{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.016137165910237,
  "eval_steps": 500,
  "global_step": 2000,
  "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
    }
  ],
  "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": 7387990357248000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}