{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.032274331820474,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742588329315186,
      "epoch": 0.005042864346949067,
      "grad_norm": 4.78125,
      "learning_rate": 2e-06,
      "loss": 10.7888,
      "mean_token_accuracy": 6.958942394703626e-05,
      "num_tokens": 11445.0,
      "step": 5
    },
    {
      "entropy": 10.742593383789062,
      "epoch": 0.010085728693898134,
      "grad_norm": 5.25,
      "learning_rate": 4.5e-06,
      "loss": 10.7696,
      "mean_token_accuracy": 0.0,
      "num_tokens": 22095.0,
      "step": 10
    },
    {
      "entropy": 10.742596626281738,
      "epoch": 0.015128593040847202,
      "grad_norm": 4.5625,
      "learning_rate": 7e-06,
      "loss": 10.7579,
      "mean_token_accuracy": 0.00017376195173710585,
      "num_tokens": 33847.0,
      "step": 15
    },
    {
      "entropy": 10.742617321014404,
      "epoch": 0.020171457387796268,
      "grad_norm": 4.53125,
      "learning_rate": 9.5e-06,
      "loss": 10.7022,
      "mean_token_accuracy": 0.0003061072085984051,
      "num_tokens": 44495.0,
      "step": 20
    },
    {
      "entropy": 10.742618179321289,
      "epoch": 0.025214321734745335,
      "grad_norm": 4.0625,
      "learning_rate": 1.2e-05,
      "loss": 10.6161,
      "mean_token_accuracy": 0.0038310963835101576,
      "num_tokens": 55298.0,
      "step": 25
    },
    {
      "entropy": 10.742384719848634,
      "epoch": 0.030257186081694403,
      "grad_norm": 3.859375,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.4928,
      "mean_token_accuracy": 0.023839639127254487,
      "num_tokens": 66391.0,
      "step": 30
    },
    {
      "entropy": 10.741421985626221,
      "epoch": 0.03530005042864347,
      "grad_norm": 3.125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.3446,
      "mean_token_accuracy": 0.037043477036058904,
      "num_tokens": 77624.0,
      "step": 35
    },
    {
      "entropy": 10.738873195648193,
      "epoch": 0.040342914775592535,
      "grad_norm": 2.578125,
      "learning_rate": 1.95e-05,
      "loss": 10.2004,
      "mean_token_accuracy": 0.04290186390280724,
      "num_tokens": 89426.0,
      "step": 40
    },
    {
      "entropy": 10.734352970123291,
      "epoch": 0.0453857791225416,
      "grad_norm": 2.1875,
      "learning_rate": 2.2e-05,
      "loss": 10.1008,
      "mean_token_accuracy": 0.04177976287901401,
      "num_tokens": 100252.0,
      "step": 45
    },
    {
      "entropy": 10.731473350524903,
      "epoch": 0.05042864346949067,
      "grad_norm": 2.0625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.0294,
      "mean_token_accuracy": 0.035502530820667746,
      "num_tokens": 110578.0,
      "step": 50
    },
    {
      "entropy": 10.731419467926026,
      "epoch": 0.05547150781643974,
      "grad_norm": 1.9453125,
      "learning_rate": 2.7e-05,
      "loss": 9.9893,
      "mean_token_accuracy": 0.03719575461000204,
      "num_tokens": 121631.0,
      "step": 55
    },
    {
      "entropy": 10.730601024627685,
      "epoch": 0.060514372163388806,
      "grad_norm": 1.96875,
      "learning_rate": 2.95e-05,
      "loss": 9.9237,
      "mean_token_accuracy": 0.043491701036691664,
      "num_tokens": 134254.0,
      "step": 60
    },
    {
      "entropy": 10.727654361724854,
      "epoch": 0.06555723651033787,
      "grad_norm": 1.734375,
      "learning_rate": 3.2e-05,
      "loss": 9.8843,
      "mean_token_accuracy": 0.03803328052163124,
      "num_tokens": 144928.0,
      "step": 65
    },
    {
      "entropy": 10.724302101135255,
      "epoch": 0.07060010085728693,
      "grad_norm": 1.8203125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.7798,
      "mean_token_accuracy": 0.039696192182600495,
      "num_tokens": 155518.0,
      "step": 70
    },
    {
      "entropy": 10.720338535308837,
      "epoch": 0.07564296520423601,
      "grad_norm": 1.9375,
      "learning_rate": 3.7e-05,
      "loss": 9.7038,
      "mean_token_accuracy": 0.04226764794439077,
      "num_tokens": 165850.0,
      "step": 75
    },
    {
      "entropy": 10.71519594192505,
      "epoch": 0.08068582955118507,
      "grad_norm": 1.8828125,
      "learning_rate": 3.95e-05,
      "loss": 9.6932,
      "mean_token_accuracy": 0.04092370942234993,
      "num_tokens": 176787.0,
      "step": 80
    },
    {
      "entropy": 10.709799861907959,
      "epoch": 0.08572869389813415,
      "grad_norm": 1.7578125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.6293,
      "mean_token_accuracy": 0.03919961098581552,
      "num_tokens": 188345.0,
      "step": 85
    },
    {
      "entropy": 10.706261253356933,
      "epoch": 0.0907715582450832,
      "grad_norm": 1.78125,
      "learning_rate": 4.45e-05,
      "loss": 9.5833,
      "mean_token_accuracy": 0.03263202589005232,
      "num_tokens": 199315.0,
      "step": 90
    },
    {
      "entropy": 10.702005004882812,
      "epoch": 0.09581442259203228,
      "grad_norm": 1.7578125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.4401,
      "mean_token_accuracy": 0.04271471928805113,
      "num_tokens": 209627.0,
      "step": 95
    },
    {
      "entropy": 10.688950729370116,
      "epoch": 0.10085728693898134,
      "grad_norm": 1.8125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.3648,
      "mean_token_accuracy": 0.04892551712691784,
      "num_tokens": 220872.0,
      "step": 100
    },
    {
      "entropy": 10.666452217102051,
      "epoch": 0.1059001512859304,
      "grad_norm": 1.8125,
      "learning_rate": 5.2e-05,
      "loss": 9.2706,
      "mean_token_accuracy": 0.04733602069318295,
      "num_tokens": 233272.0,
      "step": 105
    },
    {
      "entropy": 10.643214797973632,
      "epoch": 0.11094301563287948,
      "grad_norm": 1.8125,
      "learning_rate": 5.45e-05,
      "loss": 9.1724,
      "mean_token_accuracy": 0.05065925419330597,
      "num_tokens": 243529.0,
      "step": 110
    },
    {
      "entropy": 10.622024440765381,
      "epoch": 0.11598587997982854,
      "grad_norm": 1.640625,
      "learning_rate": 5.7e-05,
      "loss": 9.0982,
      "mean_token_accuracy": 0.04518871679902077,
      "num_tokens": 254942.0,
      "step": 115
    },
    {
      "entropy": 10.591822242736816,
      "epoch": 0.12102874432677761,
      "grad_norm": 1.7421875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.9746,
      "mean_token_accuracy": 0.05078088715672493,
      "num_tokens": 266552.0,
      "step": 120
    },
    {
      "entropy": 10.53172492980957,
      "epoch": 0.1260716086737267,
      "grad_norm": 1.59375,
      "learning_rate": 6.2e-05,
      "loss": 8.8367,
      "mean_token_accuracy": 0.049286961928009985,
      "num_tokens": 277906.0,
      "step": 125
    },
    {
      "entropy": 10.461231136322022,
