
      "epoch": 0.28240040342914774,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001395,
      "loss": 7.0984,
      "mean_token_accuracy": 0.07756517678499222,
      "num_tokens": 624851.0,
      "step": 280
    },
    {
      "entropy": 7.546781873703003,
      "epoch": 0.2874432677760968,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.1915,
      "mean_token_accuracy": 0.07566507011651993,
      "num_tokens": 635580.0,
      "step": 285
    },
    {
      "entropy": 7.662804460525512,
      "epoch": 0.2924861321230459,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001445,
      "loss": 7.2473,
      "mean_token_accuracy": 0.07517430521547794,
      "num_tokens": 646927.0,
      "step": 290
    },
    {
      "entropy": 7.587956047058105,
      "epoch": 0.297528996469995,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000147,
      "loss": 7.0551,
      "mean_token_accuracy": 0.08050617948174477,
      "num_tokens": 658116.0,
      "step": 295
    },
    {
      "entropy": 7.642880249023437,
      "epoch": 0.30257186081694404,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001495,
      "loss": 7.0584,
      "mean_token_accuracy": 0.07990356534719467,
      "num_tokens": 669054.0,
      "step": 300
    },
    {
      "entropy": 7.5717206478118895,
      "epoch": 0.3076147251638931,
      "grad_norm": 1.328125,
      "learning_rate": 0.000152,
      "loss": 7.1075,
      "mean_token_accuracy": 0.0826454609632492,
      "num_tokens": 680439.0,
      "step": 305
    },
    {
      "entropy": 7.453785753250122,
      "epoch": 0.31265758951084216,
      "grad_norm": 1.375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0661,
      "mean_token_accuracy": 0.08129210546612739,
      "num_tokens": 690909.0,
      "step": 310
    },
    {
      "entropy": 7.5011971950531,
      "epoch": 0.3177004538577912,
      "grad_norm": 1.25,
      "learning_rate": 0.000157,
      "loss": 7.048,
      "mean_token_accuracy": 0.07969367988407612,
      "num_tokens": 701744.0,
      "step": 315
    },
    {
      "entropy": 7.467720365524292,
      "epoch": 0.3227433182047403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001595,
      "loss": 7.0817,
      "mean_token_accuracy": 0.08535065315663815,
      "num_tokens": 712268.0,
      "step": 320
    },
    {
      "entropy": 7.59487681388855,
      "epoch": 0.32778618255168934,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000162,
      "loss": 7.1079,
      "mean_token_accuracy": 0.07844235375523567,
      "num_tokens": 722978.0,
      "step": 325
    },
    {
      "entropy": 7.482430076599121,
      "epoch": 0.3328290468986384,
      "grad_norm": 1.234375,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.1011,
      "mean_token_accuracy": 0.08162309601902962,
      "num_tokens": 734278.0,
      "step": 330
    },
    {
      "entropy": 7.5462571144104,
      "epoch": 0.3378719112455875,
      "grad_norm": 1.0625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.0141,
      "mean_token_accuracy": 0.0776023030281067,
      "num_tokens": 746109.0,
      "step": 335
    },
    {
      "entropy": 7.374982213973999,
      "epoch": 0.3429147755925366,
      "grad_norm": 1.46875,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.9356,
      "mean_token_accuracy": 0.08509711548686028,
      "num_tokens": 757519.0,
      "step": 340
    },
    {
      "entropy": 7.443374300003052,
      "epoch": 0.34795763993948564,
      "grad_norm": 1.421875,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9149,
      "mean_token_accuracy": 0.08727683797478676,
      "num_tokens": 767665.0,
      "step": 345
    },
    {
      "entropy": 7.435840845108032,
      "epoch": 0.3530005042864347,
      "grad_norm": 1.203125,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9537,
      "mean_token_accuracy": 0.08823386654257774,
      "num_tokens": 778880.0,
      "step": 350
    },
    {
      "entropy": 7.310178327560425,
      "epoch": 0.35804336863338376,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000177,
      "loss": 6.9559,
      "mean_token_accuracy": 0.09013235792517663,
      "num_tokens": 790360.0,
      "step": 355
    },
    {
      "entropy": 7.387903451919556,
      "epoch": 0.3630862329803328,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0001795,
      "loss": 6.9182,
      "mean_token_accuracy": 0.08709552362561226,
      "num_tokens": 800889.0,
      "step": 360
    },
    {
      "entropy": 7.419098663330078,
      "epoch": 0.3681290973272819,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000182,
      "loss": 6.9171,
      "mean_token_accuracy": 0.08571921288967133,
      "num_tokens": 811414.0,
      "step": 365
    },
    {
      "entropy": 7.240619897842407,
      "epoch": 0.37317196167423095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001845,
      "loss": 6.9305,
      "mean_token_accuracy": 0.0850242905318737,
      "num_tokens": 822822.0,
      "step": 370
    },
    {
      "entropy": 7.271744012832642,
      "epoch": 0.37821482602118,
      "grad_norm": 1.234375,
      "learning_rate": 0.000187,
      "loss": 6.829,
      "mean_token_accuracy": 0.09168825820088386,
      "num_tokens": 833145.0,
      "step": 375
    },
    {
      "entropy": 7.2786908626556395,
      "epoch": 0.3832576903681291,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001895,
      "loss": 6.9405,
      "mean_token_accuracy": 0.0836037501692772,
      "num_tokens": 844015.0,
      "step": 380
    },
    {
      "entropy": 7.349726581573487,
      "epoch": 0.3883005547150782,
      "grad_norm": 1.03125,
      "learning_rate": 0.000192,
      "loss": 6.8706,
      "mean_token_accuracy": 0.08842808604240418,
      "num_tokens": 855806.0,
      "step": 385
    },
    {
      "entropy": 7.209612941741943,
      "epoch": 0.39334341906202724,
      "grad_norm": 1.125,
      "learning_rate": 0.0001945,
      "loss": 6.7935,
      "mean_token_accuracy": 0.08768922984600067,
      "num_tokens": 865750.0,
      "step": 390
    },
    {
      "entropy": 7.334837818145752,
      "epoch": 0.3983862834089763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.9248,
      "mean_token_accuracy": 0.08573650568723679,
      "num_tokens": 877630.0,
      "step": 395
    },
    {
      "entropy": 7.157733392715454,
      "epoch": 0.40342914775592537,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8054,
      "mean_token_accuracy": 0.09393974989652634,
      "num_tokens": 888667.0,
      "step": 400
    },
    {
      "entropy": 7.3036473274230955,
      "epoch": 0.4084720121028744,
      "grad_norm": 1.09375,
      "learning_rate": 0.000202,
      "loss": 6.8838,
      "mean_token_accuracy": 0.08392557948827743,
      "num_tokens": 900989.0,
      "step": 405
    },
    {
      "entropy": 7.183315801620483,
      "epoch": 0.4135148764498235,
      "grad_norm": 1.125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7766,
      "mean_token_accuracy": 0.08522050082683563,
      "num_tokens": 911479.0,
      "step": 410
    },
    {
      "entropy": 7.263188695907592,
      "epoch": 0.41855774079677255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000207,
      "loss": 6.7898,
      "mean_token_accuracy": 0.08779488950967788,
      "num_tokens": 922434.0,
      "step": 415
    },
    {
      "entropy": 7.146833324432373,
      "epoch": 0.4236006051437216,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002095,
      "loss": 6.7957,
      "mean_token_accuracy": 0.08833592608571053,
      "num_tokens": 933405.0,
      "step": 420
    },
    {
      "entropy": 7.173463964462281,
      "epoch": 0.4286434694906707,
      "grad_norm": 1.328125,
      "learning_rate": 0.000212,
      "loss": 6.7996,
      "mean_token_accuracy": 0.0885507345199585,
      "num_tokens": 943567.0,
      "step": 425
    },
    {
      "entropy": 7.099128866195679,
      "epoch": 0.4336863338376198,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002145,
      "loss": 6.8177,
      "mean_token_accuracy": 0.08491698876023293,
      "num_tokens": 955123.0,
      "step": 430
    },
    {
      "entropy": 7.277088403701782,
      "epoch": 0.43872919818456885,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.8615,
      "mean_token_accuracy": 0.08992773070931434,
      "num_tokens": 965565.0,
      "step": 435
    },
    {
      "entropy": 7.159026861190796,
      "epoch": 0.4437720625315179,
      "grad_norm": 1.125,
      "learning_rate": 0.0002195,
      "loss": 6.78,
      "mean_token_accuracy": 0.08791229128837585,
      "num_tokens": 977101.0,
      "step": 440
    },
    {
      "entropy": 7.1419289112091064,
      "epoch": 0.44881492687846697,
      "grad_norm": 1.078125,
      "learning_rate": 0.000222,
      "loss": 6.8314,
      "mean_token_accuracy": 0.08858354166150093,
      "num_tokens": 987445.0,
      "step": 445
    },
    {
      "entropy": 7.118024682998657,
      "epoch": 0.45385779122541603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002245,
      "loss": 6.7034,
      "mean_token_accuracy": 0.09504426047205924,
      "num_tokens": 999379.0,
      "step": 450
    },
    {
      "entropy": 6.998180723190307,
      "epoch": 0.4589006555723651,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6412,
      "mean_token_accuracy": 0.09876157343387604,
      "num_tokens": 1010744.0,
      "step": 455
    },
    {
      "entropy": 7.0912974834442135,
      "epoch": 0.46394351991931415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7395,
      "mean_token_accuracy": 0.0892062947154045,
      "num_tokens": 1021808.0,
      "step": 460
    },
    {
      "entropy": 7.138762521743774,
      "epoch": 0.4689863842662632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.787,
      "mean_token_accuracy": 0.0881411112844944,
      "num_tokens": 1033694.0,
      "step": 465
    },
    {
      "entropy": 7.05295991897583,
      "epoch": 0.47402924861321233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.7531,
      "mean_token_accuracy": 0.09112738594412803,
      "num_tokens": 1044690.0,
      "step": 470
    },
    {
      "entropy": 7.110706329345703,
      "epoch": 0.4790721129601614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000237,
      "loss": 6.7543,
      "mean_token_accuracy": 0.0913363330066204,
      "num_tokens": 1055557.0,
      "step": 475
    },
    {
      "entropy": 7.020756387710572,
      "epoch": 0.48411497730711045,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002395,
      "loss": 6.6926,
      "mean_token_accuracy": 0.09522736445069313,
      "num_tokens": 1067304.0,
      "step": 480
    },
    {
      "entropy": 7.031440687179566,
      "epoch": 0.4891578416540595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000242,
      "loss": 6.7153,
      "mean_token_accuracy": 0.09247291311621667,
      "num_tokens": 1078955.0,
      "step": 485
    },
    {
      "entropy": 7.019560098648071,
      "epoch": 0.49420070600100857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002445,
      "loss": 6.7608,
      "mean_token_accuracy": 0.09506335482001305,
      "num_tokens": 1091405.0,
      "step": 490
    },
    {
      "entropy": 7.067246627807617,
      "epoch": 0.49924357034795763,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000247,
      "loss": 6.7344,
      "mean_token_accuracy": 0.09412561133503913,
      "num_tokens": 1103227.0,
      "step": 495
    },
    {
      "entropy": 7.013670969009399,
      "epoch": 0.5042864346949067,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002495,
      "loss": 6.6734,
      "mean_token_accuracy": 0.09890775382518768,
      "num_tokens": 1114485.0,
      "step": 500
    },
    {
      "epoch": 0.5042864346949067,
      "eval_entropy": 6.978830953894074,
      "eval_loss": 6.783904552459717,
      "eval_mean_token_accuracy": 0.09532982510995622,
      "eval_num_tokens": 1114485.0,
      "eval_runtime": 2.7002,
      "eval_samples_per_second": 1162.13,
      "eval_steps_per_second": 145.544,
      "step": 500
    },
    {
      "entropy": 7.006657695770263,
      "epoch": 0.5093292990418558,
      "grad_norm": 1.28125,
      "learning_rate": 0.000252,
      "loss": 6.6997,
      "mean_token_accuracy": 0.0941694438457489,
      "num_tokens": 1126789.0,
      "step": 505
    },
    {
      "entropy": 7.028399562835693,
      "epoch": 0.5143721633888049,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002545,
      "loss": 6.6884,
      "mean_token_accuracy": 0.09093117266893387,
      "num_tokens": 1138192.0,
      "step": 510
    },
    {
      "entropy": 6.900946187973022,
      "epoch": 0.5194150277357539,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000257,
      "loss": 6.6217,
      "mean_token_accuracy": 0.09346246346831322,
      "num_tokens": 1148526.0,
      "step": 515
    },
    {
      "entropy": 7.066805505752564,
      "epoch": 0.524457892082703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002595,
      "loss": 6.7189,
      "mean_token_accuracy": 0.08981634229421616,
      "num_tokens": 1159194.0,
      "step": 520
    },
    {
      "entropy": 6.8563981533050535,
      "epoch": 0.529500756429652,
      "grad_norm": 1.109375,
      "learning_rate": 0.000262,
      "loss": 6.6209,
      "mean_token_accuracy": 0.09972566738724709,
      "num_tokens": 1170601.0,
      "step": 525
    },
    {
      "entropy": 7.004376411437988,
      "epoch": 0.5345436207766011,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6214,
      "mean_token_accuracy": 0.09583464860916138,
      "num_tokens": 1182441.0,
      "step": 530
    },
    {
      "entropy": 6.822463846206665,
      "epoch": 0.5395864851235502,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.535,
      "mean_token_accuracy": 0.10155294314026833,
      "num_tokens": 1192530.0,
      "step": 535
    },
    {
      "entropy": 6.8695042610168455,
      "epoch": 0.5446293494704992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6516,
      "mean_token_accuracy": 0.09302628561854362,
      "num_tokens": 1203831.0,
      "step": 540
    },
    {
      "entropy": 6.884985733032226,
      "epoch": 0.5496722138174484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.5849,
      "mean_token_accuracy": 0.09827456548810005,
      "num_tokens": 1214164.0,
      "step": 545
    },
    {
      "entropy": 6.899946403503418,
      "epoch": 0.5547150781643974,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002745,
      "loss": 6.6103,
      "mean_token_accuracy": 0.09503386095166207,
      "num_tokens": 1226932.0,
      "step": 550
    },
    {
      "entropy": 7.018468141555786,
      "epoch": 0.5597579425113465,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000277,
      "loss": 6.6799,
      "mean_token_accuracy": 0.09099587425589561,
      "num_tokens": 1237212.0,
      "step": 555
    },
    {
      "entropy": 6.843923902511596,
      "epoch": 0.5648008068582955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002795,
      "loss": 6.5802,
      "mean_token_accuracy": 0.095969757437706,
      "num_tokens": 1249340.0,
      "step": 560
    },
    {
      "entropy": 6.853866147994995,
      "epoch": 0.5698436712052446,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5839,
      "mean_token_accuracy": 0.0990279607474804,
      "num_tokens": 1260778.0,
      "step": 565
    },
    {
      "entropy": 6.846533918380738,
      "epoch": 0.5748865355521936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002845,
      "loss": 6.5885,
      "mean_token_accuracy": 0.0959065780043602,
      "num_tokens": 1272216.0,
      "step": 570
    },
    {
      "entropy": 6.885388469696045,
      "epoch": 0.5799293998991427,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000287,
      "loss": 6.6706,
      "mean_token_accuracy": 0.09878646060824395,
      "num_tokens": 1283558.0,
      "step": 575
    },
    {
      "entropy": 6.949707126617431,
      "epoch": 0.5849722642460918,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002895,
      "loss": 6.729,
      "mean_token_accuracy": 0.09222633503377438,
      "num_tokens": 1294638.0,
      "step": 580
    },
    {
      "entropy": 6.937385559082031,
      "epoch": 0.5900151285930408,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000292,
      "loss": 6.6493,
      "mean_token_accuracy": 0.09726304411888123,
      "num_tokens": 1304697.0,
      "step": 585
    },
    {
      "entropy": 6.841894626617432,
      "epoch": 0.59505799293999,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002945,
      "loss": 6.6087,
      "mean_token_accuracy": 0.09355146735906601,
      "num_tokens": 1316618.0,
      "step": 590
    },
    {
      "entropy": 6.907530450820923,
      "epoch": 0.600100857286939,
      "grad_norm": 1.109375,
      "learning_rate": 0.000297,
      "loss": 6.7174,
      "mean_token_accuracy": 0.08738479986786843,
      "num_tokens": 1328134.0,
      "step": 595
    },
    {
      "entropy": 6.918644523620605,
      "epoch": 0.6051437216338881,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002995,
      "loss": 6.6175,
      "mean_token_accuracy": 0.09422577098011971,
      "num_tokens": 1339101.0,
      "step": 600
    },
    {
      "entropy": 6.793248319625855,
      "epoch": 0.6101865859808371,
      "grad_norm": 1.109375,
      "learning_rate": 0.000302,
      "loss": 6.4922,
      "mean_token_accuracy": 0.09907795786857605,
      "num_tokens": 1349181.0,
      "step": 605
    },
    {
      "entropy": 6.85523247718811,
      "epoch": 0.6152294503277862,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0003045,
      "loss": 6.6357,
      "mean_token_accuracy": 0.10059619322419167,
      "num_tokens": 1361443.0,
      "step": 610
    },
    {
      "entropy": 6.753997278213501,
      "epoch": 0.6202723146747352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000307,
      "loss": 6.5688,
      "mean_token_accuracy": 0.10217763558030128,
      "num_tokens": 1372807.0,
      "step": 615
    },
    {
      "entropy": 6.8447350025177,
      "epoch": 0.6253151790216843,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003095,
      "loss": 6.6514,
      "mean_token_accuracy": 0.09630123376846314,
      "num_tokens": 1385509.0,
      "step": 620
    },
    {
      "entropy": 6.80165662765503,
      "epoch": 0.6303580433686334,
      "grad_norm": 1.234375,
      "learning_rate": 0.000312,
      "loss": 6.5447,
      "mean_token_accuracy": 0.09622347950935364,
      "num_tokens": 1396035.0,
      "step": 625
    },
    {
      "entropy": 6.750753402709961,
      "epoch": 0.6354009077155824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003145,
      "loss": 6.5012,
      "mean_token_accuracy": 0.10519321039319038,
      "num_tokens": 1406839.0,
      "step": 630
    },
    {
      "entropy": 6.73918046951294,
      "epoch": 0.6404437720625316,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000317,
      "loss": 6.4651,
      "mean_token_accuracy": 0.09995857030153274,
      "num_tokens": 1417708.0,
      "step": 635
    },
    {
      "entropy": 6.814885568618775,
      "epoch": 0.6454866364094806,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003195,
      "loss": 6.5956,
      "mean_token_accuracy": 0.09352405592799187,
      "num_tokens": 1428985.0,
      "step": 640
    },
    {
      "entropy": 6.812655353546143,
      "epoch": 0.6505295007564297,
      "grad_norm": 1.21875,
      "learning_rate": 0.000322,
      "loss": 6.6272,
      "mean_token_accuracy": 0.10007955580949783,
      "num_tokens": 1440232.0,
      "step": 645
    },
    {
      "entropy": 6.6787721633911135,
      "epoch": 0.6555723651033787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.3639,
      "mean_token_accuracy": 0.10900432094931603,
      "num_tokens": 1451735.0,
      "step": 650
    },
    {
      "entropy": 6.713140487670898,
      "epoch": 0.6606152294503278,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5218,
      "mean_token_accuracy": 0.09929425492882729,
      "num_tokens": 1463477.0,
      "step": 655
    },
    {
      "entropy": 6.6867029666900635,
      "epoch": 0.6656580937972768,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5508,
      "mean_token_accuracy": 0.09417604357004165,
      "num_tokens": 1473802.0,
      "step": 660
    },
    {
      "entropy": 6.759989881515503,
      "epoch": 0.6707009581442259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5538,
      "mean_token_accuracy": 0.09805640652775764,
      "num_tokens": 1485519.0,
      "step": 665
    },
    {
      "entropy": 6.729797887802124,
      "epoch": 0.675743822491175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5546,
      "mean_token_accuracy": 0.09775246679782867,
      "num_tokens": 1496957.0,
      "step": 670
    },
    {
      "entropy": 6.685538005828858,
      "epoch": 0.680786686838124,
      "grad_norm": 1.203125,
      "learning_rate": 0.000337,
      "loss": 6.5624,
      "mean_token_accuracy": 0.1045648567378521,
      "num_tokens": 1508336.0,
      "step": 675
    },
    {
      "entropy": 6.67319016456604,
      "epoch": 0.6858295511850732,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003395,
      "loss": 6.5527,
      "mean_token_accuracy": 0.10019075870513916,
      "num_tokens": 1520016.0,
      "step": 680
    },
    {
      "entropy": 6.664720964431763,
      "epoch": 0.6908724155320222,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342,
      "loss": 6.3887,
      "mean_token_accuracy": 0.10588083937764167,
      "num_tokens": 1531877.0,
      "step": 685
    },
    {
      "entropy": 6.550206899642944,
      "epoch": 0.6959152798789713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.3834,
      "mean_token_accuracy": 0.1071600653231144,
      "num_tokens": 1541553.0,
      "step": 690
    },
    {
      "entropy": 6.740174055099487,
      "epoch": 0.7009581442259203,
      "grad_norm": 1.03125,
      "learning_rate": 0.000347,
      "loss": 6.5491,
      "mean_token_accuracy": 0.10118654146790504,
      "num_tokens": 1552491.0,
      "step": 695
    },
    {
      "entropy": 6.554775905609131,
      "epoch": 0.7060010085728694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003495,
      "loss": 6.4546,
      "mean_token_accuracy": 0.10349184796214103,
