{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0080685829551186,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.432322406768799,
      "epoch": 0.005042864346949067,
      "grad_norm": 1.09375,
      "learning_rate": 2e-06,
      "loss": 6.8944,
      "mean_token_accuracy": 0.08793810680508614,
      "num_tokens": 10258.0,
      "step": 5
    },
    {
      "entropy": 7.435951519012451,
      "epoch": 0.010085728693898134,
      "grad_norm": 0.93359375,
      "learning_rate": 4.5e-06,
      "loss": 6.8302,
      "mean_token_accuracy": 0.08808585703372955,
      "num_tokens": 21166.0,
      "step": 10
    },
    {
      "entropy": 7.430681848526001,
      "epoch": 0.015128593040847202,
      "grad_norm": 1.0703125,
      "learning_rate": 7e-06,
      "loss": 6.9437,
      "mean_token_accuracy": 0.08197088204324246,
      "num_tokens": 31390.0,
      "step": 15
    },
    {
      "entropy": 7.407356119155883,
      "epoch": 0.020171457387796268,
      "grad_norm": 1.015625,
      "learning_rate": 9.5e-06,
      "loss": 6.7616,
      "mean_token_accuracy": 0.09650376290082932,
      "num_tokens": 40983.0,
      "step": 20
    },
    {
      "entropy": 7.418797492980957,
      "epoch": 0.025214321734745335,
      "grad_norm": 0.96875,
      "learning_rate": 1.2e-05,
      "loss": 6.9135,
      "mean_token_accuracy": 0.09085965603590011,
      "num_tokens": 52673.0,
      "step": 25
    },
    {
      "entropy": 7.349091672897339,
      "epoch": 0.030257186081694403,
      "grad_norm": 0.85546875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.8145,
      "mean_token_accuracy": 0.08841377198696136,
      "num_tokens": 64956.0,
      "step": 30
    },
    {
      "entropy": 7.354793548583984,
      "epoch": 0.03530005042864347,
      "grad_norm": 0.97265625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.8421,
      "mean_token_accuracy": 0.08988854065537452,
      "num_tokens": 76492.0,
      "step": 35
    },
    {
      "entropy": 7.330742835998535,
      "epoch": 0.040342914775592535,
      "grad_norm": 0.93359375,
      "learning_rate": 1.95e-05,
      "loss": 6.8362,
      "mean_token_accuracy": 0.08618292436003686,
      "num_tokens": 89191.0,
      "step": 40
    },
    {
      "entropy": 7.338322544097901,
      "epoch": 0.0453857791225416,
      "grad_norm": 0.94921875,
      "learning_rate": 2.2e-05,
      "loss": 6.8356,
      "mean_token_accuracy": 0.09781758561730385,
      "num_tokens": 100077.0,
      "step": 45
    },
    {
      "entropy": 7.363365173339844,
      "epoch": 0.05042864346949067,
      "grad_norm": 1.0546875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.9068,
      "mean_token_accuracy": 0.08456681296229362,
      "num_tokens": 110530.0,
      "step": 50
    },
    {
      "entropy": 7.334746885299682,
      "epoch": 0.05547150781643974,
      "grad_norm": 1.0234375,
      "learning_rate": 2.7e-05,
      "loss": 6.8925,
      "mean_token_accuracy": 0.08198079392313957,
      "num_tokens": 121113.0,
      "step": 55
    },
    {
      "entropy": 7.350234079360962,
      "epoch": 0.060514372163388806,
      "grad_norm": 0.94140625,
      "learning_rate": 2.95e-05,
      "loss": 7.055,
      "mean_token_accuracy": 0.08448217362165451,
      "num_tokens": 133912.0,
      "step": 60
    },
    {
      "entropy": 7.348407411575318,
      "epoch": 0.06555723651033787,
      "grad_norm": 1.0625,
      "learning_rate": 3.2e-05,
      "loss": 6.9755,
      "mean_token_accuracy": 0.08814884722232819,
      "num_tokens": 145322.0,
      "step": 65
    },
    {
      "entropy": 7.38183012008667,
      "epoch": 0.07060010085728693,
      "grad_norm": 0.953125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.8601,
      "mean_token_accuracy": 0.08610649928450584,
      "num_tokens": 156252.0,
      "step": 70
    },
    {
      "entropy": 7.35923204421997,
      "epoch": 0.07564296520423601,
      "grad_norm": 1.0234375,
      "learning_rate": 3.7e-05,
      "loss": 6.8284,
      "mean_token_accuracy": 0.09050894230604171,
      "num_tokens": 166622.0,
      "step": 75
    },
    {
      "entropy": 7.273055934906006,
      "epoch": 0.08068582955118507,
      "grad_norm": 0.88671875,
      "learning_rate": 3.95e-05,
      "loss": 6.6887,
      "mean_token_accuracy": 0.0989742748439312,
      "num_tokens": 177591.0,
      "step": 80
    },
    {
      "entropy": 7.254341888427734,
      "epoch": 0.08572869389813415,
      "grad_norm": 0.9140625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.7421,
      "mean_token_accuracy": 0.09370373338460922,
      "num_tokens": 187764.0,
      "step": 85
    },
    {
      "entropy": 7.251728105545044,
      "epoch": 0.0907715582450832,
      "grad_norm": 1.0,
      "learning_rate": 4.45e-05,
      "loss": 6.7251,
      "mean_token_accuracy": 0.09580637812614441,
      "num_tokens": 197810.0,
      "step": 90
    },
    {
      "entropy": 7.275364923477173,
      "epoch": 0.09581442259203228,
      "grad_norm": 1.015625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.8163,
      "mean_token_accuracy": 0.0883400447666645,
      "num_tokens": 208552.0,
      "step": 95
    },
    {
      "entropy": 7.275786542892456,
      "epoch": 0.10085728693898134,
      "grad_norm": 0.765625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.826,
      "mean_token_accuracy": 0.0916892133653164,
      "num_tokens": 221628.0,
      "step": 100
    },
    {
      "entropy": 7.29000997543335,
      "epoch": 0.1059001512859304,
      "grad_norm": 0.9453125,
      "learning_rate": 5.2e-05,
      "loss": 6.7912,
      "mean_token_accuracy": 0.09033632948994637,
      "num_tokens": 231033.0,
      "step": 105
    },
    {
      "entropy": 7.300166130065918,
      "epoch": 0.11094301563287948,
      "grad_norm": 1.03125,
      "learning_rate": 5.45e-05,
      "loss": 6.8315,
      "mean_token_accuracy": 0.08547956682741642,
      "num_tokens": 242259.0,
      "step": 110
    },
    {
      "entropy": 7.323574876785278,
      "epoch": 0.11598587997982854,
      "grad_norm": 0.87109375,
      "learning_rate": 5.7e-05,
      "loss": 6.8614,
      "mean_token_accuracy": 0.08920874819159508,
      "num_tokens": 254116.0,
      "step": 115
    },
    {
      "entropy": 7.342547845840454,
      "epoch": 0.12102874432677761,
      "grad_norm": 0.92578125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.8024,
      "mean_token_accuracy": 0.09266475886106491,
      "num_tokens": 264518.0,
      "step": 120
    },
    {
      "entropy": 7.254070568084717,
      "epoch": 0.1260716086737267,
      "grad_norm": 1.0625,