      "epoch": 0.13111447302067575,
      "grad_norm": 1.6328125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 8.7177,
      "mean_token_accuracy": 0.05140590891242027,
      "num_tokens": 289079.0,
      "step": 130
    },
    {
      "entropy": 10.385406398773194,
      "epoch": 0.1361573373676248,
      "grad_norm": 1.5,
      "learning_rate": 6.7e-05,
      "loss": 8.6441,
      "mean_token_accuracy": 0.04429629724472761,
      "num_tokens": 300107.0,
      "step": 135
    },
    {
      "entropy": 10.312572288513184,
      "epoch": 0.14120020171457387,
      "grad_norm": 1.5234375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.4806,
      "mean_token_accuracy": 0.05319313518702984,
      "num_tokens": 310779.0,
      "step": 140
    },
    {
      "entropy": 10.203822803497314,
      "epoch": 0.14624306606152296,
      "grad_norm": 1.453125,
      "learning_rate": 7.2e-05,
      "loss": 8.4202,
      "mean_token_accuracy": 0.051876432821154596,
      "num_tokens": 323478.0,
      "step": 145
    },
    {
      "entropy": 10.096873664855957,
      "epoch": 0.15128593040847202,
      "grad_norm": 1.3515625,
      "learning_rate": 7.45e-05,
      "loss": 8.2552,
      "mean_token_accuracy": 0.05630794763565063,
      "num_tokens": 335255.0,
      "step": 150
    },
    {
      "entropy": 9.960623264312744,
      "epoch": 0.15632879475542108,
      "grad_norm": 1.3046875,
      "learning_rate": 7.7e-05,
      "loss": 8.1385,
      "mean_token_accuracy": 0.05476163513958454,
      "num_tokens": 344824.0,
      "step": 155
    },
    {
      "entropy": 9.771032524108886,
      "epoch": 0.16137165910237014,
      "grad_norm": 1.265625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.0754,
      "mean_token_accuracy": 0.05333587229251861,
      "num_tokens": 356283.0,
      "step": 160
    },
    {
      "entropy": 9.623997592926026,
      "epoch": 0.1664145234493192,
      "grad_norm": 1.015625,
      "learning_rate": 8.2e-05,
      "loss": 7.996,
      "mean_token_accuracy": 0.05669583305716515,
      "num_tokens": 368098.0,
      "step": 165
    },
    {
      "entropy": 9.383370685577393,
      "epoch": 0.1714573877962683,
      "grad_norm": 1.03125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8796,
      "mean_token_accuracy": 0.059235493466258046,
      "num_tokens": 380948.0,
      "step": 170
    },
    {
      "entropy": 9.174373054504395,
      "epoch": 0.17650025214321735,
      "grad_norm": 1.0234375,
      "learning_rate": 8.7e-05,
      "loss": 7.7633,
      "mean_token_accuracy": 0.05942847430706024,
      "num_tokens": 390392.0,
      "step": 175
    },
    {
      "entropy": 8.980836009979248,
      "epoch": 0.1815431164901664,
      "grad_norm": 1.3203125,
      "learning_rate": 8.95e-05,
      "loss": 7.7496,
      "mean_token_accuracy": 0.06362583562731743,
      "num_tokens": 400328.0,
      "step": 180
    },
    {
      "entropy": 8.7655029296875,
      "epoch": 0.18658598083711547,
      "grad_norm": 0.828125,
      "learning_rate": 9.2e-05,
      "loss": 7.7544,
      "mean_token_accuracy": 0.06182343065738678,
      "num_tokens": 410909.0,
      "step": 185
    },
    {
      "entropy": 8.639219570159913,
      "epoch": 0.19162884518406456,
      "grad_norm": 0.9375,
      "learning_rate": 9.45e-05,
      "loss": 7.691,
      "mean_token_accuracy": 0.06425628773868083,
      "num_tokens": 422726.0,
      "step": 190
    },
    {
      "entropy": 8.559172058105469,
      "epoch": 0.19667170953101362,
      "grad_norm": 0.91796875,
      "learning_rate": 9.7e-05,
      "loss": 7.5594,
      "mean_token_accuracy": 0.07219541147351265,
      "num_tokens": 433376.0,
      "step": 195
    },
    {
      "entropy": 8.397376251220702,
      "epoch": 0.20171457387796268,
      "grad_norm": 1.0078125,
      "learning_rate": 9.95e-05,
      "loss": 7.6275,
      "mean_token_accuracy": 0.06337097361683845,
      "num_tokens": 443940.0,
      "step": 200
    },
    {
      "entropy": 8.361763286590577,
      "epoch": 0.20675743822491174,
      "grad_norm": 0.78125,
      "learning_rate": 0.000102,
      "loss": 7.5213,
      "mean_token_accuracy": 0.06769266054034233,
      "num_tokens": 455016.0,
      "step": 205
    },
    {
      "entropy": 8.303247165679931,
      "epoch": 0.2118003025718608,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.553,
      "mean_token_accuracy": 0.06669275388121605,
      "num_tokens": 466380.0,
      "step": 210
    },
    {
      "entropy": 8.23162612915039,
      "epoch": 0.2168431669188099,
      "grad_norm": 1.140625,
      "learning_rate": 0.000107,
      "loss": 7.4065,
      "mean_token_accuracy": 0.07161542177200317,
      "num_tokens": 476915.0,
      "step": 215
    },
    {
      "entropy": 8.226816177368164,
      "epoch": 0.22188603126575895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001095,
      "loss": 7.4817,
      "mean_token_accuracy": 0.07145784981548786,
      "num_tokens": 488924.0,
      "step": 220
    },
    {
      "entropy": 8.18234987258911,
      "epoch": 0.22692889561270801,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000112,
      "loss": 7.4263,
      "mean_token_accuracy": 0.07183554545044898,
      "num_tokens": 498965.0,
      "step": 225
    },
    {
      "entropy": 8.137568473815918,
      "epoch": 0.23197175995965708,
      "grad_norm": 0.828125,
      "learning_rate": 0.0001145,
      "loss": 7.4409,
      "mean_token_accuracy": 0.07232684977352619,
      "num_tokens": 509755.0,
      "step": 230
    },
    {
      "entropy": 8.12470645904541,
      "epoch": 0.23701462430660616,
      "grad_norm": 0.796875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.4064,
      "mean_token_accuracy": 0.07139695808291435,
      "num_tokens": 521221.0,
      "step": 235
    },
    {
      "entropy": 8.080301761627197,
      "epoch": 0.24205748865355523,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.503,
      "mean_token_accuracy": 0.06776785589754582,
      "num_tokens": 533595.0,
      "step": 240
    },
    {
      "entropy": 8.111075401306152,
      "epoch": 0.24710035300050429,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000122,
      "loss": 7.4348,
      "mean_token_accuracy": 0.07060522697865963,
      "num_tokens": 543672.0,
      "step": 245
    },
    {
      "entropy": 8.046464586257935,
      "epoch": 0.2521432173474534,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0001245,
      "loss": 7.3208,
      "mean_token_accuracy": 0.07489579617977142,
      "num_tokens": 554906.0,
      "step": 250
    },
    {
      "entropy": 7.984986257553101,
      "epoch": 0.25718608169440244,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000127,
      "loss": 7.3941,
      "mean_token_accuracy": 0.07390029206871987,
      "num_tokens": 566419.0,