      "num_tokens": 1563695.0,
      "step": 700
    },
    {
      "entropy": 6.716359710693359,
      "epoch": 0.7110438729198184,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000352,
      "loss": 6.4772,
      "mean_token_accuracy": 0.10427189692854881,
      "num_tokens": 1573467.0,
      "step": 705
    },
    {
      "entropy": 6.579839134216309,
      "epoch": 0.7160867372667675,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003545,
      "loss": 6.4774,
      "mean_token_accuracy": 0.10680482760071755,
      "num_tokens": 1584367.0,
      "step": 710
    },
    {
      "entropy": 6.704503059387207,
      "epoch": 0.7211296016137166,
      "grad_norm": 1.0625,
      "learning_rate": 0.000357,
      "loss": 6.4049,
      "mean_token_accuracy": 0.11087890490889549,
      "num_tokens": 1595301.0,
      "step": 715
    },
    {
      "entropy": 6.5911847114562985,
      "epoch": 0.7261724659606656,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003595,
      "loss": 6.4628,
      "mean_token_accuracy": 0.09945516511797906,
      "num_tokens": 1605916.0,
      "step": 720
    },
    {
      "entropy": 6.6797267436981205,
      "epoch": 0.7312153303076148,
      "grad_norm": 1.0,
      "learning_rate": 0.000362,
      "loss": 6.4645,
      "mean_token_accuracy": 0.10250937268137932,
      "num_tokens": 1616455.0,
      "step": 725
    },
    {
      "entropy": 6.535722160339356,
      "epoch": 0.7362581946545638,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003645,
      "loss": 6.4454,
      "mean_token_accuracy": 0.1064433366060257,
      "num_tokens": 1627396.0,
      "step": 730
    },
    {
      "entropy": 6.790402030944824,
      "epoch": 0.7413010590015129,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000367,
      "loss": 6.5785,
      "mean_token_accuracy": 0.09991721957921981,
      "num_tokens": 1639073.0,
      "step": 735
    },
    {
      "entropy": 6.5861907482147215,
      "epoch": 0.7463439233484619,
      "grad_norm": 1.125,
      "learning_rate": 0.0003695,
      "loss": 6.4448,
      "mean_token_accuracy": 0.10222182348370552,
      "num_tokens": 1649787.0,
      "step": 740
    },
    {
      "entropy": 6.561935520172119,
      "epoch": 0.751386787695411,
      "grad_norm": 0.96875,
      "learning_rate": 0.000372,
      "loss": 6.3867,
      "mean_token_accuracy": 0.10436921566724777,
      "num_tokens": 1658883.0,
      "step": 745
    },
    {
      "entropy": 6.535987997055054,
      "epoch": 0.75642965204236,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003745,
      "loss": 6.3549,
      "mean_token_accuracy": 0.1072971761226654,
      "num_tokens": 1669099.0,
      "step": 750
    },
    {
      "entropy": 6.504691028594971,
      "epoch": 0.7614725163893091,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000377,
      "loss": 6.373,
      "mean_token_accuracy": 0.10971354767680168,
      "num_tokens": 1680227.0,
      "step": 755
    },
    {
      "entropy": 6.545934200286865,
      "epoch": 0.7665153807362582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003795,
      "loss": 6.3708,
      "mean_token_accuracy": 0.10627648830413819,
      "num_tokens": 1689901.0,
      "step": 760
    },
    {
      "entropy": 6.492063522338867,
      "epoch": 0.7715582450832073,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000382,
      "loss": 6.4469,
      "mean_token_accuracy": 0.10555314868688584,
      "num_tokens": 1700498.0,
      "step": 765
    },
    {
      "entropy": 6.565538930892944,
      "epoch": 0.7766011094301564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003845,
      "loss": 6.3697,
      "mean_token_accuracy": 0.10695412680506707,
      "num_tokens": 1710715.0,
      "step": 770
    },
    {
      "entropy": 6.589845371246338,
      "epoch": 0.7816439737771054,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4913,
      "mean_token_accuracy": 0.1047224260866642,
      "num_tokens": 1723858.0,
      "step": 775
    },
    {
      "entropy": 6.568187999725342,
      "epoch": 0.7866868381240545,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4495,
      "mean_token_accuracy": 0.09719336107373237,
      "num_tokens": 1734678.0,
      "step": 780
    },
    {
      "entropy": 6.497055625915527,
      "epoch": 0.7917297024710035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3665,
      "mean_token_accuracy": 0.1123119130730629,
      "num_tokens": 1743869.0,
      "step": 785
    },
    {
      "entropy": 6.632794427871704,
      "epoch": 0.7967725668179526,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4189,
      "mean_token_accuracy": 0.1059940479695797,
      "num_tokens": 1755266.0,
      "step": 790
    },
    {
      "entropy": 6.40299015045166,
      "epoch": 0.8018154311649016,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.3884,
      "mean_token_accuracy": 0.10287827029824256,
      "num_tokens": 1766089.0,
      "step": 795
    },
    {
      "entropy": 6.549517488479614,
      "epoch": 0.8068582955118507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003995,
      "loss": 6.3782,
      "mean_token_accuracy": 0.10602757111191749,
      "num_tokens": 1776057.0,
      "step": 800
    },
    {
      "entropy": 6.550864315032959,
      "epoch": 0.8119011598587998,
      "grad_norm": 1.046875,
      "learning_rate": 0.000402,
      "loss": 6.4463,
      "mean_token_accuracy": 0.10098104774951935,
      "num_tokens": 1786729.0,
      "step": 805
    },
    {
      "entropy": 6.510504293441772,
      "epoch": 0.8169440242057489,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004045,
      "loss": 6.3717,
      "mean_token_accuracy": 0.10247773975133896,
      "num_tokens": 1797708.0,
      "step": 810
    },
    {
      "entropy": 6.48493595123291,
      "epoch": 0.821986888552698,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4006,
      "mean_token_accuracy": 0.11212195381522179,
      "num_tokens": 1807326.0,
      "step": 815
    },
    {
      "entropy": 6.447803640365601,
      "epoch": 0.827029752899647,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004095,
      "loss": 6.3169,
      "mean_token_accuracy": 0.11085391044616699,
      "num_tokens": 1817266.0,
      "step": 820
    },
    {
      "entropy": 6.538650226593018,
      "epoch": 0.8320726172465961,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000412,
      "loss": 6.3561,
      "mean_token_accuracy": 0.11818044185638428,
      "num_tokens": 1827840.0,
      "step": 825
    },
    {
      "entropy": 6.338764762878418,
      "epoch": 0.8371154815935451,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004145,
      "loss": 6.3387,
      "mean_token_accuracy": 0.11126410290598869,
      "num_tokens": 1838158.0,
      "step": 830
    },
    {
      "entropy": 6.599035263061523,
      "epoch": 0.8421583459404942,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000417,
      "loss": 6.4225,
      "mean_token_accuracy": 0.10363714173436164,
      "num_tokens": 1848317.0,
      "step": 835
    },
    {
      "entropy": 6.398059988021851,
      "epoch": 0.8472012102874432,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004195,
      "loss": 6.3394,
      "mean_token_accuracy": 0.10549739152193069,
      "num_tokens": 1858751.0,
      "step": 840
    },
    {
      "entropy": 6.4699845790863035,
      "epoch": 0.8522440746343923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000422,
      "loss": 6.365,
      "mean_token_accuracy": 0.1064521923661232,
      "num_tokens": 1870739.0,
      "step": 845
    },
    {
      "entropy": 6.610909032821655,
      "epoch": 0.8572869389813415,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004245,
      "loss": 6.4852,
      "mean_token_accuracy": 0.10624399855732918,
      "num_tokens": 1880995.0,
      "step": 850
    },
    {
      "entropy": 6.465567302703858,
      "epoch": 0.8623298033282905,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000427,
      "loss": 6.4162,
      "mean_token_accuracy": 0.10567688420414925,
      "num_tokens": 1892633.0,
      "step": 855
    },
    {
      "entropy": 6.5905781269073485,
      "epoch": 0.8673726676752396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004295,
      "loss": 6.5011,
      "mean_token_accuracy": 0.09975023046135903,
      "num_tokens": 1904802.0,
      "step": 860
    },
    {
      "entropy": 6.505033540725708,
      "epoch": 0.8724155320221886,
      "grad_norm": 1.09375,
      "learning_rate": 0.000432,
      "loss": 6.492,
      "mean_token_accuracy": 0.10669593513011932,
      "num_tokens": 1915755.0,
      "step": 865
    },
    {
      "entropy": 6.387980556488037,
      "epoch": 0.8774583963691377,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004345,
      "loss": 6.2328,
      "mean_token_accuracy": 0.10829911306500435,
      "num_tokens": 1926098.0,
      "step": 870
    },
    {
      "entropy": 6.521399021148682,
      "epoch": 0.8825012607160867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000437,
      "loss": 6.4569,
      "mean_token_accuracy": 0.10727727934718131,
      "num_tokens": 1936607.0,
      "step": 875
    },
    {
      "entropy": 6.403067350387573,
      "epoch": 0.8875441250630358,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004395,
      "loss": 6.3421,
      "mean_token_accuracy": 0.10624040365219116,
      "num_tokens": 1948596.0,
      "step": 880
    },
    {
      "entropy": 6.43417534828186,
      "epoch": 0.8925869894099848,
      "grad_norm": 0.953125,
      "learning_rate": 0.000442,
      "loss": 6.4263,
      "mean_token_accuracy": 0.11106917932629586,
      "num_tokens": 1960322.0,
      "step": 885
    },
    {
      "entropy": 6.524694728851318,
      "epoch": 0.8976298537569339,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004445,
      "loss": 6.3527,
      "mean_token_accuracy": 0.11080583110451699,
      "num_tokens": 1971052.0,
      "step": 890
    },
    {
      "entropy": 6.44118709564209,
      "epoch": 0.902672718103883,
      "grad_norm": 1.03125,
      "learning_rate": 0.000447,
      "loss": 6.3485,
      "mean_token_accuracy": 0.10465901717543602,
      "num_tokens": 1980917.0,
      "step": 895
    },
    {
      "entropy": 6.377262210845947,
      "epoch": 0.9077155824508321,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3261,
      "mean_token_accuracy": 0.11112496107816697,
      "num_tokens": 1991035.0,
      "step": 900
    },
    {
      "entropy": 6.4430749893188475,
      "epoch": 0.9127584467977812,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3419,
      "mean_token_accuracy": 0.10812708809971809,
      "num_tokens": 2000587.0,
      "step": 905
    },
    {
      "entropy": 6.400346374511718,
      "epoch": 0.9178013111447302,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3057,
      "mean_token_accuracy": 0.11217240318655967,
      "num_tokens": 2011087.0,
      "step": 910
    },
    {
      "entropy": 6.43453688621521,
      "epoch": 0.9228441754916793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3459,
      "mean_token_accuracy": 0.10732184424996376,
      "num_tokens": 2023472.0,
      "step": 915
    },
    {
      "entropy": 6.432980585098266,
      "epoch": 0.9278870398386283,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3346,
      "mean_token_accuracy": 0.1071820467710495,
      "num_tokens": 2033783.0,
      "step": 920
    },
    {
      "entropy": 6.326636266708374,
      "epoch": 0.9329299041855774,
      "grad_norm": 1.0,
      "learning_rate": 0.000462,
      "loss": 6.3581,
      "mean_token_accuracy": 0.11071096733212471,
      "num_tokens": 2045882.0,
      "step": 925
    },
    {
      "entropy": 6.479465627670288,
      "epoch": 0.9379727685325264,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004645,
      "loss": 6.3877,
      "mean_token_accuracy": 0.10236008018255234,
      "num_tokens": 2056549.0,
      "step": 930
    },
    {
      "entropy": 6.462947130203247,
      "epoch": 0.9430156328794755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000467,
      "loss": 6.3814,
      "mean_token_accuracy": 0.11182121336460113,
      "num_tokens": 2067540.0,
      "step": 935
    },
    {
      "entropy": 6.290677881240844,
      "epoch": 0.9480584972264247,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004695,
      "loss": 6.276,
      "mean_token_accuracy": 0.10857773795723916,
      "num_tokens": 2077386.0,
      "step": 940
    },
    {
      "entropy": 6.355180120468139,
      "epoch": 0.9531013615733737,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000472,
      "loss": 6.2417,
      "mean_token_accuracy": 0.11178631335496902,
      "num_tokens": 2087777.0,
      "step": 945
    },
    {
      "entropy": 6.281049156188965,
      "epoch": 0.9581442259203228,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745,
      "loss": 6.2404,
      "mean_token_accuracy": 0.11323176622390747,
      "num_tokens": 2099071.0,
      "step": 950
    },
    {
      "entropy": 6.204891157150269,
      "epoch": 0.9631870902672718,
      "grad_norm": 1.15625,
      "learning_rate": 0.000477,
      "loss": 6.3123,
      "mean_token_accuracy": 0.1093476541340351,
      "num_tokens": 2109168.0,
      "step": 955
    },
    {
      "entropy": 6.614584875106812,
      "epoch": 0.9682299546142209,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004795,
      "loss": 6.4215,
      "mean_token_accuracy": 0.10730769336223603,
      "num_tokens": 2120696.0,
      "step": 960
    },
    {
      "entropy": 6.28635835647583,
      "epoch": 0.9732728189611699,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000482,
      "loss": 6.37,
      "mean_token_accuracy": 0.10793575569987297,
      "num_tokens": 2132427.0,
      "step": 965
    },
    {
      "entropy": 6.3901630401611325,
      "epoch": 0.978315683308119,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004845,
      "loss": 6.3436,
      "mean_token_accuracy": 0.1093747541308403,
      "num_tokens": 2143120.0,
      "step": 970
    },
    {
      "entropy": 6.467719650268554,
      "epoch": 0.983358547655068,
      "grad_norm": 1.015625,
      "learning_rate": 0.000487,
      "loss": 6.3473,
      "mean_token_accuracy": 0.10565910637378692,
      "num_tokens": 2154154.0,
      "step": 975
    },
    {
      "entropy": 6.314101648330689,
      "epoch": 0.9884014120020171,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004895,
      "loss": 6.4482,
      "mean_token_accuracy": 0.10238303095102311,
      "num_tokens": 2167000.0,
      "step": 980
    },
    {
      "entropy": 6.403873634338379,
      "epoch": 0.9934442763489663,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000492,
      "loss": 6.3777,
      "mean_token_accuracy": 0.10910984128713608,
      "num_tokens": 2178100.0,
      "step": 985
    },
    {
      "entropy": 6.348555374145508,
      "epoch": 0.9984871406959153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004945,
      "loss": 6.3173,
      "mean_token_accuracy": 0.10381858721375466,
      "num_tokens": 2189063.0,
      "step": 990
    },
    {
      "entropy": 6.4779570367601185,
      "epoch": 1.0030257186081695,
      "grad_norm": 1.03125,
      "learning_rate": 0.000497,
      "loss": 6.2356,
      "mean_token_accuracy": 0.11199044684569041,
      "num_tokens": 2198302.0,
      "step": 995
    },
    {
      "entropy": 6.333552265167237,
      "epoch": 1.0080685829551186,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995,
      "loss": 6.1906,
      "mean_token_accuracy": 0.1150072768330574,
      "num_tokens": 2209648.0,
      "step": 1000
    },
    {
      "epoch": 1.0080685829551186,
      "eval_entropy": 6.4261680559347605,
      "eval_loss": 6.417810440063477,
      "eval_mean_token_accuracy": 0.10917834623813932,
      "eval_num_tokens": 2209648.0,
      "eval_runtime": 2.6477,
      "eval_samples_per_second": 1185.162,
      "eval_steps_per_second": 148.429,
      "step": 1000
    },
    {
      "entropy": 6.252728891372681,
      "epoch": 1.0131114473020675,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999997762223182,
      "loss": 6.1308,
      "mean_token_accuracy": 0.11400617361068725,
      "num_tokens": 2220806.0,
      "step": 1005
    },
    {
      "entropy": 6.257132434844971,
      "epoch": 1.0181543116490166,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999988671262489,
      "loss": 6.1254,
      "mean_token_accuracy": 0.11259887367486954,
      "num_tokens": 2231806.0,
      "step": 1010
    },
    {
      "entropy": 6.295977258682251,
      "epoch": 1.0231971759959657,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999972587285105,
      "loss": 6.1821,
      "mean_token_accuracy": 0.11922840997576714,
      "num_tokens": 2243835.0,
      "step": 1015
    },
    {
      "entropy": 6.117448616027832,
      "epoch": 1.0282400403429148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999949510341015,
      "loss": 6.0891,
      "mean_token_accuracy": 0.11738667264580727,
      "num_tokens": 2254722.0,
      "step": 1020
    },
    {
      "entropy": 6.269881677627564,
      "epoch": 1.033282904689864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999919440501947,
      "loss": 6.0294,
      "mean_token_accuracy": 0.12477270513772964,
      "num_tokens": 2264806.0,
      "step": 1025
    },
    {
      "entropy": 6.358243036270141,
      "epoch": 1.0383257690368128,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999882377861355,
      "loss": 6.2479,
      "mean_token_accuracy": 0.11410975456237793,
      "num_tokens": 2277037.0,
      "step": 1030
    },
    {
      "entropy": 6.277375936508179,
      "epoch": 1.043368633383762,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999838322534433,
      "loss": 6.1785,
      "mean_token_accuracy": 0.1084089681506157,
      "num_tokens": 2289829.0,
      "step": 1035
    },
    {
      "entropy": 6.295744276046753,
      "epoch": 1.048411497730711,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999787274658107,
      "loss": 6.0711,
      "mean_token_accuracy": 0.11793513298034668,
      "num_tokens": 2300340.0,
      "step": 1040
    },
    {
      "entropy": 6.199061441421509,
      "epoch": 1.0534543620776602,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999729234391032,
      "loss": 6.1059,
      "mean_token_accuracy": 0.12160981521010399,
      "num_tokens": 2310009.0,
      "step": 1045
    },
    {
      "entropy": 6.198751163482666,
      "epoch": 1.058497226424609,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049996642019136,
      "loss": 6.1428,
      "mean_token_accuracy": 0.11531612128019333,
      "num_tokens": 2319290.0,
      "step": 1050
    },
    {
      "entropy": 6.297437572479248,
      "epoch": 1.0635400907715582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999592177427933,
      "loss": 6.0857,
      "mean_token_accuracy": 0.12294981405138969,
      "num_tokens": 2329170.0,
      "step": 1055
    },
    {
      "entropy": 6.1483875751495365,
      "epoch": 1.0685829551185073,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999513161157885,
      "loss": 6.105,
      "mean_token_accuracy": 0.11440970972180367,
      "num_tokens": 2339617.0,
      "step": 1060
    },
    {
      "entropy": 6.260158061981201,
      "epoch": 1.0736258194654564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999427153349041,
      "loss": 6.0714,
      "mean_token_accuracy": 0.12414619326591492,
      "num_tokens": 2349548.0,
      "step": 1065
    },
    {
      "entropy": 6.0723645210266115,
      "epoch": 1.0786686838124055,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999334154268714,
      "loss": 6.0752,
      "mean_token_accuracy": 0.11783618927001953,
      "num_tokens": 2359116.0,
      "step": 1070
    },
    {
      "entropy": 6.276753377914429,
      "epoch": 1.0837115481593544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999234164205949,
      "loss": 6.1327,
      "mean_token_accuracy": 0.12174399495124817,
      "num_tokens": 2369412.0,
      "step": 1075
    },
    {
      "entropy": 6.1624593257904055,
      "epoch": 1.0887544125063036,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999127183471517,
      "loss": 6.2097,
      "mean_token_accuracy": 0.11544428989291192,
      "num_tokens": 2381276.0,
      "step": 1080
    },
    {
      "entropy": 6.245842838287354,
      "epoch": 1.0937972768532527,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999013212397915,
      "loss": 6.1076,
      "mean_token_accuracy": 0.11853586509823799,
      "num_tokens": 2392135.0,
      "step": 1085
    },
    {
      "entropy": 6.093086051940918,
      "epoch": 1.0988401412002018,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998892251339368,
      "loss": 6.0149,
      "mean_token_accuracy": 0.1269606776535511,
      "num_tokens": 2402957.0,
      "step": 1090
    },
    {
      "entropy": 6.243423843383789,
      "epoch": 1.1038830055471507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998764300671827,
      "loss": 6.1579,
      "mean_token_accuracy": 0.12277726605534553,
      "num_tokens": 2413386.0,
      "step": 1095
    },
    {
      "entropy": 6.157353162765503,
      "epoch": 1.1089258698940998,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998629360792965,
      "loss": 6.1449,
      "mean_token_accuracy": 0.11866232082247734,
      "num_tokens": 2424441.0,
      "step": 1100
    },
    {
      "entropy": 6.189147281646728,
      "epoch": 1.113968734241049,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998487432122179,
      "loss": 6.1766,
      "mean_token_accuracy": 0.11409586071968078,
      "num_tokens": 2436144.0,
      "step": 1105
    },
    {
      "entropy": 6.2349693298339846,
      "epoch": 1.119011598587998,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998338515100585,
      "loss": 6.0813,
      "mean_token_accuracy": 0.12133537903428078,
      "num_tokens": 2447393.0,
      "step": 1110
    },
    {
      "entropy": 6.068421173095703,
      "epoch": 1.1240544629349472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998182610191022,
      "loss": 6.0428,
      "mean_token_accuracy": 0.12405308485031127,
      "num_tokens": 2458632.0,