      "learning_rate": 6.2e-05,
      "loss": 6.6783,
      "mean_token_accuracy": 0.09979371726512909,
      "num_tokens": 273783.0,
      "step": 125
    },
    {
      "entropy": 7.216696405410767,
      "epoch": 0.13111447302067575,
      "grad_norm": 0.85546875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.8518,
      "mean_token_accuracy": 0.09130219370126724,
      "num_tokens": 284568.0,
      "step": 130
    },
    {
      "entropy": 7.356634569168091,
      "epoch": 0.1361573373676248,
      "grad_norm": 0.921875,
      "learning_rate": 6.7e-05,
      "loss": 6.8489,
      "mean_token_accuracy": 0.08760668933391572,
      "num_tokens": 296499.0,
      "step": 135
    },
    {
      "entropy": 7.341347980499267,
      "epoch": 0.14120020171457387,
      "grad_norm": 0.94921875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.7447,
      "mean_token_accuracy": 0.09345789924263954,
      "num_tokens": 307013.0,
      "step": 140
    },
    {
      "entropy": 7.212394428253174,
      "epoch": 0.14624306606152296,
      "grad_norm": 0.859375,
      "learning_rate": 7.2e-05,
      "loss": 6.8123,
      "mean_token_accuracy": 0.08974087685346603,
      "num_tokens": 318604.0,
      "step": 145
    },
    {
      "entropy": 7.267185878753662,
      "epoch": 0.15128593040847202,
      "grad_norm": 0.9453125,
      "learning_rate": 7.45e-05,
      "loss": 6.757,
      "mean_token_accuracy": 0.08906264528632164,
      "num_tokens": 328968.0,
      "step": 150
    },
    {
      "entropy": 7.338633060455322,
      "epoch": 0.15632879475542108,
      "grad_norm": 1.03125,
      "learning_rate": 7.7e-05,
      "loss": 6.7681,
      "mean_token_accuracy": 0.09419665113091469,
      "num_tokens": 338950.0,
      "step": 155
    },
    {
      "entropy": 7.276337623596191,
      "epoch": 0.16137165910237014,
      "grad_norm": 0.93359375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.8953,
      "mean_token_accuracy": 0.09276964589953422,
      "num_tokens": 350667.0,
      "step": 160
    },
    {
      "entropy": 7.386541271209717,
      "epoch": 0.1664145234493192,
      "grad_norm": 1.1171875,
      "learning_rate": 8.2e-05,
      "loss": 6.9239,
      "mean_token_accuracy": 0.09228215366601944,
      "num_tokens": 361781.0,
      "step": 165
    },
    {
      "entropy": 7.312812662124633,
      "epoch": 0.1714573877962683,
      "grad_norm": 0.90234375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.8138,
      "mean_token_accuracy": 0.09285491853952407,
      "num_tokens": 373410.0,
      "step": 170
    },
    {
      "entropy": 7.218026733398437,
      "epoch": 0.17650025214321735,
      "grad_norm": 0.9453125,
      "learning_rate": 8.7e-05,
      "loss": 6.7653,
      "mean_token_accuracy": 0.09918249770998955,
      "num_tokens": 385302.0,
      "step": 175
    },
    {
      "entropy": 7.297417688369751,
      "epoch": 0.1815431164901664,
      "grad_norm": 1.0859375,
      "learning_rate": 8.95e-05,
      "loss": 6.7624,
      "mean_token_accuracy": 0.09598183333873749,
      "num_tokens": 395183.0,
      "step": 180
    },
    {
      "entropy": 7.290459728240966,
      "epoch": 0.18658598083711547,
      "grad_norm": 0.9375,
      "learning_rate": 9.2e-05,
      "loss": 6.758,
      "mean_token_accuracy": 0.09613369405269623,
      "num_tokens": 405356.0,
      "step": 185
    },
    {
      "entropy": 7.29151873588562,
      "epoch": 0.19162884518406456,
      "grad_norm": 1.03125,
      "learning_rate": 9.45e-05,
      "loss": 6.7434,
      "mean_token_accuracy": 0.09359151124954224,
      "num_tokens": 415864.0,
      "step": 190
    },
    {
      "entropy": 7.1764427661895756,
      "epoch": 0.19667170953101362,
      "grad_norm": 0.9609375,
      "learning_rate": 9.7e-05,
      "loss": 6.6799,
      "mean_token_accuracy": 0.0958583302795887,
      "num_tokens": 426225.0,
      "step": 195
    },
    {
      "entropy": 7.251503944396973,
      "epoch": 0.20171457387796268,
      "grad_norm": 1.0,
      "learning_rate": 9.95e-05,
      "loss": 6.8068,
      "mean_token_accuracy": 0.09116024672985076,
      "num_tokens": 436847.0,
      "step": 200
    },
    {
      "entropy": 7.271901845932007,
      "epoch": 0.20675743822491174,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000102,
      "loss": 6.7809,
      "mean_token_accuracy": 0.09316601976752281,
      "num_tokens": 448807.0,
      "step": 205
    },
    {
      "entropy": 7.312121725082397,
      "epoch": 0.2118003025718608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.7572,
      "mean_token_accuracy": 0.09582966193556786,
      "num_tokens": 459691.0,
      "step": 210
    },
    {
      "entropy": 7.235625171661377,
      "epoch": 0.2168431669188099,
      "grad_norm": 0.90625,
      "learning_rate": 0.000107,
      "loss": 6.7575,
      "mean_token_accuracy": 0.08772426173090934,
      "num_tokens": 470097.0,
      "step": 215
    },
    {
      "entropy": 7.220300769805908,
      "epoch": 0.22188603126575895,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0001095,
      "loss": 6.6645,
      "mean_token_accuracy": 0.0966582477092743,
      "num_tokens": 482006.0,
      "step": 220
    },
    {
      "entropy": 7.262398672103882,
      "epoch": 0.22692889561270801,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000112,
      "loss": 6.8118,
      "mean_token_accuracy": 0.09521539956331253,
      "num_tokens": 493134.0,
      "step": 225
    },
    {
      "entropy": 7.295605850219727,
      "epoch": 0.23197175995965708,
      "grad_norm": 1.0,
      "learning_rate": 0.0001145,
      "loss": 6.743,
      "mean_token_accuracy": 0.0946677066385746,
      "num_tokens": 504801.0,
      "step": 230
    },
    {
      "entropy": 7.281533098220825,
      "epoch": 0.23701462430660616,
      "grad_norm": 1.015625,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.8161,
      "mean_token_accuracy": 0.09522581920027733,
      "num_tokens": 517523.0,
      "step": 235
    },
    {
      "entropy": 7.199113416671753,
      "epoch": 0.24205748865355523,
      "grad_norm": 1.0625,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.7182,
      "mean_token_accuracy": 0.09818312823772431,
      "num_tokens": 528267.0,
      "step": 240
    },
    {
      "entropy": 7.226230573654175,
      "epoch": 0.24710035300050429,
      "grad_norm": 1.03125,
      "learning_rate": 0.000122,
      "loss": 6.7107,
      "mean_token_accuracy": 0.09675659462809563,
      "num_tokens": 538750.0,
      "step": 245
    },
    {
      "entropy": 7.268562889099121,
      "epoch": 0.2521432173474534,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001245,