      "step": 255
    },
    {
      "entropy": 8.038315868377685,
      "epoch": 0.2622289460413515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0001295,
      "loss": 7.3625,
      "mean_token_accuracy": 0.07229974381625652,
      "num_tokens": 578848.0,
      "step": 260
    },
    {
      "entropy": 7.896354866027832,
      "epoch": 0.26727181038830056,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000132,
      "loss": 7.305,
      "mean_token_accuracy": 0.07605222910642624,
      "num_tokens": 590601.0,
      "step": 265
    },
    {
      "entropy": 7.9424131393432615,
      "epoch": 0.2723146747352496,
      "grad_norm": 0.9375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.3548,
      "mean_token_accuracy": 0.07202886901795864,
      "num_tokens": 602021.0,
      "step": 270
    },
    {
      "entropy": 8.004773712158203,
      "epoch": 0.2773575390821987,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.4443,
      "mean_token_accuracy": 0.06653593964874745,
      "num_tokens": 613966.0,
      "step": 275
    },
    {
      "entropy": 7.990764474868774,
      "epoch": 0.28240040342914774,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001395,
      "loss": 7.2762,
      "mean_token_accuracy": 0.07293831780552865,
      "num_tokens": 624851.0,
      "step": 280
    },
    {
      "entropy": 7.808479070663452,
      "epoch": 0.2874432677760968,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.3518,
      "mean_token_accuracy": 0.07462373450398445,
      "num_tokens": 635580.0,
      "step": 285
    },
    {
      "entropy": 7.965742826461792,
      "epoch": 0.2924861321230459,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0001445,
      "loss": 7.4071,
      "mean_token_accuracy": 0.0715030949562788,
      "num_tokens": 646927.0,
      "step": 290
    },
    {
      "entropy": 7.943162631988526,
      "epoch": 0.297528996469995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000147,
      "loss": 7.2414,
      "mean_token_accuracy": 0.07769636139273643,
      "num_tokens": 658116.0,
      "step": 295
    },
    {
      "entropy": 7.861419916152954,
      "epoch": 0.30257186081694404,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001495,
      "loss": 7.2404,
      "mean_token_accuracy": 0.07680558413267136,
      "num_tokens": 669054.0,
      "step": 300
    },
    {
      "entropy": 7.859039735794068,
      "epoch": 0.3076147251638931,
      "grad_norm": 1.203125,
      "learning_rate": 0.000152,
      "loss": 7.2846,
      "mean_token_accuracy": 0.07930284813046455,
      "num_tokens": 680439.0,
      "step": 305
    },
    {
      "entropy": 7.81926383972168,
      "epoch": 0.31265758951084216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.2209,
      "mean_token_accuracy": 0.0794371448457241,
      "num_tokens": 690909.0,
      "step": 310
    },
    {
      "entropy": 7.75442066192627,
      "epoch": 0.3177004538577912,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000157,
      "loss": 7.209,
      "mean_token_accuracy": 0.07925689667463302,
      "num_tokens": 701744.0,
      "step": 315
    },
    {
      "entropy": 7.746679449081421,
      "epoch": 0.3227433182047403,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0001595,
      "loss": 7.2338,
      "mean_token_accuracy": 0.08032149039208888,
      "num_tokens": 712268.0,
      "step": 320
    },
    {
      "entropy": 7.765822124481201,
      "epoch": 0.32778618255168934,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000162,
      "loss": 7.2609,
      "mean_token_accuracy": 0.07947171069681644,
      "num_tokens": 722978.0,
      "step": 325
    },
    {
      "entropy": 7.870586299896241,
      "epoch": 0.3328290468986384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.253,
      "mean_token_accuracy": 0.07748774662613869,
      "num_tokens": 734278.0,
      "step": 330
    },
    {
      "entropy": 7.8009439468383786,
      "epoch": 0.3378719112455875,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.1867,
      "mean_token_accuracy": 0.07406368069350719,
      "num_tokens": 746109.0,
      "step": 335
    },
    {
      "entropy": 7.751057529449463,
      "epoch": 0.3429147755925366,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.1104,
      "mean_token_accuracy": 0.07908297292888164,
      "num_tokens": 757519.0,
      "step": 340
    },
    {
      "entropy": 7.711837816238403,
      "epoch": 0.34795763993948564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.0784,
      "mean_token_accuracy": 0.08319444209337234,
      "num_tokens": 767665.0,
      "step": 345
    },
    {
      "entropy": 7.729198932647705,
      "epoch": 0.3530005042864347,
      "grad_norm": 1.125,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.118,
      "mean_token_accuracy": 0.08570092767477036,
      "num_tokens": 778880.0,
      "step": 350
    },
    {
      "entropy": 7.589126014709473,
      "epoch": 0.35804336863338376,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000177,
      "loss": 7.1275,
      "mean_token_accuracy": 0.08408626839518547,
      "num_tokens": 790360.0,
      "step": 355
    },
    {
      "entropy": 7.650224447250366,
      "epoch": 0.3630862329803328,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001795,
      "loss": 7.0884,
      "mean_token_accuracy": 0.08589479774236679,
      "num_tokens": 800889.0,
      "step": 360
    },
    {
      "entropy": 7.698587417602539,
      "epoch": 0.3681290973272819,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000182,
      "loss": 7.0774,
      "mean_token_accuracy": 0.08224210813641548,
      "num_tokens": 811414.0,
      "step": 365
    },
    {
      "entropy": 7.551970529556274,
      "epoch": 0.37317196167423095,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0001845,
      "loss": 7.0729,
      "mean_token_accuracy": 0.08361641615629196,
      "num_tokens": 822822.0,
      "step": 370
    },
    {
      "entropy": 7.61205792427063,
      "epoch": 0.37821482602118,
      "grad_norm": 1.09375,
      "learning_rate": 0.000187,
      "loss": 6.9949,
      "mean_token_accuracy": 0.09048431143164634,
      "num_tokens": 833145.0,
      "step": 375
    },
    {
      "entropy": 7.606620168685913,
      "epoch": 0.3832576903681291,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001895,
      "loss": 7.1067,
      "mean_token_accuracy": 0.08064279481768608,
      "num_tokens": 844015.0,
      "step": 380
    },
    {
      "entropy": 7.537594938278199,
      "epoch": 0.3883005547150782,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000192,
      "loss": 7.0227,
      "mean_token_accuracy": 0.08781845718622208,
      "num_tokens": 855806.0,
      "step": 385
    },
    {
      "entropy": 7.601389598846436,