      "step": 1115
    },
    {
      "entropy": 6.2577080726623535,
      "epoch": 1.129097327281896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998019717878047,
      "loss": 6.1288,
      "mean_token_accuracy": 0.11385903730988503,
      "num_tokens": 2469331.0,
      "step": 1120
    },
    {
      "entropy": 6.227204322814941,
      "epoch": 1.1341401916288452,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499784983866793,
      "loss": 6.1671,
      "mean_token_accuracy": 0.11437357068061829,
      "num_tokens": 2479501.0,
      "step": 1125
    },
    {
      "entropy": 6.0807130336761475,
      "epoch": 1.1391830559757943,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997672973088661,
      "loss": 6.0287,
      "mean_token_accuracy": 0.12107328623533249,
      "num_tokens": 2490087.0,
      "step": 1130
    },
    {
      "entropy": 6.214322185516357,
      "epoch": 1.1442259203227434,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997489121689944,
      "loss": 6.0967,
      "mean_token_accuracy": 0.11605892032384872,
      "num_tokens": 2501130.0,
      "step": 1135
    },
    {
      "entropy": 6.246376657485962,
      "epoch": 1.1492687846696923,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997298285043189,
      "loss": 6.2169,
      "mean_token_accuracy": 0.11053206399083138,
      "num_tokens": 2513999.0,
      "step": 1140
    },
    {
      "entropy": 6.166038084030151,
      "epoch": 1.1543116490166414,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997100463741527,
      "loss": 6.1287,
      "mean_token_accuracy": 0.11388898193836212,
      "num_tokens": 2525579.0,
      "step": 1145
    },
    {
      "entropy": 6.167115879058838,
      "epoch": 1.1593545133635905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996895658399785,
      "loss": 6.0949,
      "mean_token_accuracy": 0.12076913714408874,
      "num_tokens": 2536091.0,
      "step": 1150
    },
    {
      "entropy": 6.126659774780274,
      "epoch": 1.1643973777105396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996683869654507,
      "loss": 6.1302,
      "mean_token_accuracy": 0.12046247944235802,
      "num_tokens": 2546073.0,
      "step": 1155
    },
    {
      "entropy": 6.1831365585327145,
      "epoch": 1.1694402420574885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996465098163934,
      "loss": 6.0907,
      "mean_token_accuracy": 0.11801456063985824,
      "num_tokens": 2556626.0,
      "step": 1160
    },
    {
      "entropy": 6.1779944896698,
      "epoch": 1.1744831064044376,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996239344608015,
      "loss": 6.1206,
      "mean_token_accuracy": 0.12329623401165009,
      "num_tokens": 2567304.0,
      "step": 1165
    },
    {
      "entropy": 6.129946279525757,
      "epoch": 1.1795259707513868,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996006609688398,
      "loss": 5.9896,
      "mean_token_accuracy": 0.1163274422287941,
      "num_tokens": 2577242.0,
      "step": 1170
    },
    {
      "entropy": 6.059417057037353,
      "epoch": 1.1845688350983359,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995766894128426,
      "loss": 6.0956,
      "mean_token_accuracy": 0.11244072541594505,
      "num_tokens": 2588330.0,
      "step": 1175
    },
    {
      "entropy": 6.160956144332886,
      "epoch": 1.189611699445285,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995520198673143,
      "loss": 6.0936,
      "mean_token_accuracy": 0.12082465887069702,
      "num_tokens": 2598471.0,
      "step": 1180
    },
    {
      "entropy": 6.21316089630127,
      "epoch": 1.194654563792234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995266524089283,
      "loss": 6.1817,
      "mean_token_accuracy": 0.11847011670470238,
      "num_tokens": 2610238.0,
      "step": 1185
    },
    {
      "entropy": 6.094186544418335,
      "epoch": 1.199697428139183,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995005871165271,
      "loss": 6.0416,
      "mean_token_accuracy": 0.1227484904229641,
      "num_tokens": 2619870.0,
      "step": 1190
    },
    {
      "entropy": 6.185560274124145,
      "epoch": 1.2047402924861321,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994738240711225,
      "loss": 6.1908,
      "mean_token_accuracy": 0.11310086920857429,
      "num_tokens": 2631541.0,
      "step": 1195
    },
    {
      "entropy": 6.106654500961303,
      "epoch": 1.2097831568330812,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994463633558945,
      "loss": 6.0229,
      "mean_token_accuracy": 0.12247754111886025,
      "num_tokens": 2641896.0,
      "step": 1200
    },
    {
      "entropy": 6.082180595397949,
      "epoch": 1.2148260211800301,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994182050561917,
      "loss": 6.1038,
      "mean_token_accuracy": 0.12054873630404472,
      "num_tokens": 2654208.0,
      "step": 1205
    },
    {
      "entropy": 6.143400192260742,
      "epoch": 1.2198688855269793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993893492595306,
      "loss": 6.0785,
      "mean_token_accuracy": 0.12056932002305984,
      "num_tokens": 2665115.0,
      "step": 1210
    },
    {
      "entropy": 6.13805513381958,
      "epoch": 1.2249117498739284,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993597960555959,
      "loss": 6.1659,
      "mean_token_accuracy": 0.11709513887763023,
      "num_tokens": 2676093.0,
      "step": 1215
    },
    {
      "entropy": 6.216587829589844,
      "epoch": 1.2299546142208775,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993295455362395,
      "loss": 6.1297,
      "mean_token_accuracy": 0.12260301932692527,
      "num_tokens": 2687866.0,
      "step": 1220
    },
    {
      "entropy": 6.135248184204102,
      "epoch": 1.2349974785678266,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992985977954808,
      "loss": 6.1339,
      "mean_token_accuracy": 0.11458160728216171,
      "num_tokens": 2699814.0,
      "step": 1225
    },
    {
      "entropy": 6.165663337707519,
      "epoch": 1.2400403429147757,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499266952929506,
      "loss": 6.0974,
      "mean_token_accuracy": 0.12700220122933387,
      "num_tokens": 2710384.0,
      "step": 1230
    },
    {
      "entropy": 6.150148534774781,
      "epoch": 1.2450832072617246,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992346110366682,
      "loss": 6.0655,
      "mean_token_accuracy": 0.11997778415679931,
      "num_tokens": 2721442.0,
      "step": 1235
    },
    {
      "entropy": 6.090733575820923,
      "epoch": 1.2501260716086737,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992015722174865,
      "loss": 6.0888,
      "mean_token_accuracy": 0.11658885851502418,
      "num_tokens": 2733605.0,
      "step": 1240
    },
    {
      "entropy": 6.171622323989868,
      "epoch": 1.2551689359556228,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991678365746466,
      "loss": 6.128,
      "mean_token_accuracy": 0.12494012638926506,
      "num_tokens": 2744853.0,
      "step": 1245
    },
    {
      "entropy": 6.0140642642974855,
      "epoch": 1.2602118003025717,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991334042129996,
      "loss": 6.023,
      "mean_token_accuracy": 0.12799779549241067,
      "num_tokens": 2754477.0,
      "step": 1250
    },
    {
      "entropy": 6.170590877532959,
      "epoch": 1.2652546646495209,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990982752395618,
      "loss": 6.0337,
      "mean_token_accuracy": 0.12020832374691963,
      "num_tokens": 2765895.0,
      "step": 1255
    },
    {
      "entropy": 6.077351331710815,
      "epoch": 1.27029752899647,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990624497635152,
      "loss": 6.0866,
      "mean_token_accuracy": 0.11977267414331436,
      "num_tokens": 2776897.0,
      "step": 1260
    },
    {
      "entropy": 6.209859228134155,
      "epoch": 1.275340393343419,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499025927896206,
      "loss": 6.1038,
      "mean_token_accuracy": 0.12009399533271789,
      "num_tokens": 2788666.0,
      "step": 1265
    },
    {
      "entropy": 6.031207752227783,
      "epoch": 1.2803832576903682,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989887097511451,
      "loss": 5.9986,
      "mean_token_accuracy": 0.12084411457180977,
      "num_tokens": 2799167.0,
      "step": 1270
    },
    {
      "entropy": 6.0679034233093265,
      "epoch": 1.2854261220373173,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989507954440074,
      "loss": 6.0037,
      "mean_token_accuracy": 0.11892658248543739,
      "num_tokens": 2809774.0,
      "step": 1275
    },
    {
      "entropy": 6.099821901321411,
      "epoch": 1.2904689863842662,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004989121850926313,
      "loss": 6.1233,
      "mean_token_accuracy": 0.11551008000969887,
      "num_tokens": 2820515.0,
      "step": 1280
    },
    {
      "entropy": 6.086860656738281,
      "epoch": 1.2955118507312153,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004988728788170188,
      "loss": 6.0611,
      "mean_token_accuracy": 0.12110966295003892,
      "num_tokens": 2832163.0,
      "step": 1285
    },
    {
      "entropy": 6.091725492477417,
      "epoch": 1.3005547150781644,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988328767393346,
      "loss": 6.0574,
      "mean_token_accuracy": 0.12182244136929513,
      "num_tokens": 2843717.0,
      "step": 1290
    },
    {
      "entropy": 5.981151485443116,
      "epoch": 1.3055975794251133,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987921789839061,
      "loss": 6.0158,
      "mean_token_accuracy": 0.12019825726747513,
      "num_tokens": 2854489.0,
      "step": 1295
    },
    {
      "entropy": 6.1015993595123295,
      "epoch": 1.3106404437720625,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987507856772229,
      "loss": 6.0627,
      "mean_token_accuracy": 0.1227644480764866,
      "num_tokens": 2865317.0,
      "step": 1300
    },
    {
      "entropy": 6.058894157409668,
      "epoch": 1.3156833081190116,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987086969479364,
      "loss": 5.9809,
      "mean_token_accuracy": 0.12572994455695152,
      "num_tokens": 2877466.0,
      "step": 1305
    },
    {
      "entropy": 6.00954852104187,
      "epoch": 1.3207261724659607,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004986659129268592,
      "loss": 6.0129,
      "mean_token_accuracy": 0.12269562259316444,
      "num_tokens": 2887698.0,
      "step": 1310
    },
    {
      "entropy": 5.980123376846313,
      "epoch": 1.3257690368129098,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986224337469652,
      "loss": 5.9609,
      "mean_token_accuracy": 0.12754920348525048,
      "num_tokens": 2899339.0,
      "step": 1315
    },
    {
      "entropy": 6.056573486328125,
      "epoch": 1.330811901159859,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985782595433886,
      "loss": 6.0081,
      "mean_token_accuracy": 0.12356857880949974,
      "num_tokens": 2908340.0,
      "step": 1320
    },
    {
      "entropy": 6.078930139541626,
      "epoch": 1.3358547655068078,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985333904534237,
      "loss": 6.1003,
      "mean_token_accuracy": 0.12403591498732566,
      "num_tokens": 2920432.0,
      "step": 1325
    },
    {
      "entropy": 6.12197904586792,
      "epoch": 1.340897629853757,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498487826616525,
      "loss": 6.055,
      "mean_token_accuracy": 0.12046747058629989,
      "num_tokens": 2930385.0,
      "step": 1330
    },
    {
      "entropy": 6.096244239807129,
      "epoch": 1.345940494200706,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004984415681743056,
      "loss": 5.9999,
      "mean_token_accuracy": 0.12470603808760643,
      "num_tokens": 2940760.0,
      "step": 1335
    },
    {
      "entropy": 6.053582334518433,
      "epoch": 1.350983358547655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983946152705382,
      "loss": 6.0303,
      "mean_token_accuracy": 0.1265427976846695,
      "num_tokens": 2952441.0,
      "step": 1340
    },
    {
      "entropy": 6.091333484649658,
      "epoch": 1.356026222894604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983469680511531,
      "loss": 6.0466,
      "mean_token_accuracy": 0.12300621345639229,
      "num_tokens": 2962766.0,
      "step": 1345
    },
    {
      "entropy": 5.971969509124756,
      "epoch": 1.3610690872415532,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004982986266642393,
      "loss": 5.9324,
      "mean_token_accuracy": 0.12989450618624687,
      "num_tokens": 2974239.0,
      "step": 1350
    },
    {
      "entropy": 6.044353055953979,
      "epoch": 1.3661119515885023,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004982495912600426,
      "loss": 5.9819,
      "mean_token_accuracy": 0.12527156621217728,
      "num_tokens": 2984497.0,
      "step": 1355
    },
    {
      "entropy": 6.031408786773682,
      "epoch": 1.3711548159354514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004981998619909664,
      "loss": 5.9519,
      "mean_token_accuracy": 0.13078300505876542,
      "num_tokens": 2995367.0,
      "step": 1360
    },
    {
      "entropy": 6.021449947357178,
      "epoch": 1.3761976802824005,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981494390115704,
      "loss": 5.9832,
      "mean_token_accuracy": 0.1252884514629841,
      "num_tokens": 3007687.0,
      "step": 1365
    },
    {
      "entropy": 6.037176036834717,
      "epoch": 1.3812405446293494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980983224785704,
      "loss": 5.9876,
      "mean_token_accuracy": 0.12274198979139328,
      "num_tokens": 3017505.0,
      "step": 1370
    },
    {
      "entropy": 5.987515687942505,
      "epoch": 1.3862834089762985,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004980465125508376,
      "loss": 5.9261,
      "mean_token_accuracy": 0.12372706532478332,
      "num_tokens": 3028271.0,
      "step": 1375
    },
    {
      "entropy": 6.0109961986541744,
      "epoch": 1.3913262733232477,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979940093893986,
      "loss": 5.9846,
      "mean_token_accuracy": 0.12520991489291192,
      "num_tokens": 3040440.0,
      "step": 1380
    },
    {
      "entropy": 5.929500961303711,
      "epoch": 1.3963691376701965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004979408131574345,
      "loss": 5.9556,
      "mean_token_accuracy": 0.12497764006257057,
      "num_tokens": 3050999.0,
      "step": 1385
    },
    {
      "entropy": 6.053280925750732,
      "epoch": 1.4014120020171457,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978869240202802,
      "loss": 5.9992,
      "mean_token_accuracy": 0.12776673436164857,
      "num_tokens": 3063374.0,
      "step": 1390
    },
    {
      "entropy": 6.028737211227417,
      "epoch": 1.4064548663640948,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004978323421454246,
      "loss": 6.0395,
      "mean_token_accuracy": 0.12454756051301956,
      "num_tokens": 3075464.0,
      "step": 1395
    },
    {
      "entropy": 5.949777460098266,
      "epoch": 1.411497730711044,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977770677025092,
      "loss": 5.9103,
      "mean_token_accuracy": 0.13049111217260362,
      "num_tokens": 3086398.0,
      "step": 1400
    },
    {
      "entropy": 5.946027517318726,
      "epoch": 1.416540595057993,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004977211008633284,
      "loss": 6.0461,
      "mean_token_accuracy": 0.11787689551711082,
      "num_tokens": 3097777.0,
      "step": 1405
    },
    {
      "entropy": 6.0807273387908936,
      "epoch": 1.4215834594049421,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004976644418018285,
      "loss": 6.0026,
      "mean_token_accuracy": 0.13193272352218627,
      "num_tokens": 3108711.0,
      "step": 1410
    },
    {
      "entropy": 6.04375581741333,
      "epoch": 1.426626323751891,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497607090694107,
      "loss": 6.0231,
      "mean_token_accuracy": 0.12179360166192055,
      "num_tokens": 3119589.0,
      "step": 1415
    },
    {
      "entropy": 6.017640018463135,
      "epoch": 1.4316691880988401,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975490477184126,
      "loss": 5.9568,
      "mean_token_accuracy": 0.13265052065253258,
      "num_tokens": 3131744.0,
      "step": 1420
    },
    {
      "entropy": 6.093933820724487,
      "epoch": 1.4367120524457893,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497490313055144,
      "loss": 6.003,
      "mean_token_accuracy": 0.12273363247513772,
      "num_tokens": 3143526.0,
      "step": 1425
    },
    {
      "entropy": 5.946259832382202,
      "epoch": 1.4417549167927382,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004974308868868501,
      "loss": 5.9407,
      "mean_token_accuracy": 0.12551197931170463,
      "num_tokens": 3154561.0,
      "step": 1430
    },
    {
      "entropy": 6.044280624389648,
      "epoch": 1.4467977811396873,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973707693982287,
      "loss": 6.1069,
      "mean_token_accuracy": 0.1201934814453125,
      "num_tokens": 3166052.0,
      "step": 1435
    },
    {
      "entropy": 6.130475854873657,
      "epoch": 1.4518406454866364,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004973099607761264,
      "loss": 6.0729,
      "mean_token_accuracy": 0.12008198499679565,
      "num_tokens": 3179335.0,
      "step": 1440
    },
    {
      "entropy": 5.9311377048492435,
      "epoch": 1.4568835098335855,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004972484612095378,
      "loss": 5.957,
      "mean_token_accuracy": 0.12554447948932648,
      "num_tokens": 3192129.0,
      "step": 1445
    },
    {
      "entropy": 6.110207796096802,
      "epoch": 1.4619263741805346,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497186270889605,
      "loss": 6.0633,
      "mean_token_accuracy": 0.12692347541451454,
      "num_tokens": 3202792.0,
      "step": 1450
    },
    {
      "entropy": 6.0356773853302,
      "epoch": 1.4669692385274837,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004971233900096169,
      "loss": 6.0074,
      "mean_token_accuracy": 0.12310663536190987,
      "num_tokens": 3213707.0,
      "step": 1455
    },
    {
      "entropy": 6.061852645874024,
      "epoch": 1.4720121028744326,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004970598187650085,
      "loss": 5.9785,
      "mean_token_accuracy": 0.1272933304309845,
      "num_tokens": 3224531.0,
      "step": 1460
    },
    {
      "entropy": 5.997699165344239,
      "epoch": 1.4770549672213817,
      "grad_norm": 1.0,
      "learning_rate": 0.000496995557353361,
      "loss": 5.9685,
      "mean_token_accuracy": 0.12390959337353706,
      "num_tokens": 3234656.0,
      "step": 1465
    },
    {
      "entropy": 5.925370979309082,
      "epoch": 1.4820978315683309,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969306059744004,
      "loss": 5.9988,
      "mean_token_accuracy": 0.12714771330356597,
      "num_tokens": 3246745.0,
      "step": 1470
    },
    {
      "entropy": 6.007743883132934,
      "epoch": 1.4871406959152798,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004968649648299968,
      "loss": 5.9677,
      "mean_token_accuracy": 0.12777481228113174,
      "num_tokens": 3257277.0,
      "step": 1475
    },
    {
      "entropy": 6.022071361541748,
      "epoch": 1.4921835602622289,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967986341241646,
      "loss": 5.9527,
      "mean_token_accuracy": 0.12892103865742682,
      "num_tokens": 3267704.0,
      "step": 1480
    },
    {
      "entropy": 6.055181169509888,
      "epoch": 1.497226424609178,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004967316140630611,
      "loss": 6.0201,
      "mean_token_accuracy": 0.1264706738293171,
      "num_tokens": 3279772.0,
      "step": 1485
    },
    {
      "entropy": 5.977132177352905,
      "epoch": 1.502269288956127,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966639048549862,
      "loss": 5.9196,
      "mean_token_accuracy": 0.13101670071482657,
      "num_tokens": 3290536.0,
      "step": 1490
    },
    {
      "entropy": 5.940878009796142,
      "epoch": 1.5073121533030762,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965955067103816,
      "loss": 6.0031,
      "mean_token_accuracy": 0.13029861822724342,
      "num_tokens": 3302051.0,
      "step": 1495
    },
    {
      "entropy": 6.055119800567627,
      "epoch": 1.5123550176500253,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004965264198418305,
      "loss": 5.9755,
      "mean_token_accuracy": 0.11771604046225548,
      "num_tokens": 3314324.0,
      "step": 1500
    },
    {
      "epoch": 1.5123550176500253,
      "eval_entropy": 5.820931597823709,
      "eval_loss": 6.134761333465576,
      "eval_mean_token_accuracy": 0.125417051083259,
      "eval_num_tokens": 3314324.0,
      "eval_runtime": 2.6525,
      "eval_samples_per_second": 1183.028,
      "eval_steps_per_second": 148.161,
      "step": 1500
    },
    {
      "entropy": 5.888900661468506,
      "epoch": 1.5173978819969742,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004964566444640564,
      "loss": 5.9638,
      "mean_token_accuracy": 0.12984286695718766,
      "num_tokens": 3326246.0,
      "step": 1505
    },
    {
      "entropy": 5.980616760253906,
      "epoch": 1.5224407463439233,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004963861807939227,
      "loss": 5.9547,
      "mean_token_accuracy": 0.12878640592098237,
      "num_tokens": 3336308.0,
      "step": 1510
    },
    {
      "entropy": 5.875748062133789,
      "epoch": 1.5274836106908725,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004963150290504321,