      "loss": 6.7346,
      "mean_token_accuracy": 0.097308711707592,
      "num_tokens": 549380.0,
      "step": 250
    },
    {
      "entropy": 7.262276458740234,
      "epoch": 0.25718608169440244,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000127,
      "loss": 6.7255,
      "mean_token_accuracy": 0.09805365800857543,
      "num_tokens": 559643.0,
      "step": 255
    },
    {
      "entropy": 7.235341882705688,
      "epoch": 0.2622289460413515,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001295,
      "loss": 6.7632,
      "mean_token_accuracy": 0.09189907684922219,
      "num_tokens": 571250.0,
      "step": 260
    },
    {
      "entropy": 7.196493816375733,
      "epoch": 0.26727181038830056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000132,
      "loss": 6.7442,
      "mean_token_accuracy": 0.10515325963497162,
      "num_tokens": 582406.0,
      "step": 265
    },
    {
      "entropy": 7.352744102478027,
      "epoch": 0.2723146747352496,
      "grad_norm": 1.015625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.8521,
      "mean_token_accuracy": 0.09412558823823929,
      "num_tokens": 593115.0,
      "step": 270
    },
    {
      "entropy": 7.18369722366333,
      "epoch": 0.2773575390821987,
      "grad_norm": 1.03125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.7276,
      "mean_token_accuracy": 0.09833462312817573,
      "num_tokens": 603938.0,
      "step": 275
    },
    {
      "entropy": 7.247702264785767,
      "epoch": 0.28240040342914774,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001395,
      "loss": 6.771,
      "mean_token_accuracy": 0.09771163985133172,
      "num_tokens": 614067.0,
      "step": 280
    },
    {
      "entropy": 7.243542671203613,
      "epoch": 0.2874432677760968,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7654,
      "mean_token_accuracy": 0.09758238643407821,
      "num_tokens": 626271.0,
      "step": 285
    },
    {
      "entropy": 7.291503524780273,
      "epoch": 0.2924861321230459,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001445,
      "loss": 6.7423,
      "mean_token_accuracy": 0.09706803634762765,
      "num_tokens": 636580.0,
      "step": 290
    },
    {
      "entropy": 7.140781545639038,
      "epoch": 0.297528996469995,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000147,
      "loss": 6.6248,
      "mean_token_accuracy": 0.1026174359023571,
      "num_tokens": 648202.0,
      "step": 295
    },
    {
      "entropy": 7.259130382537842,
      "epoch": 0.30257186081694404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001495,
      "loss": 6.7082,
      "mean_token_accuracy": 0.09900272190570832,
      "num_tokens": 659274.0,
      "step": 300
    },
    {
      "entropy": 7.239129590988159,
      "epoch": 0.3076147251638931,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000152,
      "loss": 6.7624,
      "mean_token_accuracy": 0.09335382506251336,
      "num_tokens": 670782.0,
      "step": 305
    },
    {
      "entropy": 7.163164758682251,
      "epoch": 0.31265758951084216,
      "grad_norm": 1.046875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.6747,
      "mean_token_accuracy": 0.09428300261497498,
      "num_tokens": 681392.0,
      "step": 310
    },
    {
      "entropy": 7.21016297340393,
      "epoch": 0.3177004538577912,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000157,
      "loss": 6.7037,
      "mean_token_accuracy": 0.10150701180100441,
      "num_tokens": 692779.0,
      "step": 315
    },
    {
      "entropy": 7.229485654830933,
      "epoch": 0.3227433182047403,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001595,
      "loss": 6.7269,
      "mean_token_accuracy": 0.10247877538204193,
      "num_tokens": 702200.0,
      "step": 320
    },
    {
      "entropy": 7.1149742126464846,
      "epoch": 0.32778618255168934,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000162,
      "loss": 6.7218,
      "mean_token_accuracy": 0.09391158893704414,
      "num_tokens": 714817.0,
      "step": 325
    },
    {
      "entropy": 7.174094009399414,
      "epoch": 0.3328290468986384,
      "grad_norm": 1.046875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.6052,
      "mean_token_accuracy": 0.0979647196829319,
      "num_tokens": 725555.0,
      "step": 330
    },
    {
      "entropy": 7.13692626953125,
      "epoch": 0.3378719112455875,
      "grad_norm": 1.03125,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.6195,
      "mean_token_accuracy": 0.09633892104029655,
      "num_tokens": 737372.0,
      "step": 335
    },
    {
      "entropy": 7.185083723068237,
      "epoch": 0.3429147755925366,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.6887,
      "mean_token_accuracy": 0.09809593409299851,
      "num_tokens": 748515.0,
      "step": 340
    },
    {
      "entropy": 7.191352319717407,
      "epoch": 0.34795763993948564,
      "grad_norm": 1.03125,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7335,
      "mean_token_accuracy": 0.09944713562726974,
      "num_tokens": 760586.0,
      "step": 345
    },
    {
      "entropy": 7.127374887466431,
      "epoch": 0.3530005042864347,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.6665,
      "mean_token_accuracy": 0.09854965955018997,
      "num_tokens": 772228.0,
      "step": 350
    },
    {
      "entropy": 7.2189923286437985,
      "epoch": 0.35804336863338376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000177,
      "loss": 6.7453,
      "mean_token_accuracy": 0.0974178895354271,
      "num_tokens": 784035.0,
      "step": 355
    },
    {
      "entropy": 7.065363168716431,
      "epoch": 0.3630862329803328,
      "grad_norm": 1.125,
      "learning_rate": 0.0001795,
      "loss": 6.6418,
      "mean_token_accuracy": 0.10008223056793213,
      "num_tokens": 794383.0,
      "step": 360
    },
    {
      "entropy": 7.085238981246948,
      "epoch": 0.3681290973272819,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000182,
      "loss": 6.5745,
      "mean_token_accuracy": 0.10618990287184715,
      "num_tokens": 805192.0,
      "step": 365
    },
    {
      "entropy": 7.1900914192199705,
      "epoch": 0.37317196167423095,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001845,
      "loss": 6.7213,
      "mean_token_accuracy": 0.10011058002710342,
      "num_tokens": 816794.0,
      "step": 370
    },
    {
      "entropy": 7.214580726623535,
      "epoch": 0.37821482602118,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000187,
      "loss": 6.7313,