      "epoch": 0.39334341906202724,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001945,
      "loss": 6.952,
      "mean_token_accuracy": 0.08714486509561539,
      "num_tokens": 865750.0,
      "step": 390
    },
    {
      "entropy": 7.603587245941162,
      "epoch": 0.3983862834089763,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.0798,
      "mean_token_accuracy": 0.08171837627887726,
      "num_tokens": 877630.0,
      "step": 395
    },
    {
      "entropy": 7.46077446937561,
      "epoch": 0.40342914775592537,
      "grad_norm": 0.921875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.9621,
      "mean_token_accuracy": 0.08956382796168327,
      "num_tokens": 888667.0,
      "step": 400
    },
    {
      "entropy": 7.612736940383911,
      "epoch": 0.4084720121028744,
      "grad_norm": 1.0625,
      "learning_rate": 0.000202,
      "loss": 7.0383,
      "mean_token_accuracy": 0.0815612755715847,
      "num_tokens": 900989.0,
      "step": 405
    },
    {
      "entropy": 7.4917103290557865,
      "epoch": 0.4135148764498235,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.9247,
      "mean_token_accuracy": 0.08268933743238449,
      "num_tokens": 911479.0,
      "step": 410
    },
    {
      "entropy": 7.538983345031738,
      "epoch": 0.41855774079677255,
      "grad_norm": 0.96875,
      "learning_rate": 0.000207,
      "loss": 6.9579,
      "mean_token_accuracy": 0.0860539086163044,
      "num_tokens": 922434.0,
      "step": 415
    },
    {
      "entropy": 7.502885150909424,
      "epoch": 0.4236006051437216,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002095,
      "loss": 6.9417,
      "mean_token_accuracy": 0.08750939816236496,
      "num_tokens": 933405.0,
      "step": 420
    },
    {
      "entropy": 7.503132390975952,
      "epoch": 0.4286434694906707,
      "grad_norm": 1.09375,
      "learning_rate": 0.000212,
      "loss": 6.9414,
      "mean_token_accuracy": 0.0863698624074459,
      "num_tokens": 943567.0,
      "step": 425
    },
    {
      "entropy": 7.433593511581421,
      "epoch": 0.4336863338376198,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002145,
      "loss": 6.9729,
      "mean_token_accuracy": 0.08398969918489456,
      "num_tokens": 955123.0,
      "step": 430
    },
    {
      "entropy": 7.4864178657531735,
      "epoch": 0.43872919818456885,
      "grad_norm": 1.140625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.9878,
      "mean_token_accuracy": 0.08637007214128971,
      "num_tokens": 965565.0,
      "step": 435
    },
    {
      "entropy": 7.533525228500366,
      "epoch": 0.4437720625315179,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002195,
      "loss": 6.9331,
      "mean_token_accuracy": 0.09186888933181762,
      "num_tokens": 977101.0,
      "step": 440
    },
    {
      "entropy": 7.447479343414306,
      "epoch": 0.44881492687846697,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000222,
      "loss": 6.9676,
      "mean_token_accuracy": 0.08773500472307205,
      "num_tokens": 987445.0,
      "step": 445
    },
    {
      "entropy": 7.413919162750244,
      "epoch": 0.45385779122541603,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002245,
      "loss": 6.8527,
      "mean_token_accuracy": 0.0943599946796894,
      "num_tokens": 999379.0,
      "step": 450
    },
    {
      "entropy": 7.366581916809082,
      "epoch": 0.4589006555723651,
      "grad_norm": 1.0625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7887,
      "mean_token_accuracy": 0.09886589571833611,
      "num_tokens": 1010744.0,
      "step": 455
    },
    {
      "entropy": 7.395628499984741,
      "epoch": 0.46394351991931415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8751,
      "mean_token_accuracy": 0.09002586975693702,
      "num_tokens": 1021808.0,
      "step": 460
    },
    {
      "entropy": 7.436336374282837,
      "epoch": 0.4689863842662632,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.9197,
      "mean_token_accuracy": 0.08965835645794869,
      "num_tokens": 1033694.0,
      "step": 465
    },
    {
      "entropy": 7.366260051727295,
      "epoch": 0.47402924861321233,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8866,
      "mean_token_accuracy": 0.08940717950463295,
      "num_tokens": 1044690.0,
      "step": 470
    },
    {
      "entropy": 7.3759284019470215,
      "epoch": 0.4790721129601614,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000237,
      "loss": 6.883,
      "mean_token_accuracy": 0.09023785665631294,
      "num_tokens": 1055557.0,
      "step": 475
    },
    {
      "entropy": 7.412348127365112,
      "epoch": 0.48411497730711045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002395,
      "loss": 6.8225,
      "mean_token_accuracy": 0.09234220460057259,
      "num_tokens": 1067304.0,
      "step": 480
    },
    {
      "entropy": 7.280447006225586,
      "epoch": 0.4891578416540595,
      "grad_norm": 1.15625,
      "learning_rate": 0.000242,
      "loss": 6.8498,
      "mean_token_accuracy": 0.09138206690549851,
      "num_tokens": 1078955.0,
      "step": 485
    },
    {
      "entropy": 7.339125728607177,
      "epoch": 0.49420070600100857,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002445,
      "loss": 6.8907,
      "mean_token_accuracy": 0.09167812019586563,
      "num_tokens": 1091405.0,
      "step": 490
    },
    {
      "entropy": 7.334241151809692,
      "epoch": 0.49924357034795763,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000247,
      "loss": 6.8625,
      "mean_token_accuracy": 0.09363722130656242,
      "num_tokens": 1103227.0,
      "step": 495
    },
    {
      "entropy": 7.351650381088257,
      "epoch": 0.5042864346949067,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002495,
      "loss": 6.7968,
      "mean_token_accuracy": 0.09819836765527726,
      "num_tokens": 1114485.0,
      "step": 500
    },
    {
      "epoch": 0.5042864346949067,
      "eval_entropy": 7.210145644559205,
      "eval_loss": 6.899664402008057,
      "eval_mean_token_accuracy": 0.09479910408267538,
      "eval_num_tokens": 1114485.0,
      "eval_runtime": 2.1175,
      "eval_samples_per_second": 1481.964,
      "eval_steps_per_second": 185.6,
      "step": 500
    },
    {
      "entropy": 7.260896873474121,
      "epoch": 0.5093292990418558,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000252,
      "loss": 6.8181,
      "mean_token_accuracy": 0.09441837519407273,
      "num_tokens": 1126789.0,
      "step": 505
    },
    {
      "entropy": 7.352414274215699,
      "epoch": 0.5143721633888049,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002545,
      "loss": 6.7989,
      "mean_token_accuracy": 0.09330016896128654,
      "num_tokens": 1138192.0,
      "step": 510
    },