      "loss": 5.8941,
      "mean_token_accuracy": 0.1307993695139885,
      "num_tokens": 3347596.0,
      "step": 1515
    },
    {
      "entropy": 5.888290357589722,
      "epoch": 1.5325264750378214,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004962431894547259,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1301721751689911,
      "num_tokens": 3358370.0,
      "step": 1520
    },
    {
      "entropy": 6.044656658172608,
      "epoch": 1.5375693393847705,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004961706622300834,
      "loss": 5.8976,
      "mean_token_accuracy": 0.12725164890289306,
      "num_tokens": 3368630.0,
      "step": 1525
    },
    {
      "entropy": 5.95916166305542,
      "epoch": 1.5426122037317196,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004960974476019206,
      "loss": 5.9281,
      "mean_token_accuracy": 0.13028757646679878,
      "num_tokens": 3379952.0,
      "step": 1530
    },
    {
      "entropy": 5.915283250808716,
      "epoch": 1.5476550680786687,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960235457977905,
      "loss": 5.8544,
      "mean_token_accuracy": 0.1292065441608429,
      "num_tokens": 3390097.0,
      "step": 1535
    },
    {
      "entropy": 5.88209900856018,
      "epoch": 1.5526979324256178,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959489570473814,
      "loss": 5.8503,
      "mean_token_accuracy": 0.13215067759156227,
      "num_tokens": 3399778.0,
      "step": 1540
    },
    {
      "entropy": 5.941514205932617,
      "epoch": 1.557740796772567,
      "grad_norm": 0.890625,
      "learning_rate": 0.000495873681582517,
      "loss": 6.007,
      "mean_token_accuracy": 0.12837614566087724,
      "num_tokens": 3410213.0,
      "step": 1545
    },
    {
      "entropy": 6.009626150131226,
      "epoch": 1.5627836611195158,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957977196371552,
      "loss": 6.0055,
      "mean_token_accuracy": 0.12803423553705215,
      "num_tokens": 3422071.0,
      "step": 1550
    },
    {
      "entropy": 6.056908845901489,
      "epoch": 1.567826525466465,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957210714473873,
      "loss": 5.9833,
      "mean_token_accuracy": 0.13199174925684928,
      "num_tokens": 3432102.0,
      "step": 1555
    },
    {
      "entropy": 5.918883609771728,
      "epoch": 1.572869389813414,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004956437372514378,
      "loss": 5.9638,
      "mean_token_accuracy": 0.12932863235473632,
      "num_tokens": 3442821.0,
      "step": 1560
    },
    {
      "entropy": 5.923231077194214,
      "epoch": 1.577912254160363,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004955657172896631,
      "loss": 5.9211,
      "mean_token_accuracy": 0.12785825058817862,
      "num_tokens": 3454378.0,
      "step": 1565
    },
    {
      "entropy": 5.945256757736206,
      "epoch": 1.582955118507312,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004954870118045511,
      "loss": 5.941,
      "mean_token_accuracy": 0.13135533407330513,
      "num_tokens": 3463933.0,
      "step": 1570
    },
    {
      "entropy": 6.024400615692139,
      "epoch": 1.5879979828542612,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004954076210407203,
      "loss": 5.8713,
      "mean_token_accuracy": 0.12804725468158723,
      "num_tokens": 3473993.0,
      "step": 1575
    },
    {
      "entropy": 5.913674783706665,
      "epoch": 1.5930408472012103,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004953275452449188,
      "loss": 6.0226,
      "mean_token_accuracy": 0.12443782985210419,
      "num_tokens": 3484882.0,
      "step": 1580
    },
    {
      "entropy": 5.996275997161865,
      "epoch": 1.5980837115481594,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004952467846660243,
      "loss": 5.9353,
      "mean_token_accuracy": 0.12367153093218804,
      "num_tokens": 3496503.0,
      "step": 1585
    },
    {
      "entropy": 5.989222621917724,
      "epoch": 1.6031265758951085,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004951653395550423,
      "loss": 6.016,
      "mean_token_accuracy": 0.1291416108608246,
      "num_tokens": 3508696.0,
      "step": 1590
    },
    {
      "entropy": 6.024582004547119,
      "epoch": 1.6081694402420574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004950832101651062,
      "loss": 5.9267,
      "mean_token_accuracy": 0.13258333280682563,
      "num_tokens": 3519542.0,
      "step": 1595
    },
    {
      "entropy": 5.871862697601318,
      "epoch": 1.6132123045890066,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950003967514762,
      "loss": 5.9008,
      "mean_token_accuracy": 0.13136171102523803,
      "num_tokens": 3531576.0,
      "step": 1600
    },
    {
      "entropy": 5.985661792755127,
      "epoch": 1.6182551689359557,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000494916899571538,
      "loss": 5.8534,
      "mean_token_accuracy": 0.13423650190234185,
      "num_tokens": 3542816.0,
      "step": 1605
    },
    {
      "entropy": 5.966505241394043,
      "epoch": 1.6232980332829046,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004948327188848029,
      "loss": 5.9948,
      "mean_token_accuracy": 0.1255824327468872,
      "num_tokens": 3554100.0,
      "step": 1610
    },
    {
      "entropy": 6.077696657180786,
      "epoch": 1.6283408976298537,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004947478549529065,
      "loss": 5.8494,
      "mean_token_accuracy": 0.12402525842189789,
      "num_tokens": 3565048.0,
      "step": 1615
    },
    {
      "entropy": 5.930281019210815,
      "epoch": 1.6333837619768028,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004946623080396077,
      "loss": 5.9554,
      "mean_token_accuracy": 0.12922262623906136,
      "num_tokens": 3575587.0,
      "step": 1620
    },
    {
      "entropy": 5.955634641647339,
      "epoch": 1.638426626323752,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945760784107884,
      "loss": 6.0183,
      "mean_token_accuracy": 0.12855926603078843,
      "num_tokens": 3587059.0,
      "step": 1625
    },
    {
      "entropy": 5.9010406017303465,
      "epoch": 1.643469490670701,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004944891663344522,
      "loss": 5.8267,
      "mean_token_accuracy": 0.1329905480146408,
      "num_tokens": 3598492.0,
      "step": 1630
    },
    {
      "entropy": 5.891679334640503,
      "epoch": 1.6485123550176501,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004944015720807239,
      "loss": 5.9012,
      "mean_token_accuracy": 0.13142564743757248,
      "num_tokens": 3609513.0,
      "step": 1635
    },
    {
      "entropy": 5.980248069763183,
      "epoch": 1.653555219364599,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004943132959218484,
      "loss": 5.9307,
      "mean_token_accuracy": 0.13032082244753837,
      "num_tokens": 3621327.0,
      "step": 1640
    },
    {
      "entropy": 5.883354759216308,
      "epoch": 1.6585980837115482,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049422433813219,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1323447398841381,
      "num_tokens": 3632136.0,
      "step": 1645
    },
    {
      "entropy": 5.896560430526733,
      "epoch": 1.6636409480584973,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941346989882318,
      "loss": 5.8008,
      "mean_token_accuracy": 0.1278294138610363,
      "num_tokens": 3642271.0,
      "step": 1650
    },
    {
      "entropy": 5.7824047088623045,
      "epoch": 1.6686838124054462,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004940443787685739,
      "loss": 5.9415,
      "mean_token_accuracy": 0.12705803215503692,
      "num_tokens": 3652355.0,
      "step": 1655
    },
    {
      "entropy": 6.048564910888672,
      "epoch": 1.6737266767523953,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004939533777539341,
      "loss": 5.9195,
      "mean_token_accuracy": 0.13076957687735558,
      "num_tokens": 3663714.0,
      "step": 1660
    },
    {
      "entropy": 5.867811965942383,
      "epoch": 1.6787695410993444,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938616962271453,
      "loss": 5.8732,
      "mean_token_accuracy": 0.1254029929637909,
      "num_tokens": 3674675.0,
      "step": 1665
    },
    {
      "entropy": 5.935502529144287,
      "epoch": 1.6838124054462935,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004937693344731561,
      "loss": 5.9575,
      "mean_token_accuracy": 0.1326349377632141,
      "num_tokens": 3686403.0,
      "step": 1670
    },
    {
      "entropy": 5.9425211429595945,
      "epoch": 1.6888552697932426,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004936762927790286,
      "loss": 5.8342,
      "mean_token_accuracy": 0.13906463459134102,
      "num_tokens": 3697259.0,
      "step": 1675
    },
    {
      "entropy": 5.867275047302246,
      "epoch": 1.6938981341401917,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935825714339389,
      "loss": 5.9001,
      "mean_token_accuracy": 0.13478674218058587,
      "num_tokens": 3708536.0,
      "step": 1680
    },
    {
      "entropy": 6.012644243240357,
      "epoch": 1.6989409984871406,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004934881707291748,
      "loss": 5.9065,
      "mean_token_accuracy": 0.12929113656282426,
      "num_tokens": 3719181.0,
      "step": 1685
    },
    {
      "entropy": 5.826500463485718,
      "epoch": 1.7039838628340898,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933930909581362,
      "loss": 5.7538,
      "mean_token_accuracy": 0.13596596643328668,
      "num_tokens": 3729325.0,
      "step": 1690
    },
    {
      "entropy": 5.904290437698364,
      "epoch": 1.7090267271810389,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004932973324163328,
      "loss": 5.8745,
      "mean_token_accuracy": 0.13066629618406295,
      "num_tokens": 3740462.0,
      "step": 1695
    },
    {
      "entropy": 5.937609672546387,
      "epoch": 1.7140695915279878,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004932008954013846,
      "loss": 5.9797,
      "mean_token_accuracy": 0.12895917668938636,
      "num_tokens": 3752523.0,
      "step": 1700
    },
    {
      "entropy": 5.854755926132202,
      "epoch": 1.7191124558749369,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004931037802130201,
      "loss": 5.8405,
      "mean_token_accuracy": 0.13127497360110282,
      "num_tokens": 3764120.0,
      "step": 1705
    },
    {
      "entropy": 5.907617568969727,
      "epoch": 1.724155320221886,
      "grad_norm": 1.0,
      "learning_rate": 0.0004930059871530755,
      "loss": 5.972,
      "mean_token_accuracy": 0.12837913185358046,
      "num_tokens": 3776733.0,
      "step": 1710
    },
    {
      "entropy": 6.01545033454895,
      "epoch": 1.7291981845688351,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004929075165254938,
      "loss": 5.9337,
      "mean_token_accuracy": 0.12843016907572746,
      "num_tokens": 3787799.0,
      "step": 1715
    },
    {
      "entropy": 5.798501253128052,
      "epoch": 1.7342410489157842,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004928083686363241,
      "loss": 5.8562,
      "mean_token_accuracy": 0.1305542565882206,
      "num_tokens": 3797995.0,
      "step": 1720
    },
    {
      "entropy": 5.994955348968506,
      "epoch": 1.7392839132627333,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004927085437937204,
      "loss": 5.9414,
      "mean_token_accuracy": 0.12499688416719437,
      "num_tokens": 3809580.0,
      "step": 1725
    },
    {
      "entropy": 5.964277935028076,
      "epoch": 1.7443267776096822,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004926080423079404,
      "loss": 5.9669,
      "mean_token_accuracy": 0.13099026530981064,
      "num_tokens": 3820150.0,
      "step": 1730
    },
    {
      "entropy": 5.898656749725342,
      "epoch": 1.7493696419566314,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492506864491345,
      "loss": 5.813,
      "mean_token_accuracy": 0.12974431589245797,
      "num_tokens": 3830391.0,
      "step": 1735
    },
    {
      "entropy": 5.816258382797241,
      "epoch": 1.7544125063035805,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924050106583976,
      "loss": 5.8311,
      "mean_token_accuracy": 0.13523281887173652,
      "num_tokens": 3842773.0,
      "step": 1740
    },
    {
      "entropy": 5.9264304637908936,
      "epoch": 1.7594553706505294,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004923024811256618,
      "loss": 5.8615,
      "mean_token_accuracy": 0.13258160129189492,
      "num_tokens": 3853646.0,
      "step": 1745
    },
    {
      "entropy": 5.904428911209107,
      "epoch": 1.7644982349974785,
      "grad_norm": 0.9375,
      "learning_rate": 0.000492199276211802,
      "loss": 5.8332,
      "mean_token_accuracy": 0.1314268544316292,
      "num_tokens": 3865319.0,
      "step": 1750
    },
    {
      "entropy": 5.870353078842163,
      "epoch": 1.7695410993444276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004920953962375813,
      "loss": 5.9019,
      "mean_token_accuracy": 0.1266103930771351,
      "num_tokens": 3877107.0,
      "step": 1755
    },
    {
      "entropy": 5.8814592361450195,
      "epoch": 1.7745839636913767,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919908415258612,
      "loss": 5.9306,
      "mean_token_accuracy": 0.13517394736409188,
      "num_tokens": 3889334.0,
      "step": 1760
    },
    {
      "entropy": 5.958156681060791,
      "epoch": 1.7796268280383258,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004918856124016002,
      "loss": 5.8145,
      "mean_token_accuracy": 0.13674092888832093,
      "num_tokens": 3899482.0,
      "step": 1765
    },
    {
      "entropy": 5.764415264129639,
      "epoch": 1.784669692385275,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004917797091918526,
      "loss": 5.7577,
      "mean_token_accuracy": 0.1344623699784279,
      "num_tokens": 3910216.0,
      "step": 1770
    },
    {
      "entropy": 5.990887451171875,
      "epoch": 1.789712556732224,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004916731322257683,
      "loss": 5.9112,
      "mean_token_accuracy": 0.13275586888194085,
      "num_tokens": 3922246.0,
      "step": 1775
    },
    {
      "entropy": 5.829149913787842,
      "epoch": 1.794755421079173,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004915658818345908,
      "loss": 5.8148,
      "mean_token_accuracy": 0.140170219540596,
      "num_tokens": 3933688.0,
      "step": 1780
    },
    {
      "entropy": 5.8959815979003904,
      "epoch": 1.799798285426122,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004914579583516568,
      "loss": 5.9084,
      "mean_token_accuracy": 0.12661692649126052,
      "num_tokens": 3944168.0,
      "step": 1785
    },
    {
      "entropy": 5.872848033905029,
      "epoch": 1.804841149773071,
      "grad_norm": 1.015625,
      "learning_rate": 0.000491349362112395,
      "loss": 5.8429,
      "mean_token_accuracy": 0.13449825197458268,
      "num_tokens": 3954346.0,
      "step": 1790
    },
    {
      "entropy": 5.932722806930542,
      "epoch": 1.80988401412002,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000491240093454325,
      "loss": 5.946,
      "mean_token_accuracy": 0.13098567500710487,
      "num_tokens": 3966375.0,
      "step": 1795
    },
    {
      "entropy": 5.91828465461731,
      "epoch": 1.8149268784669692,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004911301527170563,
      "loss": 5.8963,
      "mean_token_accuracy": 0.1361001491546631,
      "num_tokens": 3977030.0,
      "step": 1800
    },
    {
      "entropy": 5.834843254089355,
      "epoch": 1.8199697428139183,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004910195402422873,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1292489267885685,
      "num_tokens": 3988064.0,
      "step": 1805
    },
    {
      "entropy": 5.923317813873291,
      "epoch": 1.8250126071608674,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004909082563738042,
      "loss": 5.863,
      "mean_token_accuracy": 0.13194844424724578,
      "num_tokens": 3997994.0,
      "step": 1810
    },
    {
      "entropy": 5.94038462638855,
      "epoch": 1.8300554715078166,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004907963014574797,
      "loss": 5.9382,
      "mean_token_accuracy": 0.12465495020151138,
      "num_tokens": 4008645.0,
      "step": 1815
    },
    {
      "entropy": 5.909869003295898,
      "epoch": 1.8350983358547657,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004906836758412722,
      "loss": 5.8625,
      "mean_token_accuracy": 0.13498851880431176,
      "num_tokens": 4020387.0,
      "step": 1820
    },
    {
      "entropy": 5.740508127212524,
      "epoch": 1.8401412002017146,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004905703798752248,
      "loss": 5.7427,
      "mean_token_accuracy": 0.1360771492123604,
      "num_tokens": 4032720.0,
      "step": 1825
    },
    {
      "entropy": 5.804358339309692,
      "epoch": 1.8451840645486637,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000490456413911464,
      "loss": 5.8135,
      "mean_token_accuracy": 0.1373528338968754,
      "num_tokens": 4043438.0,
      "step": 1830
    },
    {
      "entropy": 5.825810146331787,
      "epoch": 1.8502269288956126,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004903417783041987,
      "loss": 5.7941,
      "mean_token_accuracy": 0.14117383807897568,
      "num_tokens": 4054467.0,
      "step": 1835
    },
    {
      "entropy": 5.871437740325928,
      "epoch": 1.8552697932425617,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004902264734097188,
      "loss": 5.8019,
      "mean_token_accuracy": 0.13363157361745834,
      "num_tokens": 4066418.0,
      "step": 1840
    },
    {
      "entropy": 5.954784202575683,
      "epoch": 1.8603126575895108,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004901104995863948,
      "loss": 5.8892,
      "mean_token_accuracy": 0.1288447819650173,
      "num_tokens": 4078770.0,
      "step": 1845
    },
    {
      "entropy": 5.824422836303711,
      "epoch": 1.86535552193646,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004899938571946759,
      "loss": 5.9346,
      "mean_token_accuracy": 0.129574766010046,
      "num_tokens": 4089998.0,
      "step": 1850
    },
    {
      "entropy": 5.926247024536133,
      "epoch": 1.870398386283409,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004898765465970894,
      "loss": 5.8355,
      "mean_token_accuracy": 0.13571253269910813,
      "num_tokens": 4101409.0,
      "step": 1855
    },
    {
      "entropy": 5.943756437301635,
      "epoch": 1.8754412506303582,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004897585681582393,
      "loss": 5.9734,
      "mean_token_accuracy": 0.12765001207590104,
      "num_tokens": 4113598.0,
      "step": 1860
    },
    {
      "entropy": 5.809826135635376,
      "epoch": 1.8804841149773073,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004896399222448053,
      "loss": 5.8918,
      "mean_token_accuracy": 0.13419346362352372,
      "num_tokens": 4124639.0,
      "step": 1865
    },
    {
      "entropy": 5.998760461807251,
      "epoch": 1.8855269793242562,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004895206092255417,
      "loss": 5.9613,
      "mean_token_accuracy": 0.12527743875980377,
      "num_tokens": 4136280.0,
      "step": 1870
    },
    {
      "entropy": 5.878151702880859,
      "epoch": 1.8905698436712053,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004894006294712761,
      "loss": 5.7956,
      "mean_token_accuracy": 0.12782858684659004,
      "num_tokens": 4148251.0,
      "step": 1875
    },
    {
      "entropy": 5.898477172851562,
      "epoch": 1.8956127080181542,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004892799833549083,
      "loss": 5.8888,
      "mean_token_accuracy": 0.1343490734696388,
      "num_tokens": 4159298.0,
      "step": 1880
    },
    {
      "entropy": 5.876490736007691,
      "epoch": 1.9006555723651033,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004891586712514091,
      "loss": 5.8005,
      "mean_token_accuracy": 0.13247711658477784,
      "num_tokens": 4169974.0,
      "step": 1885
    },
    {
      "entropy": 5.7899970531463625,
      "epoch": 1.9056984367120524,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004890366935378196,
      "loss": 5.8118,
      "mean_token_accuracy": 0.13953317031264306,
      "num_tokens": 4181442.0,
      "step": 1890
    },
    {
      "entropy": 5.876431465148926,
      "epoch": 1.9107413010590015,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004889140505932492,
      "loss": 5.8458,
      "mean_token_accuracy": 0.14020064026117324,
      "num_tokens": 4191535.0,
      "step": 1895
    },
    {
      "entropy": 5.856980180740356,
      "epoch": 1.9157841654059506,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887907427988749,
      "loss": 5.8742,
      "mean_token_accuracy": 0.13323668092489244,
      "num_tokens": 4201741.0,
      "step": 1900
    },
    {
      "entropy": 5.86411485671997,
      "epoch": 1.9208270297528998,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004886667705379403,
      "loss": 5.7619,
      "mean_token_accuracy": 0.134678003937006,
      "num_tokens": 4211762.0,
      "step": 1905
    },
    {
      "entropy": 5.894043588638306,
      "epoch": 1.9258698940998489,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004885421341957541,
      "loss": 5.8586,
      "mean_token_accuracy": 0.1314342848956585,
      "num_tokens": 4222322.0,
      "step": 1910
    },
    {
      "entropy": 5.892050552368164,
      "epoch": 1.9309127584467978,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004884168341596888,
      "loss": 5.9206,
      "mean_token_accuracy": 0.13454761281609534,
      "num_tokens": 4232166.0,
      "step": 1915
    },
    {
      "entropy": 5.859141111373901,