      "mean_token_accuracy": 0.09720972329378127,
      "num_tokens": 827505.0,
      "step": 375
    },
    {
      "entropy": 7.1219089984893795,
      "epoch": 0.3832576903681291,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001895,
      "loss": 6.5946,
      "mean_token_accuracy": 0.10458309799432755,
      "num_tokens": 837462.0,
      "step": 380
    },
    {
      "entropy": 7.100887250900269,
      "epoch": 0.3883005547150782,
      "grad_norm": 1.015625,
      "learning_rate": 0.000192,
      "loss": 6.552,
      "mean_token_accuracy": 0.10412034988403321,
      "num_tokens": 848309.0,
      "step": 385
    },
    {
      "entropy": 7.13932957649231,
      "epoch": 0.39334341906202724,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001945,
      "loss": 6.7236,
      "mean_token_accuracy": 0.09661326780915261,
      "num_tokens": 859474.0,
      "step": 390
    },
    {
      "entropy": 7.078795003890991,
      "epoch": 0.3983862834089763,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.5887,
      "mean_token_accuracy": 0.10084822699427605,
      "num_tokens": 871638.0,
      "step": 395
    },
    {
      "entropy": 7.1444016456604,
      "epoch": 0.40342914775592537,
      "grad_norm": 1.109375,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6305,
      "mean_token_accuracy": 0.10135297477245331,
      "num_tokens": 882480.0,
      "step": 400
    },
    {
      "entropy": 7.110705089569092,
      "epoch": 0.4084720121028744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000202,
      "loss": 6.6247,
      "mean_token_accuracy": 0.10225840136408806,
      "num_tokens": 893366.0,
      "step": 405
    },
    {
      "entropy": 7.121734523773194,
      "epoch": 0.4135148764498235,
      "grad_norm": 1.21875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.5849,
      "mean_token_accuracy": 0.1083330750465393,
      "num_tokens": 904678.0,
      "step": 410
    },
    {
      "entropy": 7.0099626064300535,
      "epoch": 0.41855774079677255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000207,
      "loss": 6.5967,
      "mean_token_accuracy": 0.1049334615468979,
      "num_tokens": 915756.0,
      "step": 415
    },
    {
      "entropy": 7.100469732284546,
      "epoch": 0.4236006051437216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002095,
      "loss": 6.581,
      "mean_token_accuracy": 0.09910632744431495,
      "num_tokens": 926204.0,
      "step": 420
    },
    {
      "entropy": 7.060488557815551,
      "epoch": 0.4286434694906707,
      "grad_norm": 1.3125,
      "learning_rate": 0.000212,
      "loss": 6.6065,
      "mean_token_accuracy": 0.10566477999091148,
      "num_tokens": 936894.0,
      "step": 425
    },
    {
      "entropy": 7.137387037277222,
      "epoch": 0.4336863338376198,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002145,
      "loss": 6.6479,
      "mean_token_accuracy": 0.10312608256936073,
      "num_tokens": 948987.0,
      "step": 430
    },
    {
      "entropy": 7.084998321533203,
      "epoch": 0.43872919818456885,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6537,
      "mean_token_accuracy": 0.1018140584230423,
      "num_tokens": 961036.0,
      "step": 435
    },
    {
      "entropy": 7.125703001022339,
      "epoch": 0.4437720625315179,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002195,
      "loss": 6.6763,
      "mean_token_accuracy": 0.09882956370711327,
      "num_tokens": 972445.0,
      "step": 440
    },
    {
      "entropy": 7.0276741027832035,
      "epoch": 0.44881492687846697,
      "grad_norm": 1.171875,
      "learning_rate": 0.000222,
      "loss": 6.6481,
      "mean_token_accuracy": 0.09917017221450805,
      "num_tokens": 982847.0,
      "step": 445
    },
    {
      "entropy": 7.137781095504761,
      "epoch": 0.45385779122541603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002245,
      "loss": 6.6044,
      "mean_token_accuracy": 0.10046214014291763,
      "num_tokens": 994055.0,
      "step": 450
    },
    {
      "entropy": 7.033148241043091,
      "epoch": 0.4589006555723651,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.5887,
      "mean_token_accuracy": 0.10494610518217087,
      "num_tokens": 1003565.0,
      "step": 455
    },
    {
      "entropy": 6.981668519973755,
      "epoch": 0.46394351991931415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.4992,
      "mean_token_accuracy": 0.11249068826436996,
      "num_tokens": 1013376.0,
      "step": 460
    },
    {
      "entropy": 7.059791231155396,
      "epoch": 0.4689863842662632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5858,
      "mean_token_accuracy": 0.10482539832592011,
      "num_tokens": 1024580.0,
      "step": 465
    },
    {
      "entropy": 6.999967098236084,
      "epoch": 0.47402924861321233,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5819,
      "mean_token_accuracy": 0.10121265724301338,
      "num_tokens": 1037172.0,
      "step": 470
    },
    {
      "entropy": 7.029157018661499,
      "epoch": 0.4790721129601614,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000237,
      "loss": 6.5026,
      "mean_token_accuracy": 0.10607488974928855,
      "num_tokens": 1046690.0,
      "step": 475
    },
    {
      "entropy": 6.958544731140137,
      "epoch": 0.48411497730711045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002395,
      "loss": 6.5115,
      "mean_token_accuracy": 0.10288126021623611,
      "num_tokens": 1058341.0,
      "step": 480
    },
    {
      "entropy": 6.991989469528198,
      "epoch": 0.4891578416540595,
      "grad_norm": 1.203125,
      "learning_rate": 0.000242,
      "loss": 6.4452,
      "mean_token_accuracy": 0.10999866873025894,
      "num_tokens": 1067544.0,
      "step": 485
    },
    {
      "entropy": 7.035932159423828,
      "epoch": 0.49420070600100857,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002445,
      "loss": 6.5638,
      "mean_token_accuracy": 0.10314799919724464,
      "num_tokens": 1079028.0,
      "step": 490
    },
    {
      "entropy": 6.959421110153198,
      "epoch": 0.49924357034795763,
      "grad_norm": 1.25,
      "learning_rate": 0.000247,
      "loss": 6.6327,
      "mean_token_accuracy": 0.1083416223526001,
      "num_tokens": 1090709.0,
      "step": 495
    },
    {
      "entropy": 6.954031038284302,
      "epoch": 0.5042864346949067,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002495,
      "loss": 6.5021,
      "mean_token_accuracy": 0.10255713239312172,
      "num_tokens": 1102987.0,
      "step": 500
    },
    {
      "entropy": 7.159888648986817,