    {
      "entropy": 7.2082147121429445,
      "epoch": 0.5194150277357539,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000257,
      "loss": 6.7411,
      "mean_token_accuracy": 0.09130406081676483,
      "num_tokens": 1148526.0,
      "step": 515
    },
    {
      "entropy": 7.369548559188843,
      "epoch": 0.524457892082703,
      "grad_norm": 1.0,
      "learning_rate": 0.0002595,
      "loss": 6.8496,
      "mean_token_accuracy": 0.08768555521965027,
      "num_tokens": 1159194.0,
      "step": 520
    },
    {
      "entropy": 7.23188009262085,
      "epoch": 0.529500756429652,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000262,
      "loss": 6.7433,
      "mean_token_accuracy": 0.1001415841281414,
      "num_tokens": 1170601.0,
      "step": 525
    },
    {
      "entropy": 7.288579416275025,
      "epoch": 0.5345436207766011,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7419,
      "mean_token_accuracy": 0.0954306311905384,
      "num_tokens": 1182441.0,
      "step": 530
    },
    {
      "entropy": 7.17895131111145,
      "epoch": 0.5395864851235502,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6349,
      "mean_token_accuracy": 0.10441381633281707,
      "num_tokens": 1192530.0,
      "step": 535
    },
    {
      "entropy": 7.13337082862854,
      "epoch": 0.5446293494704992,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.7521,
      "mean_token_accuracy": 0.0956033818423748,
      "num_tokens": 1203831.0,
      "step": 540
    },
    {
      "entropy": 7.2454320907592775,
      "epoch": 0.5496722138174484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6923,
      "mean_token_accuracy": 0.0969244122505188,
      "num_tokens": 1214164.0,
      "step": 545
    },
    {
      "entropy": 7.150829601287842,
      "epoch": 0.5547150781643974,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002745,
      "loss": 6.7235,
      "mean_token_accuracy": 0.09548334404826164,
      "num_tokens": 1226932.0,
      "step": 550
    },
    {
      "entropy": 7.283726549148559,
      "epoch": 0.5597579425113465,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000277,
      "loss": 6.79,
      "mean_token_accuracy": 0.08873454928398132,
      "num_tokens": 1237212.0,
      "step": 555
    },
    {
      "entropy": 7.143116188049317,
      "epoch": 0.5648008068582955,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002795,
      "loss": 6.6907,
      "mean_token_accuracy": 0.09660674557089806,
      "num_tokens": 1249340.0,
      "step": 560
    },
    {
      "entropy": 7.123078393936157,
      "epoch": 0.5698436712052446,
      "grad_norm": 1.03125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6907,
      "mean_token_accuracy": 0.09954180717468261,
      "num_tokens": 1260778.0,
      "step": 565
    },
    {
      "entropy": 7.1541259765625,
      "epoch": 0.5748865355521936,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0002845,
      "loss": 6.6986,
      "mean_token_accuracy": 0.09604307189583779,
      "num_tokens": 1272216.0,
      "step": 570
    },
    {
      "entropy": 7.223128890991211,
      "epoch": 0.5799293998991427,
      "grad_norm": 1.46875,
      "learning_rate": 0.000287,
      "loss": 6.7725,
      "mean_token_accuracy": 0.09934922829270362,
      "num_tokens": 1283558.0,
      "step": 575
    },
    {
      "entropy": 7.209529066085816,
      "epoch": 0.5849722642460918,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002895,
      "loss": 6.834,
      "mean_token_accuracy": 0.08799525387585164,
      "num_tokens": 1294638.0,
      "step": 580
    },
    {
      "entropy": 7.27526068687439,
      "epoch": 0.5900151285930408,
      "grad_norm": 1.15625,
      "learning_rate": 0.000292,
      "loss": 6.7472,
      "mean_token_accuracy": 0.09616359025239944,
      "num_tokens": 1304697.0,
      "step": 585
    },
    {
      "entropy": 7.173431634902954,
      "epoch": 0.59505799293999,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002945,
      "loss": 6.7033,
      "mean_token_accuracy": 0.09485772103071213,
      "num_tokens": 1316618.0,
      "step": 590
    },
    {
      "entropy": 7.151233100891114,
      "epoch": 0.600100857286939,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000297,
      "loss": 6.8116,
      "mean_token_accuracy": 0.08478758260607719,
      "num_tokens": 1328134.0,
      "step": 595
    },
    {
      "entropy": 7.242454576492309,
      "epoch": 0.6051437216338881,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0002995,
      "loss": 6.7227,
      "mean_token_accuracy": 0.0935085467994213,
      "num_tokens": 1339101.0,
      "step": 600
    },
    {
      "entropy": 7.143788623809814,
      "epoch": 0.6101865859808371,
      "grad_norm": 0.984375,
      "learning_rate": 0.000302,
      "loss": 6.5973,
      "mean_token_accuracy": 0.09975283592939377,
      "num_tokens": 1349181.0,
      "step": 605
    },
    {
      "entropy": 7.0980949878692625,
      "epoch": 0.6152294503277862,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0003045,
      "loss": 6.7221,
      "mean_token_accuracy": 0.09565178751945495,
      "num_tokens": 1361443.0,
      "step": 610
    },
    {
      "entropy": 7.098438501358032,
      "epoch": 0.6202723146747352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000307,
      "loss": 6.6778,
      "mean_token_accuracy": 0.10161929428577424,
      "num_tokens": 1372807.0,
      "step": 615
    },
    {
      "entropy": 7.160452890396118,
      "epoch": 0.6253151790216843,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003095,
      "loss": 6.7426,
      "mean_token_accuracy": 0.09811687916517257,
      "num_tokens": 1385509.0,
      "step": 620
    },
    {
      "entropy": 7.133578205108643,
      "epoch": 0.6303580433686334,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000312,
      "loss": 6.6405,
      "mean_token_accuracy": 0.0968807429075241,
      "num_tokens": 1396035.0,
      "step": 625
    },
    {
      "entropy": 7.047602128982544,
      "epoch": 0.6354009077155824,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003145,
      "loss": 6.5876,
      "mean_token_accuracy": 0.10460086688399314,
      "num_tokens": 1406839.0,
      "step": 630
    },
    {
      "entropy": 6.976920652389526,
      "epoch": 0.6404437720625316,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000317,
      "loss": 6.559,
      "mean_token_accuracy": 0.09928513914346696,
      "num_tokens": 1417708.0,
      "step": 635
    },
    {
      "entropy": 7.116953611373901,
      "epoch": 0.6454866364094806,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003195,
      "loss": 6.6878,
      "mean_token_accuracy": 0.0931871697306633,
      "num_tokens": 1428985.0,
      "step": 640
    },
    {