      "epoch": 1.9359556227937469,
      "grad_norm": 1.0,
      "learning_rate": 0.0004882908708191797,
      "loss": 5.8826,
      "mean_token_accuracy": 0.1334269478917122,
      "num_tokens": 4243651.0,
      "step": 1920
    },
    {
      "entropy": 5.893623161315918,
      "epoch": 1.9409984871406958,
      "grad_norm": 0.875,
      "learning_rate": 0.00048816424456572403,
      "loss": 5.7962,
      "mean_token_accuracy": 0.13634647578001022,
      "num_tokens": 4255137.0,
      "step": 1925
    },
    {
      "entropy": 5.759242677688599,
      "epoch": 1.946041351487645,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048803695579287905,
      "loss": 5.8057,
      "mean_token_accuracy": 0.13784050345420837,
      "num_tokens": 4264904.0,
      "step": 1930
    },
    {
      "entropy": 5.833326148986816,
      "epoch": 1.951084215834594,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048790900489626104,
      "loss": 5.7825,
      "mean_token_accuracy": 0.13507805094122888,
      "num_tokens": 4275305.0,
      "step": 1935
    },
    {
      "entropy": 5.866985082626343,
      "epoch": 1.9561270801815431,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004877803922735445,
      "loss": 5.8532,
      "mean_token_accuracy": 0.13358560726046562,
      "num_tokens": 4287038.0,
      "step": 1940
    },
    {
      "entropy": 5.980012845993042,
      "epoch": 1.9611699445284922,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004876511183244603,
      "loss": 5.9843,
      "mean_token_accuracy": 0.12907288521528243,
      "num_tokens": 4298668.0,
      "step": 1945
    },
    {
      "entropy": 5.7747985363006595,
      "epoch": 1.9662128088754414,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048752118345079496,
      "loss": 5.7271,
      "mean_token_accuracy": 0.13709074035286903,
      "num_tokens": 4309429.0,
      "step": 1950
    },
    {
      "entropy": 5.872978210449219,
      "epoch": 1.9712556732223905,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048739058805638905,
      "loss": 5.8107,
      "mean_token_accuracy": 0.1377664566040039,
      "num_tokens": 4321137.0,
      "step": 1955
    },
    {
      "entropy": 5.855399703979492,
      "epoch": 1.9762985375693394,
      "grad_norm": 1.0,
      "learning_rate": 0.000487259332547136,
      "loss": 5.8704,
      "mean_token_accuracy": 0.12618953958153725,
      "num_tokens": 4331249.0,
      "step": 1960
    },
    {
      "entropy": 5.832701635360718,
      "epoch": 1.9813414019162885,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000487127417330981,
      "loss": 5.8406,
      "mean_token_accuracy": 0.1293475955724716,
      "num_tokens": 4342154.0,
      "step": 1965
    },
    {
      "entropy": 5.907890939712525,
      "epoch": 1.9863842662632374,
      "grad_norm": 1.0,
      "learning_rate": 0.0004869948428179196,
      "loss": 5.891,
      "mean_token_accuracy": 0.12837604135274888,
      "num_tokens": 4353906.0,
      "step": 1970
    },
    {
      "entropy": 5.861940050125122,
      "epoch": 1.9914271306101865,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004868616094199964,
      "loss": 5.9432,
      "mean_token_accuracy": 0.12661690264940262,
      "num_tokens": 4366416.0,
      "step": 1975
    },
    {
      "entropy": 5.905925703048706,
      "epoch": 1.9964699949571356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000486727717551304,
      "loss": 5.8438,
      "mean_token_accuracy": 0.13471274226903915,
      "num_tokens": 4376743.0,
      "step": 1980
    },
    {
      "entropy": 5.746378580729167,
      "epoch": 2.00100857286939,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004865931676279813,
      "loss": 5.6089,
      "mean_token_accuracy": 0.14421919816070133,
      "num_tokens": 4385379.0,
      "step": 1985
    },
    {
      "entropy": 5.844557476043701,
      "epoch": 2.006051437216339,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004864579600682125,
      "loss": 5.5372,
      "mean_token_accuracy": 0.14184438735246657,
      "num_tokens": 4396872.0,
      "step": 1990
    },
    {
      "entropy": 5.80364499092102,
      "epoch": 2.011094301563288,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048632209529222607,
      "loss": 5.4561,
      "mean_token_accuracy": 0.14693821966648102,
      "num_tokens": 4407075.0,
      "step": 1995
    },
    {
      "entropy": 5.741332960128784,
      "epoch": 2.016137165910237,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048618557372229253,
      "loss": 5.4841,
      "mean_token_accuracy": 0.1523932471871376,
      "num_tokens": 4418020.0,
      "step": 2000
    },
    {
      "epoch": 2.016137165910237,
      "eval_entropy": 5.680995989695154,
      "eval_loss": 5.989192962646484,
      "eval_mean_token_accuracy": 0.13356172469736055,
      "eval_num_tokens": 4418020.0,
      "eval_runtime": 2.6495,
      "eval_samples_per_second": 1184.365,
      "eval_steps_per_second": 148.329,
      "step": 2000
    },
    {
      "entropy": 5.766691541671753,
      "epoch": 2.0211800302571863,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004860483957827243,
      "loss": 5.5215,
      "mean_token_accuracy": 0.14789806455373763,
      "num_tokens": 4429740.0,
      "step": 2005
    },
    {
      "entropy": 5.771888685226441,
      "epoch": 2.026222894604135,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859105618998735,
      "loss": 5.4892,
      "mean_token_accuracy": 0.14458957314491272,
      "num_tokens": 4440183.0,
      "step": 2010
    },
    {
      "entropy": 5.785090827941895,
      "epoch": 2.031265758951084,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004857720725021311,
      "loss": 5.4873,
      "mean_token_accuracy": 0.14502768814563752,
      "num_tokens": 4451750.0,
      "step": 2015
    },
    {
      "entropy": 5.645265007019043,
      "epoch": 2.036308623298033,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048563292801992534,
      "loss": 5.4364,
      "mean_token_accuracy": 0.1487667128443718,
      "num_tokens": 4462071.0,
      "step": 2020
    },
    {
      "entropy": 5.825870656967163,
      "epoch": 2.0413514876449823,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004854931288857203,
      "loss": 5.6289,
      "mean_token_accuracy": 0.13642685860395432,
      "num_tokens": 4474241.0,
      "step": 2025
    },
    {
      "entropy": 5.759660863876343,
      "epoch": 2.0463943519919314,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048535267553401505,
      "loss": 5.4615,
      "mean_token_accuracy": 0.1518971547484398,
      "num_tokens": 4487083.0,
      "step": 2030
    },
    {
      "entropy": 5.708598232269287,
      "epoch": 2.0514372163388805,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048521156840134193,
      "loss": 5.5262,
      "mean_token_accuracy": 0.1527533784508705,
      "num_tokens": 4497181.0,
      "step": 2035
    },
    {
      "entropy": 5.782727146148682,
      "epoch": 2.0564800806858297,
      "grad_norm": 1.046875,
      "learning_rate": 0.000485069807926265,
      "loss": 5.4769,
      "mean_token_accuracy": 0.14840264469385148,
      "num_tokens": 4508250.0,
      "step": 2040
    },
    {
      "entropy": 5.71816668510437,
      "epoch": 2.0615229450327788,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004849273945493792,
      "loss": 5.4992,
      "mean_token_accuracy": 0.1487680435180664,
      "num_tokens": 4519778.0,
      "step": 2045
    },
    {
      "entropy": 5.685887575149536,
      "epoch": 2.066565809379728,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004847843287133084,
      "loss": 5.4897,
      "mean_token_accuracy": 0.14126669764518737,
      "num_tokens": 4530879.0,
      "step": 2050
    },
    {
      "entropy": 5.731546783447266,
      "epoch": 2.0716086737266766,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004846406108627048,
      "loss": 5.5504,
      "mean_token_accuracy": 0.15017888098955154,
      "num_tokens": 4542560.0,
      "step": 2055
    },
    {
      "entropy": 5.62994155883789,
      "epoch": 2.0766515380736257,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048449624144424665,
      "loss": 5.4279,
      "mean_token_accuracy": 0.15739631205797194,
      "num_tokens": 4552213.0,
      "step": 2060
    },
    {
      "entropy": 5.6950929164886475,
      "epoch": 2.081694402420575,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048435122090663744,
      "loss": 5.4558,
      "mean_token_accuracy": 0.15148370563983918,
      "num_tokens": 4562798.0,
      "step": 2065
    },
    {
      "entropy": 5.685782098770142,
      "epoch": 2.086737266767524,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048420554970060424,
      "loss": 5.5126,
      "mean_token_accuracy": 0.14563852846622466,
      "num_tokens": 4574534.0,
      "step": 2070
    },
    {
      "entropy": 5.702060794830322,
      "epoch": 2.091780131114473,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048405922827889677,
      "loss": 5.4945,
      "mean_token_accuracy": 0.13938834443688392,
      "num_tokens": 4586239.0,
      "step": 2075
    },
    {
      "entropy": 5.6921937465667725,
      "epoch": 2.096822995461422,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048391225709628515,
      "loss": 5.5173,
      "mean_token_accuracy": 0.14917422384023665,
      "num_tokens": 4598018.0,
      "step": 2080
    },
    {
      "entropy": 5.688322830200195,
      "epoch": 2.1018658598083713,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004837646366095593,
      "loss": 5.497,
      "mean_token_accuracy": 0.14469251781702042,
      "num_tokens": 4608985.0,
      "step": 2085
    },
    {
      "entropy": 5.727775382995605,
      "epoch": 2.1069087241553204,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048361636727752716,
      "loss": 5.5676,
      "mean_token_accuracy": 0.14773706644773482,
      "num_tokens": 4621103.0,
      "step": 2090
    },
    {
      "entropy": 5.673424577713012,
      "epoch": 2.1119515885022695,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004834674495610131,
      "loss": 5.5367,
      "mean_token_accuracy": 0.14560986086726188,
      "num_tokens": 4632244.0,
      "step": 2095
    },
    {
      "entropy": 5.713310432434082,
      "epoch": 2.116994452849218,
      "grad_norm": 1.125,
      "learning_rate": 0.00048331788392285693,
      "loss": 5.4587,
      "mean_token_accuracy": 0.14869346618652343,
      "num_tokens": 4642387.0,
      "step": 2100
    },
    {
      "entropy": 5.717536687850952,
      "epoch": 2.1220373171961673,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048316767082791205,
      "loss": 5.5909,
      "mean_token_accuracy": 0.14217522516846656,
      "num_tokens": 4655181.0,
      "step": 2105
    },
    {
      "entropy": 5.641996097564697,
      "epoch": 2.1270801815431164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048301681074304427,
      "loss": 5.4902,
      "mean_token_accuracy": 0.14716131538152694,
      "num_tokens": 4665664.0,
      "step": 2110
    },
    {
      "entropy": 5.635302257537842,
      "epoch": 2.1321230458900655,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048286530413713016,
      "loss": 5.458,
      "mean_token_accuracy": 0.15459446460008622,
      "num_tokens": 4677683.0,
      "step": 2115
    },
    {
      "entropy": 5.712548017501831,
      "epoch": 2.1371659102370146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048271315148105574,
      "loss": 5.4957,
      "mean_token_accuracy": 0.14238945618271828,
      "num_tokens": 4688891.0,
      "step": 2120
    },
    {
      "entropy": 5.68415412902832,
      "epoch": 2.1422087745839637,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000482560353247715,
      "loss": 5.5419,
      "mean_token_accuracy": 0.14890266060829163,
      "num_tokens": 4700765.0,
      "step": 2125
    },
    {
      "entropy": 5.709896850585937,
      "epoch": 2.147251638930913,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004824069099120084,
      "loss": 5.5606,
      "mean_token_accuracy": 0.14066436737775803,
      "num_tokens": 4712004.0,
      "step": 2130
    },
    {
      "entropy": 5.670287656784057,
      "epoch": 2.152294503277862,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004822528219508413,
      "loss": 5.446,
      "mean_token_accuracy": 0.1466579720377922,
      "num_tokens": 4722427.0,
      "step": 2135
    },
    {
      "entropy": 5.709214210510254,
      "epoch": 2.157337367624811,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004820980898431227,
      "loss": 5.5448,
      "mean_token_accuracy": 0.1513879381120205,
      "num_tokens": 4733815.0,
      "step": 2140
    },
    {
      "entropy": 5.6729559898376465,
      "epoch": 2.1623802319717598,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004819427140697635,
      "loss": 5.5056,
      "mean_token_accuracy": 0.1432117849588394,
      "num_tokens": 4745070.0,
      "step": 2145
    },
    {
      "entropy": 5.68441686630249,
      "epoch": 2.167423096318709,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004817866951136753,
      "loss": 5.5624,
      "mean_token_accuracy": 0.13918726816773414,
      "num_tokens": 4756066.0,
      "step": 2150
    },
    {
      "entropy": 5.7714673519134525,
      "epoch": 2.172465960665658,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048163003345976847,
      "loss": 5.5832,
      "mean_token_accuracy": 0.14053326919674874,
      "num_tokens": 4767153.0,
      "step": 2155
    },
    {
      "entropy": 5.683536720275879,
      "epoch": 2.177508825012607,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004814727295949512,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1484683021903038,
      "num_tokens": 4777554.0,
      "step": 2160
    },
    {
      "entropy": 5.639589738845825,
      "epoch": 2.1825516893595562,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048131478400812735,
      "loss": 5.5327,
      "mean_token_accuracy": 0.1460942767560482,
      "num_tokens": 4788881.0,
      "step": 2165
    },
    {
      "entropy": 5.701530885696411,
      "epoch": 2.1875945537065054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004811561971901956,
      "loss": 5.4979,
      "mean_token_accuracy": 0.14612958282232286,
      "num_tokens": 4798680.0,
      "step": 2170
    },
    {
      "entropy": 5.667514801025391,
      "epoch": 2.1926374180534545,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004809969696340474,
      "loss": 5.5044,
      "mean_token_accuracy": 0.15070030689239503,
      "num_tokens": 4808096.0,
      "step": 2175
    },
    {
      "entropy": 5.670957708358765,
      "epoch": 2.1976802824004036,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048083710183456554,
      "loss": 5.4629,
      "mean_token_accuracy": 0.15089642405509948,
      "num_tokens": 4818778.0,
      "step": 2180
    },
    {
      "entropy": 5.57321720123291,
      "epoch": 2.2027231467473527,
      "grad_norm": 1.125,
      "learning_rate": 0.00048067659428862306,
      "loss": 5.4543,
      "mean_token_accuracy": 0.15111969113349916,
      "num_tokens": 4828586.0,
      "step": 2185
    },
    {
      "entropy": 5.672693300247192,
      "epoch": 2.2077660110943014,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048051544749508096,
      "loss": 5.5658,
      "mean_token_accuracy": 0.14453586041927338,
      "num_tokens": 4839419.0,
      "step": 2190
    },
    {
      "entropy": 5.666914653778076,
      "epoch": 2.2128088754412505,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004803536619547872,
      "loss": 5.4861,
      "mean_token_accuracy": 0.15151002258062363,
      "num_tokens": 4849700.0,
      "step": 2195
    },
    {
      "entropy": 5.6714146614074705,
      "epoch": 2.2178517397881996,
      "grad_norm": 1.015625,
      "learning_rate": 0.000480191238170575,
      "loss": 5.6584,
      "mean_token_accuracy": 0.13440413549542427,
      "num_tokens": 4860261.0,
      "step": 2200
    },
    {
      "entropy": 5.758241844177246,
      "epoch": 2.2228946041351487,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004800281766472612,
      "loss": 5.5626,
      "mean_token_accuracy": 0.14456569850444795,
      "num_tokens": 4872277.0,
      "step": 2205
    },
    {
      "entropy": 5.672854328155518,
      "epoch": 2.227937468482098,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047986447789164486,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1488087370991707,
      "num_tokens": 4883512.0,
      "step": 2210
    },
    {
      "entropy": 5.6741656303405765,
      "epoch": 2.232980332829047,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004797001424125054,
      "loss": 5.5327,
      "mean_token_accuracy": 0.14639116078615189,
      "num_tokens": 4896197.0,
      "step": 2215
    },
    {
      "entropy": 5.732757711410523,
      "epoch": 2.238023197175996,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047953517072060124,
      "loss": 5.4931,
      "mean_token_accuracy": 0.15128834694623947,
      "num_tokens": 4907397.0,
      "step": 2220
    },
    {
      "entropy": 5.637088871002197,
      "epoch": 2.243066061522945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004793695633286684,
      "loss": 5.5785,
      "mean_token_accuracy": 0.1417090468108654,
      "num_tokens": 4919008.0,
      "step": 2225
    },
    {
      "entropy": 5.6243401050567625,
      "epoch": 2.2481089258698943,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047920332075141825,
      "loss": 5.4902,
      "mean_token_accuracy": 0.14776550382375717,
      "num_tokens": 4930877.0,
      "step": 2230
    },
    {
      "entropy": 5.6722338676452635,
      "epoch": 2.253151790216843,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004790364435055367,
      "loss": 5.3959,
      "mean_token_accuracy": 0.15714615434408188,
      "num_tokens": 4942274.0,
      "step": 2235
    },
    {
      "entropy": 5.6240647315979,
      "epoch": 2.258194654563792,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000478868932109682,
      "loss": 5.491,
      "mean_token_accuracy": 0.14969139397144318,
      "num_tokens": 4953740.0,
      "step": 2240
    },
    {
      "entropy": 5.6926532745361325,
      "epoch": 2.263237518910741,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004787007870844835,
      "loss": 5.5324,
      "mean_token_accuracy": 0.14889150112867355,
      "num_tokens": 4964760.0,
      "step": 2245
    },
    {
      "entropy": 5.6714859962463375,
      "epoch": 2.2682803832576903,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047853200895253976,
      "loss": 5.5913,
      "mean_token_accuracy": 0.1424285888671875,
      "num_tokens": 4976404.0,
      "step": 2250
    },
    {
      "entropy": 5.653643035888672,
      "epoch": 2.2733232476046394,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.4754,
      "mean_token_accuracy": 0.14482759982347487,
      "num_tokens": 4987552.0,
      "step": 2255
    },
    {
      "entropy": 5.630600309371948,
      "epoch": 2.2783661119515886,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047819255546864797,
      "loss": 5.487,
      "mean_token_accuracy": 0.1477342277765274,
      "num_tokens": 4998912.0,
      "step": 2260
    },
    {
      "entropy": 5.682914304733276,
      "epoch": 2.2834089762985377,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004780218811717293,
      "loss": 5.4651,
      "mean_token_accuracy": 0.15380304604768752,
      "num_tokens": 5009556.0,
      "step": 2265
    },
    {
      "entropy": 5.624031257629395,
      "epoch": 2.288451840645487,
      "grad_norm": 1.125,
      "learning_rate": 0.00047785057587812057,
      "loss": 5.5095,
      "mean_token_accuracy": 0.14924681186676025,
      "num_tokens": 5021569.0,
      "step": 2270
    },
    {
      "entropy": 5.7075849056243895,
      "epoch": 2.293494704992436,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047767864012024266,
      "loss": 5.58,
      "mean_token_accuracy": 0.14542651996016503,
      "num_tokens": 5032994.0,
      "step": 2275
    },
    {
      "entropy": 5.7110589981079105,
      "epoch": 2.2985375693393846,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047750607443247585,
      "loss": 5.5193,
      "mean_token_accuracy": 0.14738448411226274,
      "num_tokens": 5044681.0,
      "step": 2280
    },
    {
      "entropy": 5.618875551223755,
      "epoch": 2.3035804336863337,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047733287935115837,
      "loss": 5.4435,
      "mean_token_accuracy": 0.15486137419939042,
      "num_tokens": 5055509.0,
      "step": 2285
    },
    {
      "entropy": 5.519594621658325,
      "epoch": 2.308623298033283,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004771590554145844,
      "loss": 5.4787,
      "mean_token_accuracy": 0.15279970616102218,
      "num_tokens": 5066903.0,
      "step": 2290
    },
    {
      "entropy": 5.660610771179199,
      "epoch": 2.313666162380232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047698460316300285,
      "loss": 5.5231,
      "mean_token_accuracy": 0.14691229909658432,
      "num_tokens": 5078411.0,
      "step": 2295
    },
    {
      "entropy": 5.646321582794189,
      "epoch": 2.318709026727181,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004768095231386153,
      "loss": 5.5308,
      "mean_token_accuracy": 0.14467755407094957,
      "num_tokens": 5089811.0,
      "step": 2300
    },
    {
      "entropy": 5.688470268249512,
      "epoch": 2.32375189107413,
      "grad_norm": 1.0,
      "learning_rate": 0.0004766338158855744,
      "loss": 5.5323,
      "mean_token_accuracy": 0.15395605564117432,
      "num_tokens": 5101196.0,
      "step": 2305
    },
    {
      "entropy": 5.576445579528809,
      "epoch": 2.3287947554210793,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004764574819499826,
      "loss": 5.5122,
      "mean_token_accuracy": 0.150606869161129,
      "num_tokens": 5112379.0,
      "step": 2310
    },
    {
      "entropy": 5.704416084289551,
      "epoch": 2.3338376197680284,
      "grad_norm": 1.0,