      "epoch": 0.5093292990418558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000252,
      "loss": 6.6089,
      "mean_token_accuracy": 0.09792369678616523,
      "num_tokens": 1113370.0,
      "step": 505
    },
    {
      "entropy": 6.860920858383179,
      "epoch": 0.5143721633888049,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002545,
      "loss": 6.452,
      "mean_token_accuracy": 0.11532516852021217,
      "num_tokens": 1124054.0,
      "step": 510
    },
    {
      "entropy": 7.0116840362548825,
      "epoch": 0.5194150277357539,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000257,
      "loss": 6.4394,
      "mean_token_accuracy": 0.1111026681959629,
      "num_tokens": 1134530.0,
      "step": 515
    },
    {
      "entropy": 6.8797385692596436,
      "epoch": 0.524457892082703,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002595,
      "loss": 6.4534,
      "mean_token_accuracy": 0.11024491488933563,
      "num_tokens": 1146011.0,
      "step": 520
    },
    {
      "entropy": 6.911460733413696,
      "epoch": 0.529500756429652,
      "grad_norm": 1.125,
      "learning_rate": 0.000262,
      "loss": 6.5475,
      "mean_token_accuracy": 0.10476418659090996,
      "num_tokens": 1155822.0,
      "step": 525
    },
    {
      "entropy": 6.9672643661499025,
      "epoch": 0.5345436207766011,
      "grad_norm": 0.9375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.4517,
      "mean_token_accuracy": 0.10918923616409301,
      "num_tokens": 1167398.0,
      "step": 530
    },
    {
      "entropy": 6.951054525375366,
      "epoch": 0.5395864851235502,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5179,
      "mean_token_accuracy": 0.10848165675997734,
      "num_tokens": 1179706.0,
      "step": 535
    },
    {
      "entropy": 6.842685604095459,
      "epoch": 0.5446293494704992,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.4623,
      "mean_token_accuracy": 0.10590022280812264,
      "num_tokens": 1190159.0,
      "step": 540
    },
    {
      "entropy": 6.887945699691772,
      "epoch": 0.5496722138174484,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.3957,
      "mean_token_accuracy": 0.11316062286496162,
      "num_tokens": 1200223.0,
      "step": 545
    },
    {
      "entropy": 6.8363923072814945,
      "epoch": 0.5547150781643974,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002745,
      "loss": 6.5028,
      "mean_token_accuracy": 0.10700231790542603,
      "num_tokens": 1212108.0,
      "step": 550
    },
    {
      "entropy": 6.985593938827515,
      "epoch": 0.5597579425113465,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000277,
      "loss": 6.4666,
      "mean_token_accuracy": 0.1086841069161892,
      "num_tokens": 1221940.0,
      "step": 555
    },
    {
      "entropy": 6.873191499710083,
      "epoch": 0.5648008068582955,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002795,
      "loss": 6.5171,
      "mean_token_accuracy": 0.10610352382063866,
      "num_tokens": 1232884.0,
      "step": 560
    },
    {
      "entropy": 6.956163358688355,
      "epoch": 0.5698436712052446,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.4327,
      "mean_token_accuracy": 0.10770275890827179,
      "num_tokens": 1242758.0,
      "step": 565
    },
    {
      "entropy": 6.814071226119995,
      "epoch": 0.5748865355521936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002845,
      "loss": 6.4253,
      "mean_token_accuracy": 0.11039303541183472,
      "num_tokens": 1254787.0,
      "step": 570
    },
    {
      "entropy": 6.858562231063843,
      "epoch": 0.5799293998991427,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000287,
      "loss": 6.4618,
      "mean_token_accuracy": 0.11011067107319832,
      "num_tokens": 1266364.0,
      "step": 575
    },
    {
      "entropy": 6.871497106552124,
      "epoch": 0.5849722642460918,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002895,
      "loss": 6.4473,
      "mean_token_accuracy": 0.10705809071660041,
      "num_tokens": 1277105.0,
      "step": 580
    },
    {
      "entropy": 7.04167366027832,
      "epoch": 0.5900151285930408,
      "grad_norm": 0.921875,
      "learning_rate": 0.000292,
      "loss": 6.6277,
      "mean_token_accuracy": 0.10272014588117599,
      "num_tokens": 1288470.0,
      "step": 585
    },
    {
      "entropy": 6.948342657089233,
      "epoch": 0.59505799293999,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002945,
      "loss": 6.5378,
      "mean_token_accuracy": 0.10782962441444396,
      "num_tokens": 1301596.0,
      "step": 590
    },
    {
      "entropy": 6.915746879577637,
      "epoch": 0.600100857286939,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000297,
      "loss": 6.5886,
      "mean_token_accuracy": 0.10370224639773369,
      "num_tokens": 1314102.0,
      "step": 595
    },
    {
      "entropy": 6.862001037597656,
      "epoch": 0.6051437216338881,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002995,
      "loss": 6.4664,
      "mean_token_accuracy": 0.11232062503695488,
      "num_tokens": 1326939.0,
      "step": 600
    },
    {
      "entropy": 6.942681741714478,
      "epoch": 0.6101865859808371,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000302,
      "loss": 6.5184,
      "mean_token_accuracy": 0.10658980160951614,
      "num_tokens": 1337877.0,
      "step": 605
    },
    {
      "entropy": 6.862700510025024,
      "epoch": 0.6152294503277862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003045,
      "loss": 6.4309,
      "mean_token_accuracy": 0.10537772327661514,
      "num_tokens": 1348425.0,
      "step": 610
    },
    {
      "entropy": 6.898571968078613,
      "epoch": 0.6202723146747352,
      "grad_norm": 1.09375,
      "learning_rate": 0.000307,
      "loss": 6.4755,
      "mean_token_accuracy": 0.11413818150758744,
      "num_tokens": 1359216.0,
      "step": 615
    },
    {
      "entropy": 6.801260757446289,
      "epoch": 0.6253151790216843,
      "grad_norm": 1.125,
      "learning_rate": 0.0003095,
      "loss": 6.4595,
      "mean_token_accuracy": 0.11029788851737976,
      "num_tokens": 1371181.0,
      "step": 620
    },
    {
      "entropy": 6.974251842498779,
      "epoch": 0.6303580433686334,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000312,
      "loss": 6.4479,
      "mean_token_accuracy": 0.10635321885347367,
      "num_tokens": 1382371.0,
      "step": 625
    },
    {
      "entropy": 6.8364170551300045,
      "epoch": 0.6354009077155824,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003145,