      "entropy": 7.110161256790161,
      "epoch": 0.6505295007564297,
      "grad_norm": 1.25,
      "learning_rate": 0.000322,
      "loss": 6.7068,
      "mean_token_accuracy": 0.10163840278983116,
      "num_tokens": 1440232.0,
      "step": 645
    },
    {
      "entropy": 6.962800979614258,
      "epoch": 0.6555723651033787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4498,
      "mean_token_accuracy": 0.1053794801235199,
      "num_tokens": 1451735.0,
      "step": 650
    },
    {
      "entropy": 7.015639686584473,
      "epoch": 0.6606152294503278,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.6221,
      "mean_token_accuracy": 0.09871228262782097,
      "num_tokens": 1463477.0,
      "step": 655
    },
    {
      "entropy": 7.01046838760376,
      "epoch": 0.6656580937972768,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.6263,
      "mean_token_accuracy": 0.09373534545302391,
      "num_tokens": 1473802.0,
      "step": 660
    },
    {
      "entropy": 7.009602022171021,
      "epoch": 0.6707009581442259,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.6264,
      "mean_token_accuracy": 0.09858302772045135,
      "num_tokens": 1485519.0,
      "step": 665
    },
    {
      "entropy": 7.069497060775757,
      "epoch": 0.675743822491175,
      "grad_norm": 1.015625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6323,
      "mean_token_accuracy": 0.09767952933907509,
      "num_tokens": 1496957.0,
      "step": 670
    },
    {
      "entropy": 7.0134529113769535,
      "epoch": 0.680786686838124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000337,
      "loss": 6.6387,
      "mean_token_accuracy": 0.10315296724438668,
      "num_tokens": 1508336.0,
      "step": 675
    },
    {
      "entropy": 6.98244080543518,
      "epoch": 0.6858295511850732,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003395,
      "loss": 6.6118,
      "mean_token_accuracy": 0.09836572855710983,
      "num_tokens": 1520016.0,
      "step": 680
    },
    {
      "entropy": 6.9399433612823485,
      "epoch": 0.6908724155320222,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000342,
      "loss": 6.4545,
      "mean_token_accuracy": 0.10555818825960159,
      "num_tokens": 1531877.0,
      "step": 685
    },
    {
      "entropy": 6.861659812927246,
      "epoch": 0.6959152798789713,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4507,
      "mean_token_accuracy": 0.10879426971077918,
      "num_tokens": 1541553.0,
      "step": 690
    },
    {
      "entropy": 6.983822345733643,
      "epoch": 0.7009581442259203,
      "grad_norm": 1.015625,
      "learning_rate": 0.000347,
      "loss": 6.6146,
      "mean_token_accuracy": 0.10361509844660759,
      "num_tokens": 1552491.0,
      "step": 695
    },
    {
      "entropy": 6.899546098709107,
      "epoch": 0.7060010085728694,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003495,
      "loss": 6.5076,
      "mean_token_accuracy": 0.10220511853694916,
      "num_tokens": 1563695.0,
      "step": 700
    },
    {
      "entropy": 6.986747407913208,
      "epoch": 0.7110438729198184,
      "grad_norm": 1.140625,
      "learning_rate": 0.000352,
      "loss": 6.5457,
      "mean_token_accuracy": 0.1040324755012989,
      "num_tokens": 1573467.0,
      "step": 705
    },
    {
      "entropy": 6.899123620986939,
      "epoch": 0.7160867372667675,
      "grad_norm": 0.953125,
      "learning_rate": 0.0003545,
      "loss": 6.523,
      "mean_token_accuracy": 0.1107451319694519,
      "num_tokens": 1584367.0,
      "step": 710
    },
    {
      "entropy": 6.902140855789185,
      "epoch": 0.7211296016137166,
      "grad_norm": 1.0625,
      "learning_rate": 0.000357,
      "loss": 6.4647,
      "mean_token_accuracy": 0.11195559501647949,
      "num_tokens": 1595301.0,
      "step": 715
    },
    {
      "entropy": 6.908828496932983,
      "epoch": 0.7261724659606656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003595,
      "loss": 6.5362,
      "mean_token_accuracy": 0.10281007215380669,
      "num_tokens": 1605916.0,
      "step": 720
    },
    {
      "entropy": 6.960860681533814,
      "epoch": 0.7312153303076148,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000362,
      "loss": 6.5308,
      "mean_token_accuracy": 0.10332813784480095,
      "num_tokens": 1616455.0,
      "step": 725
    },
    {
      "entropy": 6.853022289276123,
      "epoch": 0.7362581946545638,
      "grad_norm": 1.0,
      "learning_rate": 0.0003645,
      "loss": 6.4979,
      "mean_token_accuracy": 0.10751453042030334,
      "num_tokens": 1627396.0,
      "step": 730
    },
    {
      "entropy": 7.0525219440460205,
      "epoch": 0.7413010590015129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000367,
      "loss": 6.6317,
      "mean_token_accuracy": 0.09941819235682488,
      "num_tokens": 1639073.0,
      "step": 735
    },
    {
      "entropy": 6.803125476837158,
      "epoch": 0.7463439233484619,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003695,
      "loss": 6.4978,
      "mean_token_accuracy": 0.10190683454275132,
      "num_tokens": 1649787.0,
      "step": 740
    },
    {
      "entropy": 6.935564804077148,
      "epoch": 0.751386787695411,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000372,
      "loss": 6.4383,
      "mean_token_accuracy": 0.10585329979658127,
      "num_tokens": 1658883.0,
      "step": 745
    },
    {
      "entropy": 6.7616808891296385,
      "epoch": 0.75642965204236,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003745,
      "loss": 6.4186,
      "mean_token_accuracy": 0.11256190687417984,
      "num_tokens": 1669099.0,
      "step": 750
    },
    {
      "entropy": 6.824008131027222,
      "epoch": 0.7614725163893091,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000377,
      "loss": 6.4199,
      "mean_token_accuracy": 0.10491297394037247,
      "num_tokens": 1680227.0,
      "step": 755
    },
    {
      "entropy": 6.830388927459717,
      "epoch": 0.7665153807362582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003795,
      "loss": 6.4178,
      "mean_token_accuracy": 0.10797212198376656,
      "num_tokens": 1689901.0,
      "step": 760
    },
    {
      "entropy": 6.790120553970337,
      "epoch": 0.7715582450832073,
      "grad_norm": 1.0625,
      "learning_rate": 0.000382,
      "loss": 6.4932,
      "mean_token_accuracy": 0.10378503128886223,
      "num_tokens": 1700498.0,
      "step": 765
    },
    {
      "entropy": 6.854589986801147,
      "epoch": 0.7766011094301564,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003845,
      "loss": 6.3993,
      "mean_token_accuracy": 0.1122807152569294,
      "num_tokens": 1710715.0,
      "step": 770
    },
    {
      "entropy": 6.860204267501831,