      "learning_rate": 0.00047628052187988955,
      "loss": 5.6262,
      "mean_token_accuracy": 0.13457250222563744,
      "num_tokens": 5125065.0,
      "step": 2315
    },
    {
      "entropy": 5.766238880157471,
      "epoch": 2.338880484114977,
      "grad_norm": 1.0,
      "learning_rate": 0.0004761029362252914,
      "loss": 5.5993,
      "mean_token_accuracy": 0.14355267137289046,
      "num_tokens": 5137471.0,
      "step": 2320
    },
    {
      "entropy": 5.736388301849365,
      "epoch": 2.343923348461926,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047592472553812856,
      "loss": 5.5429,
      "mean_token_accuracy": 0.15075437873601913,
      "num_tokens": 5147682.0,
      "step": 2325
    },
    {
      "entropy": 5.644363355636597,
      "epoch": 2.3489662128088753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004757458903722837,
      "loss": 5.5152,
      "mean_token_accuracy": 0.14494604468345643,
      "num_tokens": 5156143.0,
      "step": 2330
    },
    {
      "entropy": 5.69903416633606,
      "epoch": 2.3540090771558244,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047556643128358075,
      "loss": 5.5571,
      "mean_token_accuracy": 0.14494485259056092,
      "num_tokens": 5167462.0,
      "step": 2335
    },
    {
      "entropy": 5.586874771118164,
      "epoch": 2.3590519415027735,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047538634882978277,
      "loss": 5.4811,
      "mean_token_accuracy": 0.1485794246196747,
      "num_tokens": 5178030.0,
      "step": 2340
    },
    {
      "entropy": 5.604873609542847,
      "epoch": 2.3640948058497226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047520564357059013,
      "loss": 5.5169,
      "mean_token_accuracy": 0.1468599297106266,
      "num_tokens": 5188296.0,
      "step": 2345
    },
    {
      "entropy": 5.548481798171997,
      "epoch": 2.3691376701966718,
      "grad_norm": 1.0,
      "learning_rate": 0.0004750243160676389,
      "loss": 5.4911,
      "mean_token_accuracy": 0.1502973645925522,
      "num_tokens": 5199058.0,
      "step": 2350
    },
    {
      "entropy": 5.66943097114563,
      "epoch": 2.374180534543621,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004748423668844992,
      "loss": 5.4603,
      "mean_token_accuracy": 0.14912888109683992,
      "num_tokens": 5209314.0,
      "step": 2355
    },
    {
      "entropy": 5.676968717575074,
      "epoch": 2.37922339889057,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004746597965866733,
      "loss": 5.588,
      "mean_token_accuracy": 0.14606355130672455,
      "num_tokens": 5220248.0,
      "step": 2360
    },
    {
      "entropy": 5.6687804698944095,
      "epoch": 2.384266263237519,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004744766057415938,
      "loss": 5.546,
      "mean_token_accuracy": 0.13689773082733153,
      "num_tokens": 5230874.0,
      "step": 2365
    },
    {
      "entropy": 5.654182243347168,
      "epoch": 2.389309127584468,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004742927949186222,
      "loss": 5.5029,
      "mean_token_accuracy": 0.15400156527757644,
      "num_tokens": 5243328.0,
      "step": 2370
    },
    {
      "entropy": 5.560733842849731,
      "epoch": 2.394351991931417,
      "grad_norm": 1.25,
      "learning_rate": 0.00047410836468904664,
      "loss": 5.4458,
      "mean_token_accuracy": 0.14656009376049042,
      "num_tokens": 5254357.0,
      "step": 2375
    },
    {
      "entropy": 5.561968517303467,
      "epoch": 2.399394856278366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047392331562608073,
      "loss": 5.4535,
      "mean_token_accuracy": 0.15026493668556212,
      "num_tokens": 5264492.0,
      "step": 2380
    },
    {
      "entropy": 5.575039291381836,
      "epoch": 2.404437720625315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004737376483048609,
      "loss": 5.4142,
      "mean_token_accuracy": 0.15389057099819184,
      "num_tokens": 5274816.0,
      "step": 2385
    },
    {
      "entropy": 5.613295602798462,
      "epoch": 2.4094805849722642,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004735513633024458,
      "loss": 5.5165,
      "mean_token_accuracy": 0.1459888331592083,
      "num_tokens": 5286855.0,
      "step": 2390
    },
    {
      "entropy": 5.683413743972778,
      "epoch": 2.4145234493192134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004733644611978134,
      "loss": 5.6262,
      "mean_token_accuracy": 0.13750532865524293,
      "num_tokens": 5299594.0,
      "step": 2395
    },
    {
      "entropy": 5.579679918289185,
      "epoch": 2.4195663136661625,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004731769425718597,
      "loss": 5.4423,
      "mean_token_accuracy": 0.1451920449733734,
      "num_tokens": 5310078.0,
      "step": 2400
    },
    {
      "entropy": 5.584488201141357,
      "epoch": 2.4246091780131116,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000472988808007397,
      "loss": 5.4762,
      "mean_token_accuracy": 0.15450869500637054,
      "num_tokens": 5319544.0,
      "step": 2405
    },
    {
      "entropy": 5.617080354690552,
      "epoch": 2.4296520423600603,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047280005808915176,
      "loss": 5.5489,
      "mean_token_accuracy": 0.14544432386755943,
      "num_tokens": 5331025.0,
      "step": 2410
    },
    {
      "entropy": 5.547574234008789,
      "epoch": 2.4346949067070094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004726106934037631,
      "loss": 5.5234,
      "mean_token_accuracy": 0.15173332393169403,
      "num_tokens": 5343086.0,
      "step": 2415
    },
    {
      "entropy": 5.712149143218994,
      "epoch": 2.4397377710539585,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.5559,
      "mean_token_accuracy": 0.15391047894954682,
      "num_tokens": 5353598.0,
      "step": 2420
    },
    {
      "entropy": 5.608832216262817,
      "epoch": 2.4447806354009076,
      "grad_norm": 1.125,
      "learning_rate": 0.00047223012208766384,
      "loss": 5.4557,
      "mean_token_accuracy": 0.14823706597089767,
      "num_tokens": 5363607.0,
      "step": 2425
    },
    {
      "entropy": 5.574588060379028,
      "epoch": 2.4498234997478567,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004720389166397777,
      "loss": 5.5194,
      "mean_token_accuracy": 0.14685906171798707,
      "num_tokens": 5375198.0,
      "step": 2430
    },
    {
      "entropy": 5.589098119735718,
      "epoch": 2.454866364094806,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047184709879039325,
      "loss": 5.4897,
      "mean_token_accuracy": 0.1493762269616127,
      "num_tokens": 5386058.0,
      "step": 2435
    },
    {
      "entropy": 5.667830181121826,
      "epoch": 2.459909228441755,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047165466913568497,
      "loss": 5.5167,
      "mean_token_accuracy": 0.14955587536096573,
      "num_tokens": 5395720.0,
      "step": 2440
    },
    {
      "entropy": 5.5903956413269045,
      "epoch": 2.464952092788704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004714616282737285,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1448132574558258,
      "num_tokens": 5406252.0,
      "step": 2445
    },
    {
      "entropy": 5.6490295886993405,
      "epoch": 2.469994957135653,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004712679768044995,
      "loss": 5.5611,
      "mean_token_accuracy": 0.14693916216492653,
      "num_tokens": 5418125.0,
      "step": 2450
    },
    {
      "entropy": 5.699648809432984,
      "epoch": 2.4750378214826023,
      "grad_norm": 1.09375,
      "learning_rate": 0.000471073715329871,
      "loss": 5.5317,
      "mean_token_accuracy": 0.15424191057682038,
      "num_tokens": 5428095.0,
      "step": 2455
    },
    {
      "entropy": 5.671195602416992,
      "epoch": 2.4800806858295514,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004708788444536123,
      "loss": 5.5417,
      "mean_token_accuracy": 0.14158782064914704,
      "num_tokens": 5438359.0,
      "step": 2460
    },
    {
      "entropy": 5.6120024681091305,
      "epoch": 2.4851235501765,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004706833647813865,
      "loss": 5.4734,
      "mean_token_accuracy": 0.15623558014631272,
      "num_tokens": 5448559.0,
      "step": 2465
    },
    {
      "entropy": 5.546583366394043,
      "epoch": 2.4901664145234492,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047048727692074893,
      "loss": 5.4917,
      "mean_token_accuracy": 0.14745179563760757,
      "num_tokens": 5459739.0,
      "step": 2470
    },
    {
      "entropy": 5.572244501113891,
      "epoch": 2.4952092788703983,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004702905814811452,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1457386389374733,
      "num_tokens": 5471160.0,
      "step": 2475
    },
    {
      "entropy": 5.528280639648438,
      "epoch": 2.5002521432173475,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00047009327907390926,
      "loss": 5.5297,
      "mean_token_accuracy": 0.14906351193785666,
      "num_tokens": 5482344.0,
      "step": 2480
    },
    {
      "entropy": 5.677863693237304,
      "epoch": 2.5052950075642966,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004698953703122616,
      "loss": 5.5462,
      "mean_token_accuracy": 0.14721935838460923,
      "num_tokens": 5493111.0,
      "step": 2485
    },
    {
      "entropy": 5.483181524276733,
      "epoch": 2.5103378719112457,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046969685581130715,
      "loss": 5.466,
      "mean_token_accuracy": 0.1502784177660942,
      "num_tokens": 5503877.0,
      "step": 2490
    },
    {
      "entropy": 5.6861711025238035,
      "epoch": 2.515380736258195,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046949773618803353,
      "loss": 5.4765,
      "mean_token_accuracy": 0.14919823929667472,
      "num_tokens": 5514067.0,
      "step": 2495
    },
    {
      "entropy": 5.66523585319519,
      "epoch": 2.5204236006051435,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046929801206130925,
      "loss": 5.5121,
      "mean_token_accuracy": 0.14804998338222503,
      "num_tokens": 5525534.0,
      "step": 2500
    },
    {
      "epoch": 2.5204236006051435,
      "eval_entropy": 5.525525642715338,
      "eval_loss": 5.9093546867370605,
      "eval_mean_token_accuracy": 0.1372140715936668,
      "eval_num_tokens": 5525534.0,
      "eval_runtime": 2.8743,
      "eval_samples_per_second": 1091.732,
      "eval_steps_per_second": 136.727,
      "step": 2500
    },
    {
      "entropy": 5.593466424942017,
      "epoch": 2.5254664649520926,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046909768405188135,
      "loss": 5.495,
      "mean_token_accuracy": 0.15152756124734879,
      "num_tokens": 5535525.0,
      "step": 2505
    },
    {
      "entropy": 5.551295900344849,
      "epoch": 2.5305093292990417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046889675278237404,
      "loss": 5.5033,
      "mean_token_accuracy": 0.15482114404439926,
      "num_tokens": 5546129.0,
      "step": 2510
    },
    {
      "entropy": 5.730327129364014,
      "epoch": 2.535552193645991,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004686952188772863,
      "loss": 5.6314,
      "mean_token_accuracy": 0.14667745381593705,
      "num_tokens": 5557548.0,
      "step": 2515
    },
    {
      "entropy": 5.676825952529907,
      "epoch": 2.54059505799294,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046849308296299005,
      "loss": 5.5418,
      "mean_token_accuracy": 0.15349968373775483,
      "num_tokens": 5567367.0,
      "step": 2520
    },
    {
      "entropy": 5.651474905014038,
      "epoch": 2.545637922339889,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004682903456677285,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1505977191030979,
      "num_tokens": 5577489.0,
      "step": 2525
    },
    {
      "entropy": 5.576373434066772,
      "epoch": 2.550680786686838,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046808700762161385,
      "loss": 5.5061,
      "mean_token_accuracy": 0.1442405879497528,
      "num_tokens": 5587877.0,
      "step": 2530
    },
    {
      "entropy": 5.603444576263428,
      "epoch": 2.5557236510337873,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004678830694566253,
      "loss": 5.5106,
      "mean_token_accuracy": 0.15525266379117966,
      "num_tokens": 5597863.0,
      "step": 2535
    },
    {
      "entropy": 5.572502756118775,
      "epoch": 2.5607665153807364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004676785318066075,
      "loss": 5.5096,
      "mean_token_accuracy": 0.14986335039138793,
      "num_tokens": 5608141.0,
      "step": 2540
    },
    {
      "entropy": 5.598837566375733,
      "epoch": 2.5658093797276855,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004674733953072682,
      "loss": 5.5944,
      "mean_token_accuracy": 0.1454719938337803,
      "num_tokens": 5619660.0,
      "step": 2545
    },
    {
      "entropy": 5.733735609054565,
      "epoch": 2.5708522440746346,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004672676605961763,
      "loss": 5.5688,
      "mean_token_accuracy": 0.1414414018392563,
      "num_tokens": 5630271.0,
      "step": 2550
    },
    {
      "entropy": 5.620486211776734,
      "epoch": 2.5758951084215833,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046706132831276007,
      "loss": 5.5717,
      "mean_token_accuracy": 0.14866012409329415,
      "num_tokens": 5642594.0,
      "step": 2555
    },
    {
      "entropy": 5.5966949462890625,
      "epoch": 2.5809379727685324,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046685439909830486,
      "loss": 5.4977,
      "mean_token_accuracy": 0.15163239985704421,
      "num_tokens": 5653361.0,
      "step": 2560
    },
    {
      "entropy": 5.535309505462647,
      "epoch": 2.5859808371154815,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004666468735959516,
      "loss": 5.4287,
      "mean_token_accuracy": 0.1503990426659584,
      "num_tokens": 5664167.0,
      "step": 2565
    },
    {
      "entropy": 5.582321882247925,
      "epoch": 2.5910237014624307,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004664387524506943,
      "loss": 5.5081,
      "mean_token_accuracy": 0.14768971651792526,
      "num_tokens": 5675903.0,
      "step": 2570
    },
    {
      "entropy": 5.615547752380371,
      "epoch": 2.59606656580938,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004662300363093784,
      "loss": 5.4954,
      "mean_token_accuracy": 0.14742980152368546,
      "num_tokens": 5686319.0,
      "step": 2575
    },
    {
      "entropy": 5.685475921630859,
      "epoch": 2.601109430156329,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004660207258206983,
      "loss": 5.6048,
      "mean_token_accuracy": 0.14207042753696442,
      "num_tokens": 5696000.0,
      "step": 2580
    },
    {
      "entropy": 5.599239778518677,
      "epoch": 2.606152294503278,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.5057,
      "mean_token_accuracy": 0.14740608334541322,
      "num_tokens": 5708145.0,
      "step": 2585
    },
    {
      "entropy": 5.531191062927246,
      "epoch": 2.6111951588502267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004656003244052582,
      "loss": 5.4622,
      "mean_token_accuracy": 0.15142022371292113,
      "num_tokens": 5718053.0,
      "step": 2590
    },
    {
      "entropy": 5.606213474273682,
      "epoch": 2.616238023197176,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004653892347851155,
      "loss": 5.5347,
      "mean_token_accuracy": 0.14673714712262154,
      "num_tokens": 5728723.0,
      "step": 2595
    },
    {
      "entropy": 5.530488300323486,
      "epoch": 2.621280887544125,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004651775534308392,
      "loss": 5.4554,
      "mean_token_accuracy": 0.1564555361866951,
      "num_tokens": 5739414.0,
      "step": 2600
    },
    {
      "entropy": 5.658373689651489,
      "epoch": 2.626323751891074,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004649652810003398,
      "loss": 5.6026,
      "mean_token_accuracy": 0.14294807687401773,
      "num_tokens": 5751247.0,
      "step": 2605
    },
    {
      "entropy": 5.584903287887573,
      "epoch": 2.631366616238023,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00046475241815336504,
      "loss": 5.4593,
      "mean_token_accuracy": 0.1572932004928589,
      "num_tokens": 5761791.0,
      "step": 2610
    },
    {
      "entropy": 5.579724216461182,
      "epoch": 2.6364094805849723,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004645389655514974,
      "loss": 5.4536,
      "mean_token_accuracy": 0.1468845620751381,
      "num_tokens": 5773597.0,
      "step": 2615
    },
    {
      "entropy": 5.573400163650513,
      "epoch": 2.6414523449319214,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004643249238581525,
      "loss": 5.4724,
      "mean_token_accuracy": 0.14879051074385644,
      "num_tokens": 5782963.0,
      "step": 2620
    },
    {
      "entropy": 5.548124313354492,
      "epoch": 2.6464952092788705,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004641102937385769,
      "loss": 5.5172,
      "mean_token_accuracy": 0.144728322327137,
      "num_tokens": 5793474.0,
      "step": 2625
    },
    {
      "entropy": 5.596257877349854,
      "epoch": 2.6515380736258196,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004638950758598461,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1495629146695137,
      "num_tokens": 5804018.0,
      "step": 2630
    },
    {
      "entropy": 5.678882837295532,
      "epoch": 2.6565809379727687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046367927089086203,
      "loss": 5.526,
      "mean_token_accuracy": 0.1477683201432228,
      "num_tokens": 5815897.0,
      "step": 2635
    },
    {
      "entropy": 5.595109319686889,
      "epoch": 2.661623802319718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046346287950235163,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1455786257982254,
      "num_tokens": 5825883.0,
      "step": 2640
    },
    {
      "entropy": 5.632021903991699,
      "epoch": 2.6666666666666665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004632459023668642,
      "loss": 5.5307,
      "mean_token_accuracy": 0.14996212124824523,
      "num_tokens": 5836668.0,
      "step": 2645
    },
    {
      "entropy": 5.66527214050293,
      "epoch": 2.6717095310136156,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004630283401587699,
      "loss": 5.6498,
      "mean_token_accuracy": 0.13901092410087584,
      "num_tokens": 5848497.0,
      "step": 2650
    },
    {
      "entropy": 5.588075113296509,
      "epoch": 2.6767523953605648,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004628101935542568,
      "loss": 5.4501,
      "mean_token_accuracy": 0.15287178307771682,
      "num_tokens": 5859231.0,
      "step": 2655
    },
    {
      "entropy": 5.612195348739624,
      "epoch": 2.681795259707514,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046259146323132986,
      "loss": 5.6019,
      "mean_token_accuracy": 0.14820815920829772,
      "num_tokens": 5871190.0,
      "step": 2660
    },
    {
      "entropy": 5.575901842117309,
      "epoch": 2.686838124054463,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046237214986980776,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1501447468996048,
      "num_tokens": 5881788.0,
      "step": 2665
    },
    {
      "entropy": 5.587506818771362,
      "epoch": 2.691880988401412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004621522541513216,
      "loss": 5.4506,
      "mean_token_accuracy": 0.15079249441623688,
      "num_tokens": 5891289.0,
      "step": 2670
    },
    {
      "entropy": 5.564803504943848,
      "epoch": 2.696923852748361,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004619317767593123,
      "loss": 5.4942,
      "mean_token_accuracy": 0.14542635530233383,
      "num_tokens": 5903594.0,
      "step": 2675
    },
    {
      "entropy": 5.707402515411377,
      "epoch": 2.70196671709531,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046171071837902873,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1481708064675331,
      "num_tokens": 5914592.0,
      "step": 2680
    },
    {
      "entropy": 5.53761305809021,
      "epoch": 2.707009581442259,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004614890796975254,
      "loss": 5.4338,
      "mean_token_accuracy": 0.15292296111583709,
      "num_tokens": 5927137.0,
      "step": 2685
    },
    {
      "entropy": 5.565995931625366,
      "epoch": 2.712052445789208,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004612668614036605,
      "loss": 5.549,
      "mean_token_accuracy": 0.14784384816884993,
      "num_tokens": 5938730.0,
      "step": 2690
    },
    {
      "entropy": 5.633968448638916,
      "epoch": 2.7170953101361572,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004610440641880934,
      "loss": 5.5492,
      "mean_token_accuracy": 0.14645204693078995,
      "num_tokens": 5948912.0,
      "step": 2695
    },
    {
      "entropy": 5.572774600982666,
      "epoch": 2.7221381744831064,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046082068874328324,
      "loss": 5.4858,
      "mean_token_accuracy": 0.15176114588975906,
      "num_tokens": 5960253.0,
      "step": 2700
    },
    {
      "entropy": 5.649148559570312,
      "epoch": 2.7271810388300555,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046059673576348596,
      "loss": 5.5845,
      "mean_token_accuracy": 0.14117211401462554,
      "num_tokens": 5972615.0,
      "step": 2705
    },
    {
      "entropy": 5.6746245384216305,
      "epoch": 2.7322239031770046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046037220594475264,
      "loss": 5.5468,
      "mean_token_accuracy": 0.15068346709012986,
      "num_tokens": 5985023.0,