      "loss": 6.4388,
      "mean_token_accuracy": 0.10637976825237275,
      "num_tokens": 1392981.0,
      "step": 630
    },
    {
      "entropy": 6.818234634399414,
      "epoch": 0.6404437720625316,
      "grad_norm": 1.015625,
      "learning_rate": 0.000317,
      "loss": 6.4792,
      "mean_token_accuracy": 0.11108815595507622,
      "num_tokens": 1405782.0,
      "step": 635
    },
    {
      "entropy": 6.831936502456665,
      "epoch": 0.6454866364094806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003195,
      "loss": 6.4194,
      "mean_token_accuracy": 0.11271615028381347,
      "num_tokens": 1415809.0,
      "step": 640
    },
    {
      "entropy": 6.828110933303833,
      "epoch": 0.6505295007564297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000322,
      "loss": 6.4539,
      "mean_token_accuracy": 0.10585936233401298,
      "num_tokens": 1427845.0,
      "step": 645
    },
    {
      "entropy": 6.850209903717041,
      "epoch": 0.6555723651033787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4514,
      "mean_token_accuracy": 0.10457649230957031,
      "num_tokens": 1439761.0,
      "step": 650
    },
    {
      "entropy": 6.716944456100464,
      "epoch": 0.6606152294503278,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4282,
      "mean_token_accuracy": 0.11434582471847535,
      "num_tokens": 1450373.0,
      "step": 655
    },
    {
      "entropy": 6.7856956958770756,
      "epoch": 0.6656580937972768,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.308,
      "mean_token_accuracy": 0.11268317475914955,
      "num_tokens": 1462134.0,
      "step": 660
    },
    {
      "entropy": 6.804158306121826,
      "epoch": 0.6707009581442259,
      "grad_norm": 1.125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.4147,
      "mean_token_accuracy": 0.11142993420362472,
      "num_tokens": 1473600.0,
      "step": 665
    },
    {
      "entropy": 6.800200128555298,
      "epoch": 0.675743822491175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.361,
      "mean_token_accuracy": 0.11059448346495629,
      "num_tokens": 1484313.0,
      "step": 670
    },
    {
      "entropy": 6.850586938858032,
      "epoch": 0.680786686838124,
      "grad_norm": 1.171875,
      "learning_rate": 0.000337,
      "loss": 6.445,
      "mean_token_accuracy": 0.10799070671200753,
      "num_tokens": 1495407.0,
      "step": 675
    },
    {
      "entropy": 6.776702213287353,
      "epoch": 0.6858295511850732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003395,
      "loss": 6.526,
      "mean_token_accuracy": 0.10533422976732254,
      "num_tokens": 1506707.0,
      "step": 680
    },
    {
      "entropy": 6.7912431240081785,
      "epoch": 0.6908724155320222,
      "grad_norm": 1.078125,
      "learning_rate": 0.000342,
      "loss": 6.3557,
      "mean_token_accuracy": 0.11059139668941498,
      "num_tokens": 1517991.0,
      "step": 685
    },
    {
      "entropy": 6.7320136547088625,
      "epoch": 0.6959152798789713,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4017,
      "mean_token_accuracy": 0.11187214478850364,
      "num_tokens": 1529538.0,
      "step": 690
    },
    {
      "entropy": 6.701151990890503,
      "epoch": 0.7009581442259203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000347,
      "loss": 6.4026,
      "mean_token_accuracy": 0.11320648416876793,
      "num_tokens": 1540183.0,
      "step": 695
    },
    {
      "entropy": 6.852318525314331,
      "epoch": 0.7060010085728694,
      "grad_norm": 1.125,
      "learning_rate": 0.0003495,
      "loss": 6.4475,
      "mean_token_accuracy": 0.10782841965556145,
      "num_tokens": 1552899.0,
      "step": 700
    },
    {
      "entropy": 6.773307371139526,
      "epoch": 0.7110438729198184,
      "grad_norm": 1.046875,
      "learning_rate": 0.000352,
      "loss": 6.424,
      "mean_token_accuracy": 0.10968175902962685,
      "num_tokens": 1563219.0,
      "step": 705
    },
    {
      "entropy": 6.795606517791748,
      "epoch": 0.7160867372667675,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003545,
      "loss": 6.3444,
      "mean_token_accuracy": 0.1151584655046463,
      "num_tokens": 1573508.0,
      "step": 710
    },
    {
      "entropy": 6.708667278289795,
      "epoch": 0.7211296016137166,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000357,
      "loss": 6.4283,
      "mean_token_accuracy": 0.10857328176498413,
      "num_tokens": 1584319.0,
      "step": 715
    },
    {
      "entropy": 6.716313695907592,
      "epoch": 0.7261724659606656,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003595,
      "loss": 6.297,
      "mean_token_accuracy": 0.12033721506595611,
      "num_tokens": 1595986.0,
      "step": 720
    },
    {
      "entropy": 6.664373064041138,
      "epoch": 0.7312153303076148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000362,
      "loss": 6.37,
      "mean_token_accuracy": 0.11175311356782913,
      "num_tokens": 1606742.0,
      "step": 725
    },
    {
      "entropy": 6.734103393554688,
      "epoch": 0.7362581946545638,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003645,
      "loss": 6.3412,
      "mean_token_accuracy": 0.11203160732984543,
      "num_tokens": 1618325.0,
      "step": 730
    },
    {
      "entropy": 6.739001607894897,
      "epoch": 0.7413010590015129,
      "grad_norm": 1.21875,
      "learning_rate": 0.000367,
      "loss": 6.3498,
      "mean_token_accuracy": 0.1113662451505661,
      "num_tokens": 1627693.0,
      "step": 735
    },
    {
      "entropy": 6.693121767044067,
      "epoch": 0.7463439233484619,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003695,
      "loss": 6.3145,
      "mean_token_accuracy": 0.11420592591166497,
      "num_tokens": 1637707.0,
      "step": 740
    },
    {
      "entropy": 6.634211206436158,
      "epoch": 0.751386787695411,
      "grad_norm": 1.171875,
      "learning_rate": 0.000372,
      "loss": 6.3234,
      "mean_token_accuracy": 0.1172995112836361,
      "num_tokens": 1647277.0,
      "step": 745
    },
    {
      "entropy": 6.757878684997559,
      "epoch": 0.75642965204236,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003745,
      "loss": 6.3699,
      "mean_token_accuracy": 0.11142466813325883,
      "num_tokens": 1657919.0,
      "step": 750
    },
    {
      "entropy": 6.714281225204468,
      "epoch": 0.7614725163893091,
      "grad_norm": 1.140625,
      "learning_rate": 0.000377,
      "loss": 6.4127,
      "mean_token_accuracy": 0.11027689799666404,
      "num_tokens": 1669107.0,