      "epoch": 0.7816439737771054,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5261,
      "mean_token_accuracy": 0.106892429292202,
      "num_tokens": 1723858.0,
      "step": 775
    },
    {
      "entropy": 6.753562068939209,
      "epoch": 0.7866868381240545,
      "grad_norm": 1.03125,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.491,
      "mean_token_accuracy": 0.10414958521723747,
      "num_tokens": 1734678.0,
      "step": 780
    },
    {
      "entropy": 6.837999296188355,
      "epoch": 0.7917297024710035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3922,
      "mean_token_accuracy": 0.11342072263360023,
      "num_tokens": 1743869.0,
      "step": 785
    },
    {
      "entropy": 6.905572748184204,
      "epoch": 0.7967725668179526,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4633,
      "mean_token_accuracy": 0.10736536532640457,
      "num_tokens": 1755266.0,
      "step": 790
    },
    {
      "entropy": 6.693229627609253,
      "epoch": 0.8018154311649016,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4254,
      "mean_token_accuracy": 0.1047304928302765,
      "num_tokens": 1766089.0,
      "step": 795
    },
    {
      "entropy": 6.800949335098267,
      "epoch": 0.8068582955118507,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003995,
      "loss": 6.441,
      "mean_token_accuracy": 0.10456803143024444,
      "num_tokens": 1776057.0,
      "step": 800
    },
    {
      "entropy": 6.839222145080567,
      "epoch": 0.8119011598587998,
      "grad_norm": 1.046875,
      "learning_rate": 0.000402,
      "loss": 6.4781,
      "mean_token_accuracy": 0.10709379687905311,
      "num_tokens": 1786729.0,
      "step": 805
    },
    {
      "entropy": 6.80726170539856,
      "epoch": 0.8169440242057489,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004045,
      "loss": 6.408,
      "mean_token_accuracy": 0.10803752541542053,
      "num_tokens": 1797708.0,
      "step": 810
    },
    {
      "entropy": 6.7316183090209964,
      "epoch": 0.821986888552698,
      "grad_norm": 0.96875,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4408,
      "mean_token_accuracy": 0.11508865803480148,
      "num_tokens": 1807326.0,
      "step": 815
    },
    {
      "entropy": 6.7365195751190186,
      "epoch": 0.827029752899647,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004095,
      "loss": 6.3771,
      "mean_token_accuracy": 0.11316760107874871,
      "num_tokens": 1817266.0,
      "step": 820
    },
    {
      "entropy": 6.765561532974243,
      "epoch": 0.8320726172465961,
      "grad_norm": 0.890625,
      "learning_rate": 0.000412,
      "loss": 6.3975,
      "mean_token_accuracy": 0.11348253637552261,
      "num_tokens": 1827840.0,
      "step": 825
    },
    {
      "entropy": 6.61966724395752,
      "epoch": 0.8371154815935451,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004145,
      "loss": 6.3751,
      "mean_token_accuracy": 0.10820141658186913,
      "num_tokens": 1838158.0,
      "step": 830
    },
    {
      "entropy": 6.799835348129273,
      "epoch": 0.8421583459404942,
      "grad_norm": 1.03125,
      "learning_rate": 0.000417,
      "loss": 6.4544,
      "mean_token_accuracy": 0.10612540319561958,
      "num_tokens": 1848317.0,
      "step": 835
    },
    {
      "entropy": 6.711783790588379,
      "epoch": 0.8472012102874432,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004195,
      "loss": 6.3855,
      "mean_token_accuracy": 0.10564651787281036,
      "num_tokens": 1858751.0,
      "step": 840
    },
    {
      "entropy": 6.7946678638458256,
      "epoch": 0.8522440746343923,
      "grad_norm": 0.96875,
      "learning_rate": 0.000422,
      "loss": 6.3903,
      "mean_token_accuracy": 0.1101529210805893,
      "num_tokens": 1870739.0,
      "step": 845
    },
    {
      "entropy": 6.8033802032470705,
      "epoch": 0.8572869389813415,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004245,
      "loss": 6.5132,
      "mean_token_accuracy": 0.11032365560531616,
      "num_tokens": 1880995.0,
      "step": 850
    },
    {
      "entropy": 6.705736064910889,
      "epoch": 0.8623298033282905,
      "grad_norm": 1.09375,
      "learning_rate": 0.000427,
      "loss": 6.4509,
      "mean_token_accuracy": 0.10846404731273651,
      "num_tokens": 1892633.0,
      "step": 855
    },
    {
      "entropy": 6.8639147758483885,
      "epoch": 0.8673726676752396,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004295,
      "loss": 6.5176,
      "mean_token_accuracy": 0.10401368141174316,
      "num_tokens": 1904802.0,
      "step": 860
    },
    {
      "entropy": 6.778040790557862,
      "epoch": 0.8724155320221886,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000432,
      "loss": 6.5199,
      "mean_token_accuracy": 0.10730656683444977,
      "num_tokens": 1915755.0,
      "step": 865
    },
    {
      "entropy": 6.688451814651489,
      "epoch": 0.8774583963691377,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004345,
      "loss": 6.2652,
      "mean_token_accuracy": 0.11115367412567138,
      "num_tokens": 1926098.0,
      "step": 870
    },
    {
      "entropy": 6.744540882110596,
      "epoch": 0.8825012607160867,
      "grad_norm": 1.03125,
      "learning_rate": 0.000437,
      "loss": 6.4883,
      "mean_token_accuracy": 0.11085206493735314,
      "num_tokens": 1936607.0,
      "step": 875
    },
    {
      "entropy": 6.703429174423218,
      "epoch": 0.8875441250630358,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004395,
      "loss": 6.377,
      "mean_token_accuracy": 0.10893332436680794,
      "num_tokens": 1948596.0,
      "step": 880
    },
    {
      "entropy": 6.669404411315918,
      "epoch": 0.8925869894099848,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000442,
      "loss": 6.4526,
      "mean_token_accuracy": 0.11500444263219833,
      "num_tokens": 1960322.0,
      "step": 885
    },
    {
      "entropy": 6.710722208023071,
      "epoch": 0.8976298537569339,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004445,
      "loss": 6.379,
      "mean_token_accuracy": 0.11063228026032448,
      "num_tokens": 1971052.0,
      "step": 890
    },
    {
      "entropy": 6.781415891647339,
      "epoch": 0.902672718103883,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000447,
      "loss": 6.3932,
      "mean_token_accuracy": 0.10573994740843773,
      "num_tokens": 1980917.0,
      "step": 895
    },
    {
      "entropy": 6.649546766281128,
      "epoch": 0.9077155824508321,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3419,
      "mean_token_accuracy": 0.11238374263048172,
      "num_tokens": 1991035.0,
      "step": 900
    },
    {
      "entropy": 6.629901695251465,