      "step": 2710
    },
    {
      "entropy": 5.565754175186157,
      "epoch": 2.7372667675239537,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000460147099984927,
      "loss": 5.4541,
      "mean_token_accuracy": 0.15249723494052886,
      "num_tokens": 5996445.0,
      "step": 2715
    },
    {
      "entropy": 5.593200969696045,
      "epoch": 2.742309631870903,
      "grad_norm": 1.0,
      "learning_rate": 0.0004599214185836438,
      "loss": 5.5025,
      "mean_token_accuracy": 0.14778674840927125,
      "num_tokens": 6007385.0,
      "step": 2720
    },
    {
      "entropy": 5.548349905014038,
      "epoch": 2.747352496217852,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045969516244232574,
      "loss": 5.5312,
      "mean_token_accuracy": 0.15147375911474228,
      "num_tokens": 6018174.0,
      "step": 2725
    },
    {
      "entropy": 5.5658186912536625,
      "epoch": 2.752395360564801,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045946833226418235,
      "loss": 5.4695,
      "mean_token_accuracy": 0.14571216255426406,
      "num_tokens": 6030049.0,
      "step": 2730
    },
    {
      "entropy": 5.591985177993775,
      "epoch": 2.7574382249117497,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004592409287542069,
      "loss": 5.4939,
      "mean_token_accuracy": 0.14509734213352204,
      "num_tokens": 6040155.0,
      "step": 2735
    },
    {
      "entropy": 5.527977037429809,
      "epoch": 2.762481089258699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004590129526191748,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1559046670794487,
      "num_tokens": 6050569.0,
      "step": 2740
    },
    {
      "entropy": 5.617687463760376,
      "epoch": 2.767523953605648,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004587844045676413,
      "loss": 5.4514,
      "mean_token_accuracy": 0.15295047760009767,
      "num_tokens": 6060871.0,
      "step": 2745
    },
    {
      "entropy": 5.529979562759399,
      "epoch": 2.772566817952597,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.4697,
      "mean_token_accuracy": 0.15294987559318543,
      "num_tokens": 6073045.0,
      "step": 2750
    },
    {
      "entropy": 5.5869945049285885,
      "epoch": 2.777609682299546,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045832559555817525,
      "loss": 5.5101,
      "mean_token_accuracy": 0.1468508929014206,
      "num_tokens": 6084798.0,
      "step": 2755
    },
    {
      "entropy": 5.612193822860718,
      "epoch": 2.7826525466464953,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045809533602623185,
      "loss": 5.4468,
      "mean_token_accuracy": 0.14791919142007828,
      "num_tokens": 6095831.0,
      "step": 2760
    },
    {
      "entropy": 5.583421993255615,
      "epoch": 2.7876954109934444,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004578645074297603,
      "loss": 5.4261,
      "mean_token_accuracy": 0.15392916053533554,
      "num_tokens": 6106958.0,
      "step": 2765
    },
    {
      "entropy": 5.504717397689819,
      "epoch": 2.792738275340393,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045763311048618144,
      "loss": 5.4444,
      "mean_token_accuracy": 0.15830655097961427,
      "num_tokens": 6118165.0,
      "step": 2770
    },
    {
      "entropy": 5.51776533126831,
      "epoch": 2.797781139687342,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004574011459146822,
      "loss": 5.454,
      "mean_token_accuracy": 0.1573981299996376,
      "num_tokens": 6129297.0,
      "step": 2775
    },
    {
      "entropy": 5.621445226669311,
      "epoch": 2.8028240040342913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045716861443621386,
      "loss": 5.4746,
      "mean_token_accuracy": 0.1501913473010063,
      "num_tokens": 6140071.0,
      "step": 2780
    },
    {
      "entropy": 5.585222578048706,
      "epoch": 2.8078668683812404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045693551677348967,
      "loss": 5.4878,
      "mean_token_accuracy": 0.15021382346749307,
      "num_tokens": 6150940.0,
      "step": 2785
    },
    {
      "entropy": 5.570693302154541,
      "epoch": 2.8129097327281896,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004567018536509823,
      "loss": 5.4688,
      "mean_token_accuracy": 0.14465128928422927,
      "num_tokens": 6161966.0,
      "step": 2790
    },
    {
      "entropy": 5.64210352897644,
      "epoch": 2.8179525970751387,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045646762579492245,
      "loss": 5.5381,
      "mean_token_accuracy": 0.14720866531133653,
      "num_tokens": 6174455.0,
      "step": 2795
    },
    {
      "entropy": 5.533165168762207,
      "epoch": 2.822995461422088,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045623283393329546,
      "loss": 5.458,
      "mean_token_accuracy": 0.14846094697713852,
      "num_tokens": 6184789.0,
      "step": 2800
    },
    {
      "entropy": 5.582577800750732,
      "epoch": 2.828038325769037,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045599747879584,
      "loss": 5.5512,
      "mean_token_accuracy": 0.14493268132209777,
      "num_tokens": 6195704.0,
      "step": 2805
    },
    {
      "entropy": 5.622770500183106,
      "epoch": 2.833081190115986,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004557615611140451,
      "loss": 5.434,
      "mean_token_accuracy": 0.15145912021398544,
      "num_tokens": 6206290.0,
      "step": 2810
    },
    {
      "entropy": 5.586861419677734,
      "epoch": 2.838124054462935,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004555250816211485,
      "loss": 5.5501,
      "mean_token_accuracy": 0.14370328187942505,
      "num_tokens": 6216660.0,
      "step": 2815
    },
    {
      "entropy": 5.588488483428955,
      "epoch": 2.8431669188098843,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045528804105213395,
      "loss": 5.4789,
      "mean_token_accuracy": 0.1566378131508827,
      "num_tokens": 6228005.0,
      "step": 2820
    },
    {
      "entropy": 5.521516418457031,
      "epoch": 2.848209783156833,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004550504401437289,
      "loss": 5.4483,
      "mean_token_accuracy": 0.14785733968019485,
      "num_tokens": 6239447.0,
      "step": 2825
    },
    {
      "entropy": 5.578975820541382,
      "epoch": 2.853252647503782,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00045481227963440235,
      "loss": 5.5596,
      "mean_token_accuracy": 0.14091725498437882,
      "num_tokens": 6252312.0,
      "step": 2830
    },
    {
      "entropy": 5.599402666091919,
      "epoch": 2.858295511850731,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00045457356026436295,
      "loss": 5.4868,
      "mean_token_accuracy": 0.14786000698804855,
      "num_tokens": 6264131.0,
      "step": 2835
    },
    {
      "entropy": 5.475321531295776,
      "epoch": 2.8633383761976803,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045433428277555577,
      "loss": 5.4332,
      "mean_token_accuracy": 0.15369775146245956,
      "num_tokens": 6275161.0,
      "step": 2840
    },
    {
      "entropy": 5.5388298511505125,
      "epoch": 2.8683812405446294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004540944479116609,
      "loss": 5.4705,
      "mean_token_accuracy": 0.14685434401035308,
      "num_tokens": 6285655.0,
      "step": 2845
    },
    {
      "entropy": 5.6022858142852785,
      "epoch": 2.8734241048915785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004538540564180906,
      "loss": 5.5786,
      "mean_token_accuracy": 0.15059732943773269,
      "num_tokens": 6296532.0,
      "step": 2850
    },
    {
      "entropy": 5.699536323547363,
      "epoch": 2.8784669692385276,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004536131090419871,
      "loss": 5.6197,
      "mean_token_accuracy": 0.1454712927341461,
      "num_tokens": 6308229.0,
      "step": 2855
    },
    {
      "entropy": 5.587550020217895,
      "epoch": 2.8835098335854763,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004533716065322205,
      "loss": 5.5321,
      "mean_token_accuracy": 0.15190866738557815,
      "num_tokens": 6321257.0,
      "step": 2860
    },
    {
      "entropy": 5.63558349609375,
      "epoch": 2.8885526979324254,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004531295496393861,
      "loss": 5.4879,
      "mean_token_accuracy": 0.14733844846487046,
      "num_tokens": 6332040.0,
      "step": 2865
    },
    {
      "entropy": 5.500462245941162,
      "epoch": 2.8935955622793745,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004528869391158023,
      "loss": 5.3662,
      "mean_token_accuracy": 0.15406107902526855,
      "num_tokens": 6342925.0,
      "step": 2870
    },
    {
      "entropy": 5.568462133407593,
      "epoch": 2.8986384266263236,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045264377571550834,
      "loss": 5.5423,
      "mean_token_accuracy": 0.14780747443437575,
      "num_tokens": 6354263.0,
      "step": 2875
    },
    {
      "entropy": 5.69741268157959,
      "epoch": 2.9036812909732728,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004524000601942615,
      "loss": 5.5499,
      "mean_token_accuracy": 0.1434149533510208,
      "num_tokens": 6365497.0,
      "step": 2880
    },
    {
      "entropy": 5.624714279174805,
      "epoch": 2.908724155320222,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004521557933095354,
      "loss": 5.589,
      "mean_token_accuracy": 0.14008525684475898,
      "num_tokens": 6376501.0,
      "step": 2885
    },
    {
      "entropy": 5.6361486434936525,
      "epoch": 2.913767019667171,
      "grad_norm": 1.015625,
      "learning_rate": 0.000451910975820517,
      "loss": 5.5345,
      "mean_token_accuracy": 0.14242256581783294,
      "num_tokens": 6387632.0,
      "step": 2890
    },
    {
      "entropy": 5.517276191711426,
      "epoch": 2.91880988401412,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045166560848810476,
      "loss": 5.5235,
      "mean_token_accuracy": 0.1575486272573471,
      "num_tokens": 6398813.0,
      "step": 2895
    },
    {
      "entropy": 5.664028882980347,
      "epoch": 2.9238527483610692,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045141969207490606,
      "loss": 5.4292,
      "mean_token_accuracy": 0.15005266070365905,
      "num_tokens": 6409532.0,
      "step": 2900
    },
    {
      "entropy": 5.5593585014343265,
      "epoch": 2.9288956127080183,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004511732273452347,
      "loss": 5.5639,
      "mean_token_accuracy": 0.14502508342266082,
      "num_tokens": 6420755.0,
      "step": 2905
    },
    {
      "entropy": 5.6420892715454105,
      "epoch": 2.9339384770549675,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004509262150651089,
      "loss": 5.5781,
      "mean_token_accuracy": 0.15124068707227706,
      "num_tokens": 6430662.0,
      "step": 2910
    },
    {
      "entropy": 5.609274435043335,
      "epoch": 2.938981341401916,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.4523,
      "mean_token_accuracy": 0.15559796392917633,
      "num_tokens": 6441230.0,
      "step": 2915
    },
    {
      "entropy": 5.6485778331756595,
      "epoch": 2.9440242057488653,
      "grad_norm": 1.0,
      "learning_rate": 0.0004504305509260724,
      "loss": 5.6092,
      "mean_token_accuracy": 0.1395176500082016,
      "num_tokens": 6452574.0,
      "step": 2920
    },
    {
      "entropy": 5.579356575012207,
      "epoch": 2.9490670700958144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004501819006076974,
      "loss": 5.4658,
      "mean_token_accuracy": 0.1548192694783211,
      "num_tokens": 6462562.0,
      "step": 2925
    },
    {
      "entropy": 5.535385990142823,
      "epoch": 2.9541099344427635,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044993270581993413,
      "loss": 5.4638,
      "mean_token_accuracy": 0.15147760957479478,
      "num_tokens": 6473594.0,
      "step": 2930
    },
    {
      "entropy": 5.600867652893067,
      "epoch": 2.9591527987897126,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00044968296733728587,
      "loss": 5.5028,
      "mean_token_accuracy": 0.15180351287126542,
      "num_tokens": 6484129.0,
      "step": 2935
    },
    {
      "entropy": 5.627963018417359,
      "epoch": 2.9641956631366617,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004494326859359456,
      "loss": 5.474,
      "mean_token_accuracy": 0.14665171951055528,
      "num_tokens": 6495584.0,
      "step": 2940
    },
    {
      "entropy": 5.542156982421875,
      "epoch": 2.969238527483611,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044918186239379363,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15177774727344512,
      "num_tokens": 6507872.0,
      "step": 2945
    },
    {
      "entropy": 5.545723962783813,
      "epoch": 2.9742813918305595,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004489304974903953,
      "loss": 5.4256,
      "mean_token_accuracy": 0.15348378866910933,
      "num_tokens": 6519172.0,
      "step": 2950
    },
    {
      "entropy": 5.558694553375244,
      "epoch": 2.9793242561775086,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004486785920069985,
      "loss": 5.5363,
      "mean_token_accuracy": 0.15182727128267287,
      "num_tokens": 6530151.0,
      "step": 2955
    },
    {
      "entropy": 5.6444190502166744,
      "epoch": 2.9843671205244577,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004484261467265313,
      "loss": 5.5246,
      "mean_token_accuracy": 0.14669964984059333,
      "num_tokens": 6541881.0,
      "step": 2960
    },
    {
      "entropy": 5.59735074043274,
      "epoch": 2.989409984871407,
      "grad_norm": 0.921875,
      "learning_rate": 0.00044817316243359956,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1458686277270317,
      "num_tokens": 6553390.0,
      "step": 2965
    },
    {
      "entropy": 5.582724618911743,
      "epoch": 2.994452849218356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044791963991448427,
      "loss": 5.3515,
      "mean_token_accuracy": 0.1610235646367073,
      "num_tokens": 6563754.0,
      "step": 2970
    },
    {
      "entropy": 5.524975347518921,
      "epoch": 2.999495713565305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044766557995713915,
      "loss": 5.4385,
      "mean_token_accuracy": 0.1464684471487999,
      "num_tokens": 6574124.0,
      "step": 2975
    },
    {
      "entropy": 5.584632873535156,
      "epoch": 3.0040342914775593,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004474109833511885,
      "loss": 5.2302,
      "mean_token_accuracy": 0.1594168312019772,
      "num_tokens": 6582795.0,
      "step": 2980
    },
    {
      "entropy": 5.487065649032592,
      "epoch": 3.0090771558245084,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004471558508879244,
      "loss": 5.1521,
      "mean_token_accuracy": 0.17096262723207473,
      "num_tokens": 6593649.0,
      "step": 2985
    },
    {
      "entropy": 5.500127458572388,
      "epoch": 3.0141200201714575,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004469001833603045,
      "loss": 5.1555,
      "mean_token_accuracy": 0.1699725031852722,
      "num_tokens": 6606370.0,
      "step": 2990
    },
    {
      "entropy": 5.523454761505127,
      "epoch": 3.0191628845184066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004466439815629493,
      "loss": 5.102,
      "mean_token_accuracy": 0.16770666241645812,
      "num_tokens": 6616599.0,
      "step": 2995
    },
    {
      "entropy": 5.58970365524292,
      "epoch": 3.0242057488653553,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044638724629214,
      "loss": 5.194,
      "mean_token_accuracy": 0.1603380024433136,
      "num_tokens": 6627707.0,
      "step": 3000
    },
    {
      "epoch": 3.0242057488653553,
      "eval_entropy": 5.464859989156553,
      "eval_loss": 5.837061405181885,
      "eval_mean_token_accuracy": 0.14111255103622683,
      "eval_num_tokens": 6627707.0,
      "eval_runtime": 2.6605,
      "eval_samples_per_second": 1179.48,
      "eval_steps_per_second": 147.717,
      "step": 3000
    },
    {
      "entropy": 5.508710479736328,
      "epoch": 3.0292486132123044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044612997834581557,
      "loss": 5.1061,
      "mean_token_accuracy": 0.166281121969223,
      "num_tokens": 6639001.0,
      "step": 3005
    },
    {
      "entropy": 5.460995626449585,
      "epoch": 3.0342914775592535,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044587217852357086,
      "loss": 5.1654,
      "mean_token_accuracy": 0.1642779991030693,
      "num_tokens": 6649318.0,
      "step": 3010
    },
    {
      "entropy": 5.485589694976807,
      "epoch": 3.0393343419062027,
      "grad_norm": 1.125,
      "learning_rate": 0.00044561384762665363,
      "loss": 5.2145,
      "mean_token_accuracy": 0.16070360392332078,
      "num_tokens": 6660584.0,
      "step": 3015
    },
    {
      "entropy": 5.4776569366455075,
      "epoch": 3.044377206253152,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004453549864579623,
      "loss": 5.0496,
      "mean_token_accuracy": 0.17387867718935013,
      "num_tokens": 6671361.0,
      "step": 3020
    },
    {
      "entropy": 5.524082851409912,
      "epoch": 3.049420070600101,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004450955958220433,
      "loss": 5.3019,
      "mean_token_accuracy": 0.15830719023942946,
      "num_tokens": 6681896.0,
      "step": 3025
    },
    {
      "entropy": 5.495604610443115,
      "epoch": 3.05446293494705,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044483567652508864,
      "loss": 5.1526,
      "mean_token_accuracy": 0.16557196527719498,
      "num_tokens": 6693360.0,
      "step": 3030
    },
    {
      "entropy": 5.50893988609314,
      "epoch": 3.059505799293999,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004445752293749337,
      "loss": 5.2173,
      "mean_token_accuracy": 0.16287909597158431,
      "num_tokens": 6704921.0,
      "step": 3035
    },
    {
      "entropy": 5.494605493545532,
      "epoch": 3.0645486636409482,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004443142551810541,
      "loss": 5.1528,
      "mean_token_accuracy": 0.1641039714217186,
      "num_tokens": 6715659.0,
      "step": 3040
    },
    {
      "entropy": 5.41555118560791,
      "epoch": 3.069591527987897,
      "grad_norm": 1.125,
      "learning_rate": 0.00044405275475456365,
      "loss": 5.0953,
      "mean_token_accuracy": 0.17433864623308182,
      "num_tokens": 6726431.0,
      "step": 3045
    },
    {
      "entropy": 5.4583734512329105,
      "epoch": 3.074634392334846,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004437907289082116,
      "loss": 5.1735,
      "mean_token_accuracy": 0.16899368017911912,
      "num_tokens": 6738002.0,
      "step": 3050
    },
    {
      "entropy": 5.508736610412598,
      "epoch": 3.079677256681795,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044352817845638053,
      "loss": 5.1267,
      "mean_token_accuracy": 0.1695928692817688,
      "num_tokens": 6748509.0,
      "step": 3055
    },
    {
      "entropy": 5.487685298919677,
      "epoch": 3.0847201210287443,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004432651042150833,
      "loss": 5.1785,
      "mean_token_accuracy": 0.16454828530550003,
      "num_tokens": 6759258.0,
      "step": 3060
    },
    {
      "entropy": 5.454632186889649,
      "epoch": 3.0897629853756934,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044300150700196054,
      "loss": 5.1982,
      "mean_token_accuracy": 0.16186091303825378,
      "num_tokens": 6771263.0,
      "step": 3065
    },
    {
      "entropy": 5.470084762573242,
      "epoch": 3.0948058497226425,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004427373876362786,
      "loss": 5.1603,
      "mean_token_accuracy": 0.16727634370326996,
      "num_tokens": 6779965.0,
      "step": 3070
    },
    {
      "entropy": 5.493002986907959,
      "epoch": 3.0998487140695916,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004424727469389265,
      "loss": 5.1377,
      "mean_token_accuracy": 0.16506763100624083,
      "num_tokens": 6791521.0,
      "step": 3075
    },
    {
      "entropy": 5.497962570190429,
      "epoch": 3.1048915784165407,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004422075857324137,
      "loss": 5.2057,
      "mean_token_accuracy": 0.16588185727596283,
      "num_tokens": 6803537.0,
      "step": 3080
    },
    {
      "entropy": 5.492455530166626,
      "epoch": 3.10993444276349,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004419419048408673,
      "loss": 5.2062,
      "mean_token_accuracy": 0.16200020164251328,
      "num_tokens": 6813811.0,
      "step": 3085
    },
    {
      "entropy": 5.410846853256226,
      "epoch": 3.1149773071104385,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004416757050900297,
      "loss": 5.1333,
      "mean_token_accuracy": 0.16689205914735794,
      "num_tokens": 6825422.0,
      "step": 3090
    },
    {
      "entropy": 5.604042100906372,
      "epoch": 3.1200201714573876,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004414089873072557,
      "loss": 5.2864,
      "mean_token_accuracy": 0.1600692167878151,
      "num_tokens": 6837086.0,
      "step": 3095
    },
    {
      "entropy": 5.485888719558716,
      "epoch": 3.1250630358043368,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004411417523215105,
      "loss": 5.129,
      "mean_token_accuracy": 0.17212664037942887,
      "num_tokens": 6847195.0,
      "step": 3100
    },
    {
      "entropy": 5.383255529403686,
      "epoch": 3.130105900151286,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004408740009633665,
      "loss": 5.1364,
      "mean_token_accuracy": 0.16953466981649398,
      "num_tokens": 6857492.0,
      "step": 3105
    },
    {
      "entropy": 5.480805826187134,
      "epoch": 3.135148764498235,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044060573406500117,