      "step": 755
    },
    {
      "entropy": 6.73455228805542,
      "epoch": 0.7665153807362582,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003795,
      "loss": 6.4201,
      "mean_token_accuracy": 0.1129143200814724,
      "num_tokens": 1681338.0,
      "step": 760
    },
    {
      "entropy": 6.710708665847778,
      "epoch": 0.7715582450832073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000382,
      "loss": 6.2621,
      "mean_token_accuracy": 0.11415471211075782,
      "num_tokens": 1691288.0,
      "step": 765
    },
    {
      "entropy": 6.568854570388794,
      "epoch": 0.7766011094301564,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003845,
      "loss": 6.3503,
      "mean_token_accuracy": 0.11261749044060707,
      "num_tokens": 1702864.0,
      "step": 770
    },
    {
      "entropy": 6.679662752151489,
      "epoch": 0.7816439737771054,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.2684,
      "mean_token_accuracy": 0.12043757885694503,
      "num_tokens": 1712454.0,
      "step": 775
    },
    {
      "entropy": 6.683392810821533,
      "epoch": 0.7866868381240545,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4333,
      "mean_token_accuracy": 0.1112380564212799,
      "num_tokens": 1723925.0,
      "step": 780
    },
    {
      "entropy": 6.6579155921936035,
      "epoch": 0.7917297024710035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.36,
      "mean_token_accuracy": 0.11308201998472214,
      "num_tokens": 1735624.0,
      "step": 785
    },
    {
      "entropy": 6.683883953094482,
      "epoch": 0.7967725668179526,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.3225,
      "mean_token_accuracy": 0.11539738923311234,
      "num_tokens": 1746288.0,
      "step": 790
    },
    {
      "entropy": 6.768927764892578,
      "epoch": 0.8018154311649016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4913,
      "mean_token_accuracy": 0.10619482472538948,
      "num_tokens": 1758145.0,
      "step": 795
    },
    {
      "entropy": 6.602115249633789,
      "epoch": 0.8068582955118507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003995,
      "loss": 6.3161,
      "mean_token_accuracy": 0.1136295534670353,
      "num_tokens": 1768586.0,
      "step": 800
    },
    {
      "entropy": 6.709038305282593,
      "epoch": 0.8119011598587998,
      "grad_norm": 1.078125,
      "learning_rate": 0.000402,
      "loss": 6.321,
      "mean_token_accuracy": 0.11802070513367653,
      "num_tokens": 1780502.0,
      "step": 805
    },
    {
      "entropy": 6.624473953247071,
      "epoch": 0.8169440242057489,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004045,
      "loss": 6.3582,
      "mean_token_accuracy": 0.11211898848414421,
      "num_tokens": 1791450.0,
      "step": 810
    },
    {
      "entropy": 6.6230384349823,
      "epoch": 0.821986888552698,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3628,
      "mean_token_accuracy": 0.1177310198545456,
      "num_tokens": 1803690.0,
      "step": 815
    },
    {
      "entropy": 6.609463548660278,
      "epoch": 0.827029752899647,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004095,
      "loss": 6.2264,
      "mean_token_accuracy": 0.11917722448706627,
      "num_tokens": 1813630.0,
      "step": 820
    },
    {
      "entropy": 6.637943077087402,
      "epoch": 0.8320726172465961,
      "grad_norm": 1.03125,
      "learning_rate": 0.000412,
      "loss": 6.2778,
      "mean_token_accuracy": 0.11222293600440025,
      "num_tokens": 1824537.0,
      "step": 825
    },
    {
      "entropy": 6.645944356918335,
      "epoch": 0.8371154815935451,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004145,
      "loss": 6.3373,
      "mean_token_accuracy": 0.1141052208840847,
      "num_tokens": 1834404.0,
      "step": 830
    },
    {
      "entropy": 6.700550842285156,
      "epoch": 0.8421583459404942,
      "grad_norm": 1.09375,
      "learning_rate": 0.000417,
      "loss": 6.3104,
      "mean_token_accuracy": 0.12324466183781624,
      "num_tokens": 1845371.0,
      "step": 835
    },
    {
      "entropy": 6.50260362625122,
      "epoch": 0.8472012102874432,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004195,
      "loss": 6.341,
      "mean_token_accuracy": 0.11797986924648285,
      "num_tokens": 1856749.0,
      "step": 840
    },
    {
      "entropy": 6.63838267326355,
      "epoch": 0.8522440746343923,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422,
      "loss": 6.3337,
      "mean_token_accuracy": 0.109927898645401,
      "num_tokens": 1866577.0,
      "step": 845
    },
    {
      "entropy": 6.72889552116394,
      "epoch": 0.8572869389813415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245,
      "loss": 6.282,
      "mean_token_accuracy": 0.1153933621942997,
      "num_tokens": 1877123.0,
      "step": 850
    },
    {
      "entropy": 6.602636194229126,
      "epoch": 0.8623298033282905,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000427,
      "loss": 6.3232,
      "mean_token_accuracy": 0.11297061070799827,
      "num_tokens": 1888171.0,
      "step": 855
    },
    {
      "entropy": 6.546885442733765,
      "epoch": 0.8673726676752396,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004295,
      "loss": 6.2226,
      "mean_token_accuracy": 0.11574787572026253,
      "num_tokens": 1898150.0,
      "step": 860
    },
    {
      "entropy": 6.649620103836059,
      "epoch": 0.8724155320221886,
      "grad_norm": 1.203125,
      "learning_rate": 0.000432,
      "loss": 6.3305,
      "mean_token_accuracy": 0.11207615062594414,
      "num_tokens": 1907641.0,
      "step": 865
    },
    {
      "entropy": 6.632721757888794,
      "epoch": 0.8774583963691377,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004345,
      "loss": 6.2845,
      "mean_token_accuracy": 0.11901096925139427,
      "num_tokens": 1918297.0,
      "step": 870
    },
    {
      "entropy": 6.578178358078003,
      "epoch": 0.8825012607160867,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000437,
      "loss": 6.3186,
      "mean_token_accuracy": 0.1133461907505989,
      "num_tokens": 1930268.0,
      "step": 875
    },
    {
      "entropy": 6.531199407577515,
      "epoch": 0.8875441250630358,
      "grad_norm": 1.0,
      "learning_rate": 0.0004395,
      "loss": 6.2465,
      "mean_token_accuracy": 0.12018090561032295,
      "num_tokens": 1941762.0,
      "step": 880
    },
    {
      "entropy": 6.59914002418518,
      "epoch": 0.8925869894099848,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000442,