      "loss": 5.3231,
      "mean_token_accuracy": 0.15309304147958755,
      "num_tokens": 6869190.0,
      "step": 3110
    },
    {
      "entropy": 5.478493595123291,
      "epoch": 3.140191628845184,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004403369524601941,
      "loss": 5.1222,
      "mean_token_accuracy": 0.16348316818475722,
      "num_tokens": 6880728.0,
      "step": 3115
    },
    {
      "entropy": 5.403803443908691,
      "epoch": 3.145234493192133,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004400676569843249,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1617557257413864,
      "num_tokens": 6892043.0,
      "step": 3120
    },
    {
      "entropy": 5.464474678039551,
      "epoch": 3.1502773575390823,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043979784847436994,
      "loss": 5.171,
      "mean_token_accuracy": 0.17526663690805436,
      "num_tokens": 6903913.0,
      "step": 3125
    },
    {
      "entropy": 5.360471248626709,
      "epoch": 3.1553202218860315,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004395275277689004,
      "loss": 5.1559,
      "mean_token_accuracy": 0.1623574376106262,
      "num_tokens": 6913951.0,
      "step": 3130
    },
    {
      "entropy": 5.369834232330322,
      "epoch": 3.16036308623298,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004392566957080793,
      "loss": 5.1603,
      "mean_token_accuracy": 0.16323016285896302,
      "num_tokens": 6924762.0,
      "step": 3135
    },
    {
      "entropy": 5.4376220703125,
      "epoch": 3.1654059505799292,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043898535313365884,
      "loss": 5.1318,
      "mean_token_accuracy": 0.1645357847213745,
      "num_tokens": 6935845.0,
      "step": 3140
    },
    {
      "entropy": 5.3934015274047855,
      "epoch": 3.1704488149268784,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000438713500888978,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1633390322327614,
      "num_tokens": 6947338.0,
      "step": 3145
    },
    {
      "entropy": 5.484806156158447,
      "epoch": 3.1754916792738275,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004384411398189599,
      "loss": 5.2055,
      "mean_token_accuracy": 0.16281524300575256,
      "num_tokens": 6958084.0,
      "step": 3150
    },
    {
      "entropy": 5.450452852249145,
      "epoch": 3.1805345436207766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000438168270770109,
      "loss": 5.1426,
      "mean_token_accuracy": 0.16840344965457915,
      "num_tokens": 6968063.0,
      "step": 3155
    },
    {
      "entropy": 5.449947214126587,
      "epoch": 3.1855774079677257,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004378948945905085,
      "loss": 5.1409,
      "mean_token_accuracy": 0.1703096568584442,
      "num_tokens": 6978424.0,
      "step": 3160
    },
    {
      "entropy": 5.359123659133911,
      "epoch": 3.190620272314675,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00043762101212981805,
      "loss": 5.141,
      "mean_token_accuracy": 0.16534530520439147,
      "num_tokens": 6991067.0,
      "step": 3165
    },
    {
      "entropy": 5.39420714378357,
      "epoch": 3.195663136661624,
      "grad_norm": 1.125,
      "learning_rate": 0.00043734662423927054,
      "loss": 5.2196,
      "mean_token_accuracy": 0.16757993549108505,
      "num_tokens": 7000642.0,
      "step": 3170
    },
    {
      "entropy": 5.552215051651001,
      "epoch": 3.200706001008573,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004370717317716699,
      "loss": 5.2428,
      "mean_token_accuracy": 0.16152413934469223,
      "num_tokens": 7011305.0,
      "step": 3175
    },
    {
      "entropy": 5.489600563049317,
      "epoch": 3.205748865355522,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004367963355813882,
      "loss": 5.1791,
      "mean_token_accuracy": 0.16151108741760253,
      "num_tokens": 7021452.0,
      "step": 3180
    },
    {
      "entropy": 5.401028251647949,
      "epoch": 3.210791729702471,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043652043652436323,
      "loss": 5.2607,
      "mean_token_accuracy": 0.1565679907798767,
      "num_tokens": 7033189.0,
      "step": 3185
    },
    {
      "entropy": 5.513680505752563,
      "epoch": 3.21583459404942,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004362440354580956,
      "loss": 5.2041,
      "mean_token_accuracy": 0.16091608256101608,
      "num_tokens": 7044153.0,
      "step": 3190
    },
    {
      "entropy": 5.465910291671753,
      "epoch": 3.220877458396369,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004359671332416462,
      "loss": 5.267,
      "mean_token_accuracy": 0.15532111674547194,
      "num_tokens": 7056679.0,
      "step": 3195
    },
    {
      "entropy": 5.456948900222779,
      "epoch": 3.225920322743318,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004356897307356335,
      "loss": 5.1268,
      "mean_token_accuracy": 0.167019584774971,
      "num_tokens": 7067428.0,
      "step": 3200
    },
    {
      "entropy": 5.431487035751343,
      "epoch": 3.2309631870902673,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004354118288022309,
      "loss": 5.2424,
      "mean_token_accuracy": 0.15960654467344285,
      "num_tokens": 7079196.0,
      "step": 3205
    },
    {
      "entropy": 5.452811622619629,
      "epoch": 3.2360060514372164,
      "grad_norm": 1.125,
      "learning_rate": 0.00043513342830516405,
      "loss": 5.128,
      "mean_token_accuracy": 0.16414072811603547,
      "num_tokens": 7088936.0,
      "step": 3210
    },
    {
      "entropy": 5.419294929504394,
      "epoch": 3.2410489157841655,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000434854530109708,
      "loss": 5.2432,
      "mean_token_accuracy": 0.16197886019945146,
      "num_tokens": 7101504.0,
      "step": 3215
    },
    {
      "entropy": 5.498925828933716,
      "epoch": 3.2460917801311147,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043457513508268487,
      "loss": 5.2075,
      "mean_token_accuracy": 0.16442779302597046,
      "num_tokens": 7114121.0,
      "step": 3220
    },
    {
      "entropy": 5.336520099639893,
      "epoch": 3.2511346444780633,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043429524409246083,
      "loss": 5.1155,
      "mean_token_accuracy": 0.16350860744714737,
      "num_tokens": 7123321.0,
      "step": 3225
    },
    {
      "entropy": 5.420355033874512,
      "epoch": 3.2561775088250124,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004340148580089436,
      "loss": 5.1863,
      "mean_token_accuracy": 0.16267697066068648,
      "num_tokens": 7134145.0,
      "step": 3230
    },
    {
      "entropy": 5.557558917999268,
      "epoch": 3.2612203731719616,
      "grad_norm": 1.125,
      "learning_rate": 0.0004337339777035794,
      "loss": 5.2407,
      "mean_token_accuracy": 0.1607226997613907,
      "num_tokens": 7144727.0,
      "step": 3235
    },
    {
      "entropy": 5.338299369812011,
      "epoch": 3.2662632375189107,
      "grad_norm": 1.0625,
      "learning_rate": 0.000433452604049351,
      "loss": 5.1327,
      "mean_token_accuracy": 0.17556441724300384,
      "num_tokens": 7155531.0,
      "step": 3240
    },
    {
      "entropy": 5.39962158203125,
      "epoch": 3.27130610186586,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00043317073792077394,
      "loss": 5.1737,
      "mean_token_accuracy": 0.16877417266368866,
      "num_tokens": 7167234.0,
      "step": 3245
    },
    {
      "entropy": 5.474264240264892,
      "epoch": 3.276348966212809,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004328883801938947,
      "loss": 5.2552,
      "mean_token_accuracy": 0.15552335381507873,
      "num_tokens": 7178556.0,
      "step": 3250
    },
    {
      "entropy": 5.365787124633789,
      "epoch": 3.281391830559758,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043260553174628776,
      "loss": 5.1134,
      "mean_token_accuracy": 0.1659499853849411,
      "num_tokens": 7189074.0,
      "step": 3255
    },
    {
      "entropy": 5.473670148849488,
      "epoch": 3.286434694906707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004323221934570524,
      "loss": 5.2605,
      "mean_token_accuracy": 0.16382382214069366,
      "num_tokens": 7200567.0,
      "step": 3260
    },
    {
      "entropy": 5.479973316192627,
      "epoch": 3.2914775592536563,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043203836620681074,
      "loss": 5.2929,
      "mean_token_accuracy": 0.15617599040269853,
      "num_tokens": 7211012.0,
      "step": 3265
    },
    {
      "entropy": 5.408528900146484,
      "epoch": 3.2965204236006054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043175405087770435,
      "loss": 5.1925,
      "mean_token_accuracy": 0.16841529905796052,
      "num_tokens": 7222271.0,
      "step": 3270
    },
    {
      "entropy": 5.414000082015991,
      "epoch": 3.301563287947554,
      "grad_norm": 0.890625,
      "learning_rate": 0.00043146924835339186,
      "loss": 5.1375,
      "mean_token_accuracy": 0.16959647685289383,
      "num_tokens": 7235067.0,
      "step": 3275
    },
    {
      "entropy": 5.410538673400879,
      "epoch": 3.306606152294503,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043118395951904605,
      "loss": 5.2133,
      "mean_token_accuracy": 0.16099776178598404,
      "num_tokens": 7246159.0,
      "step": 3280
    },
    {
      "entropy": 5.528522968292236,
      "epoch": 3.3116490166414523,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004308981852613512,
      "loss": 5.2489,
      "mean_token_accuracy": 0.15983354300260544,
      "num_tokens": 7255944.0,
      "step": 3285
    },
    {
      "entropy": 5.461428928375244,
      "epoch": 3.3166918809884014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004306119264685003,
      "loss": 5.1784,
      "mean_token_accuracy": 0.16542329639196396,
      "num_tokens": 7267273.0,
      "step": 3290
    },
    {
      "entropy": 5.382791900634766,
      "epoch": 3.3217347453353505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004303251840301923,
      "loss": 5.1822,
      "mean_token_accuracy": 0.16305022090673446,
      "num_tokens": 7277622.0,
      "step": 3295
    },
    {
      "entropy": 5.427071475982666,
      "epoch": 3.3267776096822996,
      "grad_norm": 0.90625,
      "learning_rate": 0.00043003795883762933,
      "loss": 5.2207,
      "mean_token_accuracy": 0.1590432867407799,
      "num_tokens": 7288851.0,
      "step": 3300
    },
    {
      "entropy": 5.549104166030884,
      "epoch": 3.3318204740292487,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00042975025178351386,
      "loss": 5.2771,
      "mean_token_accuracy": 0.15939899832010268,
      "num_tokens": 7302175.0,
      "step": 3305
    },
    {
      "entropy": 5.412518739700317,
      "epoch": 3.336863338376198,
      "grad_norm": 1.109375,
      "learning_rate": 0.000429462063762046,
      "loss": 5.1436,
      "mean_token_accuracy": 0.17406332194805146,
      "num_tokens": 7312693.0,
      "step": 3310
    },
    {
      "entropy": 5.383169078826905,
      "epoch": 3.3419062027231465,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042917339566892095,
      "loss": 5.1945,
      "mean_token_accuracy": 0.15786968022584916,
      "num_tokens": 7323790.0,
      "step": 3315
    },
    {
      "entropy": 5.3746013164520265,
      "epoch": 3.3469490670700957,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004288842484013257,
      "loss": 5.1682,
      "mean_token_accuracy": 0.16247669309377671,
      "num_tokens": 7333797.0,
      "step": 3320
    },
    {
      "entropy": 5.412357234954834,
      "epoch": 3.3519919314170448,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004285946228579365,
      "loss": 5.2045,
      "mean_token_accuracy": 0.17014975994825363,
      "num_tokens": 7344674.0,
      "step": 3325
    },
    {
      "entropy": 5.426461029052734,
      "epoch": 3.357034795763994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004283045199389164,
      "loss": 5.2039,
      "mean_token_accuracy": 0.16485295295715333,
      "num_tokens": 7355978.0,
      "step": 3330
    },
    {
      "entropy": 5.539542484283447,
      "epoch": 3.362077660110943,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042801394054591197,
      "loss": 5.305,
      "mean_token_accuracy": 0.15951313972473144,
      "num_tokens": 7367054.0,
      "step": 3335
    },
    {
      "entropy": 5.476680278778076,
      "epoch": 3.367120524457892,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042772288558205057,
      "loss": 5.2612,
      "mean_token_accuracy": 0.15854859352111816,
      "num_tokens": 7379588.0,
      "step": 3340
    },
    {
      "entropy": 5.364924192428589,
      "epoch": 3.3721633888048412,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004274313559519377,
      "loss": 5.2333,
      "mean_token_accuracy": 0.16314974874258042,
      "num_tokens": 7392411.0,
      "step": 3345
    },
    {
      "entropy": 5.423452281951905,
      "epoch": 3.3772062531517903,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004271393525616544,
      "loss": 5.1184,
      "mean_token_accuracy": 0.164414843916893,
      "num_tokens": 7402801.0,
      "step": 3350
    },
    {
      "entropy": 5.406583690643311,
      "epoch": 3.3822491174987395,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042684687631875377,
      "loss": 5.1875,
      "mean_token_accuracy": 0.15710319727659225,
      "num_tokens": 7413084.0,
      "step": 3355
    },
    {
      "entropy": 5.37700080871582,
      "epoch": 3.3872919818456886,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042655392813225864,
      "loss": 5.1344,
      "mean_token_accuracy": 0.16645344942808152,
      "num_tokens": 7424004.0,
      "step": 3360
    },
    {
      "entropy": 5.320040845870972,
      "epoch": 3.3923348461926373,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042626050891265884,
      "loss": 5.1608,
      "mean_token_accuracy": 0.17361150085926055,
      "num_tokens": 7434461.0,
      "step": 3365
    },
    {
      "entropy": 5.450391435623169,
      "epoch": 3.3973777105395864,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004259666195719081,
      "loss": 5.21,
      "mean_token_accuracy": 0.16076228618621827,
      "num_tokens": 7445552.0,
      "step": 3370
    },
    {
      "entropy": 5.460859870910644,
      "epoch": 3.4024205748865355,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042567226102342134,
      "loss": 5.2055,
      "mean_token_accuracy": 0.1618820071220398,
      "num_tokens": 7457073.0,
      "step": 3375
    },
    {
      "entropy": 5.386827087402343,
      "epoch": 3.4074634392334846,
      "grad_norm": 1.125,
      "learning_rate": 0.0004253774341820715,
      "loss": 5.2015,
      "mean_token_accuracy": 0.1622549772262573,
      "num_tokens": 7468197.0,
      "step": 3380
    },
    {
      "entropy": 5.428202962875366,
      "epoch": 3.4125063035804337,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004250821399641874,
      "loss": 5.158,
      "mean_token_accuracy": 0.16729331761598587,
      "num_tokens": 7480751.0,
      "step": 3385
    },
    {
      "entropy": 5.393211221694946,
      "epoch": 3.417549167927383,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004247863792875502,
      "loss": 5.1787,
      "mean_token_accuracy": 0.17017877399921416,
      "num_tokens": 7492259.0,
      "step": 3390
    },
    {
      "entropy": 5.396057796478272,
      "epoch": 3.422592032274332,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042449015307139076,
      "loss": 5.2716,
      "mean_token_accuracy": 0.15931643843650817,
      "num_tokens": 7504317.0,
      "step": 3395
    },
    {
      "entropy": 5.412095546722412,
      "epoch": 3.427634896621281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042419346223638725,
      "loss": 5.1827,
      "mean_token_accuracy": 0.15693859606981278,
      "num_tokens": 7516039.0,
      "step": 3400
    },
    {
      "entropy": 5.469624376296997,
      "epoch": 3.4326777609682297,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004238963077046616,
      "loss": 5.2225,
      "mean_token_accuracy": 0.16526761502027512,
      "num_tokens": 7526095.0,
      "step": 3405
    },
    {
      "entropy": 5.359675693511963,
      "epoch": 3.437720625315179,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004235986903997767,
      "loss": 5.1686,
      "mean_token_accuracy": 0.1636862725019455,
      "num_tokens": 7537320.0,
      "step": 3410
    },
    {
      "entropy": 5.419049978256226,
      "epoch": 3.442763489662128,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042330061124673437,
      "loss": 5.2412,
      "mean_token_accuracy": 0.16660376638174057,
      "num_tokens": 7548100.0,
      "step": 3415
    },
    {
      "entropy": 5.444922924041748,
      "epoch": 3.447806354009077,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042300207117197116,
      "loss": 5.2556,
      "mean_token_accuracy": 0.15782784521579743,
      "num_tokens": 7560483.0,
      "step": 3420
    },
    {
      "entropy": 5.43258376121521,
      "epoch": 3.452849218356026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042270307110335657,
      "loss": 5.1751,
      "mean_token_accuracy": 0.16671013683080674,
      "num_tokens": 7571256.0,
      "step": 3425
    },
    {
      "entropy": 5.433862590789795,
      "epoch": 3.4578920827029753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004224036119701899,
      "loss": 5.1606,
      "mean_token_accuracy": 0.1657199591398239,
      "num_tokens": 7581716.0,
      "step": 3430
    },
    {
      "entropy": 5.387713861465454,
      "epoch": 3.4629349470499244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042210369470319694,
      "loss": 5.187,
      "mean_token_accuracy": 0.17281804531812667,
      "num_tokens": 7591932.0,
      "step": 3435
    },
    {
      "entropy": 5.377019357681275,
      "epoch": 3.4679778113968736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004218033202345274,
      "loss": 5.1854,
      "mean_token_accuracy": 0.16309025287628173,
      "num_tokens": 7602842.0,
      "step": 3440
    },
    {
      "entropy": 5.379820871353149,
      "epoch": 3.4730206757438227,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004215024894977522,
      "loss": 5.1557,
      "mean_token_accuracy": 0.1622964009642601,
      "num_tokens": 7614075.0,
      "step": 3445
    },
    {
      "entropy": 5.48748631477356,
      "epoch": 3.478063540090772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004212012034278601,
      "loss": 5.3368,
      "mean_token_accuracy": 0.15681455731391908,
      "num_tokens": 7625642.0,
      "step": 3450
    },
    {
      "entropy": 5.437462520599365,
      "epoch": 3.4831064044377205,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042089946296125513,
      "loss": 5.1879,
      "mean_token_accuracy": 0.16691929697990418,
      "num_tokens": 7636447.0,
      "step": 3455
    },
    {
      "entropy": 5.341785097122193,
      "epoch": 3.4881492687846696,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00042059726903575373,
      "loss": 5.1152,
      "mean_token_accuracy": 0.16832868605852128,
      "num_tokens": 7648684.0,
      "step": 3460
    },
    {
      "entropy": 5.3743284225463865,
      "epoch": 3.4931921331316187,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004202946225905815,
      "loss": 5.1791,
      "mean_token_accuracy": 0.16549742221832275,
      "num_tokens": 7660864.0,
      "step": 3465
    },
    {
      "entropy": 5.370916175842285,
      "epoch": 3.498234997478568,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004199915245663706,
      "loss": 5.1951,
      "mean_token_accuracy": 0.17021624594926835,
      "num_tokens": 7671939.0,
      "step": 3470
    },
    {
      "entropy": 5.557950401306153,
      "epoch": 3.503277861825517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041968797590515673,
      "loss": 5.2232,
      "mean_token_accuracy": 0.1650526136159897,
      "num_tokens": 7681536.0,
      "step": 3475
    },
    {
      "entropy": 5.446653127670288,
      "epoch": 3.508320726172466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004193839775503761,
      "loss": 5.174,
      "mean_token_accuracy": 0.1683890178799629,
      "num_tokens": 7693674.0,
      "step": 3480
    },
    {
      "entropy": 5.334004592895508,
      "epoch": 3.513363590519415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041907953044686253,
      "loss": 5.1821,
      "mean_token_accuracy": 0.171914803981781,
      "num_tokens": 7704533.0,
      "step": 3485
    },
    {
      "entropy": 5.400468206405639,
      "epoch": 3.518406454866364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004187746355408448,
      "loss": 5.2516,
      "mean_token_accuracy": 0.16183393746614455,
      "num_tokens": 7715448.0,
      "step": 3490
    },
    {
      "entropy": 5.447520923614502,
      "epoch": 3.523449319213313,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004184692937799431,
      "loss": 5.226,
      "mean_token_accuracy": 0.16580078452825547,
      "num_tokens": 7726808.0,
      "step": 3495
    },
    {
      "entropy": 5.528063488006592,
      "epoch": 3.528492183560262,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041816350611316693,
      "loss": 5.2179,
      "mean_token_accuracy": 0.15973225831985474,
      "num_tokens": 7737899.0,
      "step": 3500
    },
    {
      "epoch": 3.528492183560262,
      "eval_entropy": 5.31953394746659,
      "eval_loss": 5.8183393478393555,
      "eval_mean_token_accuracy": 0.14342421077134956,
      "eval_num_tokens": 7737899.0,
      "eval_runtime": 2.6595,
      "eval_samples_per_second": 1179.915,
      "eval_steps_per_second": 147.771,
      "step": 3500
    }
  ],
  "logging_steps": 5,
  "max_steps": 9910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2938608827648e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}