      "loss": 6.3195,
      "mean_token_accuracy": 0.11756950095295907,
      "num_tokens": 1952416.0,
      "step": 885
    },
    {
      "entropy": 6.5428722381591795,
      "epoch": 0.8976298537569339,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004445,
      "loss": 6.2651,
      "mean_token_accuracy": 0.11572367325425148,
      "num_tokens": 1963451.0,
      "step": 890
    },
    {
      "entropy": 6.58216700553894,
      "epoch": 0.902672718103883,
      "grad_norm": 1.046875,
      "learning_rate": 0.000447,
      "loss": 6.2794,
      "mean_token_accuracy": 0.11693633943796158,
      "num_tokens": 1974360.0,
      "step": 895
    },
    {
      "entropy": 6.693656730651855,
      "epoch": 0.9077155824508321,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3962,
      "mean_token_accuracy": 0.10579293221235275,
      "num_tokens": 1985823.0,
      "step": 900
    },
    {
      "entropy": 6.767707967758179,
      "epoch": 0.9127584467977812,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4287,
      "mean_token_accuracy": 0.11259771957993507,
      "num_tokens": 1997475.0,
      "step": 905
    },
    {
      "entropy": 6.570524787902832,
      "epoch": 0.9178013111447302,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.2708,
      "mean_token_accuracy": 0.11376104354858399,
      "num_tokens": 2009029.0,
      "step": 910
    },
    {
      "entropy": 6.4297045230865475,
      "epoch": 0.9228441754916793,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.141,
      "mean_token_accuracy": 0.11884743124246597,
      "num_tokens": 2019564.0,
      "step": 915
    },
    {
      "entropy": 6.618962907791138,
      "epoch": 0.9278870398386283,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3949,
      "mean_token_accuracy": 0.11527436822652817,
      "num_tokens": 2031394.0,
      "step": 920
    },
    {
      "entropy": 6.623352861404419,
      "epoch": 0.9329299041855774,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000462,
      "loss": 6.2762,
      "mean_token_accuracy": 0.1077704131603241,
      "num_tokens": 2042132.0,
      "step": 925
    },
    {
      "entropy": 6.538896608352661,
      "epoch": 0.9379727685325264,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004645,
      "loss": 6.2666,
      "mean_token_accuracy": 0.1150762416422367,
      "num_tokens": 2054469.0,
      "step": 930
    },
    {
      "entropy": 6.48656005859375,
      "epoch": 0.9430156328794755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000467,
      "loss": 6.2445,
      "mean_token_accuracy": 0.11287088468670844,
      "num_tokens": 2065001.0,
      "step": 935
    },
    {
      "entropy": 6.535896301269531,
      "epoch": 0.9480584972264247,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004695,
      "loss": 6.2612,
      "mean_token_accuracy": 0.11251269802451133,
      "num_tokens": 2075786.0,
      "step": 940
    },
    {
      "entropy": 6.53080735206604,
      "epoch": 0.9531013615733737,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000472,
      "loss": 6.2161,
      "mean_token_accuracy": 0.1198991946876049,
      "num_tokens": 2087818.0,
      "step": 945
    },
    {
      "entropy": 6.592200040817261,
      "epoch": 0.9581442259203228,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004745,
      "loss": 6.3004,
      "mean_token_accuracy": 0.1103257454931736,
      "num_tokens": 2098771.0,
      "step": 950
    },
    {
      "entropy": 6.4617808818817135,
      "epoch": 0.9631870902672718,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000477,
      "loss": 6.2119,
      "mean_token_accuracy": 0.12062049955129624,
      "num_tokens": 2110036.0,
      "step": 955
    },
    {
      "entropy": 6.536091947555542,
      "epoch": 0.9682299546142209,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004795,
      "loss": 6.312,
      "mean_token_accuracy": 0.12011703029274941,
      "num_tokens": 2120411.0,
      "step": 960
    },
    {
      "entropy": 6.482806015014648,
      "epoch": 0.9732728189611699,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000482,
      "loss": 6.2244,
      "mean_token_accuracy": 0.11997658982872964,
      "num_tokens": 2132831.0,
      "step": 965
    },
    {
      "entropy": 6.493047666549683,
      "epoch": 0.978315683308119,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004845,
      "loss": 6.1953,
      "mean_token_accuracy": 0.11635629460215569,
      "num_tokens": 2143166.0,
      "step": 970
    },
    {
      "entropy": 6.481213617324829,
      "epoch": 0.983358547655068,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000487,
      "loss": 6.2695,
      "mean_token_accuracy": 0.10903773382306099,
      "num_tokens": 2154366.0,
      "step": 975
    },
    {
      "entropy": 6.55472240447998,
      "epoch": 0.9884014120020171,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004895,
      "loss": 6.2535,
      "mean_token_accuracy": 0.11955308020114899,
      "num_tokens": 2164396.0,
      "step": 980
    },
    {
      "entropy": 6.478724098205566,
      "epoch": 0.9934442763489663,
      "grad_norm": 1.015625,
      "learning_rate": 0.000492,
      "loss": 6.3211,
      "mean_token_accuracy": 0.11328333243727684,
      "num_tokens": 2177295.0,
      "step": 985
    },
    {
      "entropy": 6.5510862350463865,
      "epoch": 0.9984871406959153,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004945,
      "loss": 6.1761,
      "mean_token_accuracy": 0.12193462178111077,
      "num_tokens": 2188832.0,
      "step": 990
    },
    {
      "entropy": 6.45083400938246,
      "epoch": 1.0030257186081695,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000497,
      "loss": 6.1345,
      "mean_token_accuracy": 0.11941885699828465,
      "num_tokens": 2197926.0,
      "step": 995
    },
    {
      "entropy": 6.500611925125122,
      "epoch": 1.0080685829551186,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995,
      "loss": 6.173,
      "mean_token_accuracy": 0.11789006367325783,
      "num_tokens": 2209156.0,
      "step": 1000
    },
    {
      "epoch": 1.0080685829551186,
      "eval_entropy": 6.293738744337746,
      "eval_loss": 6.301364898681641,
      "eval_mean_token_accuracy": 0.11593409591655392,
      "eval_num_tokens": 2209156.0,
      "eval_runtime": 2.053,
      "eval_samples_per_second": 1528.532,
      "eval_steps_per_second": 191.432,
      "step": 1000
    }
  ],
  "logging_steps": 5,
  "max_steps": 9910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 546050741299200.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}