      "epoch": 0.9127584467977812,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3659,
      "mean_token_accuracy": 0.11216280534863472,
      "num_tokens": 2000587.0,
      "step": 905
    },
    {
      "entropy": 6.720604801177979,
      "epoch": 0.9178013111447302,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3292,
      "mean_token_accuracy": 0.11438163220882416,
      "num_tokens": 2011087.0,
      "step": 910
    },
    {
      "entropy": 6.651438570022583,
      "epoch": 0.9228441754916793,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3782,
      "mean_token_accuracy": 0.10697766989469529,
      "num_tokens": 2023472.0,
      "step": 915
    },
    {
      "entropy": 6.697429943084717,
      "epoch": 0.9278870398386283,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3639,
      "mean_token_accuracy": 0.11164266243577003,
      "num_tokens": 2033783.0,
      "step": 920
    },
    {
      "entropy": 6.520638275146484,
      "epoch": 0.9329299041855774,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000462,
      "loss": 6.3795,
      "mean_token_accuracy": 0.11479249447584153,
      "num_tokens": 2045882.0,
      "step": 925
    },
    {
      "entropy": 6.7621557235717775,
      "epoch": 0.9379727685325264,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004645,
      "loss": 6.3894,
      "mean_token_accuracy": 0.10624322891235352,
      "num_tokens": 2056549.0,
      "step": 930
    },
    {
      "entropy": 6.654461240768432,
      "epoch": 0.9430156328794755,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467,
      "loss": 6.3845,
      "mean_token_accuracy": 0.11624199599027633,
      "num_tokens": 2067540.0,
      "step": 935
    },
    {
      "entropy": 6.602401876449585,
      "epoch": 0.9480584972264247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004695,
      "loss": 6.2956,
      "mean_token_accuracy": 0.11275828555226326,
      "num_tokens": 2077386.0,
      "step": 940
    },
    {
      "entropy": 6.646829652786255,
      "epoch": 0.9531013615733737,
      "grad_norm": 1.125,
      "learning_rate": 0.000472,
      "loss": 6.2697,
      "mean_token_accuracy": 0.11356851384043694,
      "num_tokens": 2087777.0,
      "step": 945
    },
    {
      "entropy": 6.494188833236694,
      "epoch": 0.9581442259203228,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004745,
      "loss": 6.2604,
      "mean_token_accuracy": 0.11282371133565902,
      "num_tokens": 2099071.0,
      "step": 950
    },
    {
      "entropy": 6.53692774772644,
      "epoch": 0.9631870902672718,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000477,
      "loss": 6.3258,
      "mean_token_accuracy": 0.1121332198381424,
      "num_tokens": 2109168.0,
      "step": 955
    },
    {
      "entropy": 6.835064029693603,
      "epoch": 0.9682299546142209,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004795,
      "loss": 6.425,
      "mean_token_accuracy": 0.10913640931248665,
      "num_tokens": 2120696.0,
      "step": 960
    },
    {
      "entropy": 6.537368297576904,
      "epoch": 0.9732728189611699,
      "grad_norm": 1.171875,
      "learning_rate": 0.000482,
      "loss": 6.371,
      "mean_token_accuracy": 0.10968417152762414,
      "num_tokens": 2132427.0,
      "step": 965
    },
    {
      "entropy": 6.598920631408691,
      "epoch": 0.978315683308119,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845,
      "loss": 6.3618,
      "mean_token_accuracy": 0.11373768299818039,
      "num_tokens": 2143120.0,
      "step": 970
    },
    {
      "entropy": 6.740956020355225,
      "epoch": 0.983358547655068,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000487,
      "loss": 6.349,
      "mean_token_accuracy": 0.10921334773302079,
      "num_tokens": 2154154.0,
      "step": 975
    },
    {
      "entropy": 6.541984796524048,
      "epoch": 0.9884014120020171,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004895,
      "loss": 6.4499,
      "mean_token_accuracy": 0.10598269626498222,
      "num_tokens": 2167000.0,
      "step": 980
    },
    {
      "entropy": 6.663674449920654,
      "epoch": 0.9934442763489663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000492,
      "loss": 6.3823,
      "mean_token_accuracy": 0.11338229328393937,
      "num_tokens": 2178100.0,
      "step": 985
    },
    {
      "entropy": 6.6093058586120605,
      "epoch": 0.9984871406959153,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004945,
      "loss": 6.331,
      "mean_token_accuracy": 0.10886227712035179,
      "num_tokens": 2189063.0,
      "step": 990
    },
    {
      "entropy": 6.706958293914795,
      "epoch": 1.0030257186081695,
      "grad_norm": 1.046875,
      "learning_rate": 0.000497,
      "loss": 6.2723,
      "mean_token_accuracy": 0.11505810419718425,
      "num_tokens": 2198302.0,
      "step": 995
    },
    {
      "entropy": 6.57634973526001,
      "epoch": 1.0080685829551186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995,
      "loss": 6.2311,
      "mean_token_accuracy": 0.1161844253540039,
      "num_tokens": 2209648.0,
      "step": 1000
    },
    {
      "epoch": 1.0080685829551186,
      "eval_entropy": 6.5141722082181746,
      "eval_loss": 6.3961334228515625,
      "eval_mean_token_accuracy": 0.11549569743166443,
      "eval_num_tokens": 2209648.0,
      "eval_runtime": 2.054,
      "eval_samples_per_second": 1527.74,
      "eval_steps_per_second": 191.333,
      "step": 1000
    },
    {
      "entropy": 6.522863817214966,
      "epoch": 1.0131114473020675,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999997762223182,
      "loss": 6.1663,
      "mean_token_accuracy": 0.12039870694279671,
      "num_tokens": 2220806.0,
      "step": 1005
    },
    {
      "entropy": 6.48839225769043,
      "epoch": 1.0181543116490166,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999988671262489,
      "loss": 6.1552,
      "mean_token_accuracy": 0.11676756143569947,
      "num_tokens": 2231806.0,
      "step": 1010
    },
    {
      "entropy": 6.5846107006073,
      "epoch": 1.0231971759959657,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999972587285105,
      "loss": 6.2079,
      "mean_token_accuracy": 0.12164777219295501,
      "num_tokens": 2243835.0,
      "step": 1015
    },
    {
      "entropy": 6.374717426300049,
      "epoch": 1.0282400403429148,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999949510341015,
      "loss": 6.1189,
      "mean_token_accuracy": 0.11811361014842987,
      "num_tokens": 2254722.0,
      "step": 1020
    },
    {
      "entropy": 6.500878429412841,
      "epoch": 1.033282904689864,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999919440501947,
      "loss": 6.077,
      "mean_token_accuracy": 0.12784491255879402,
      "num_tokens": 2264806.0,
      "step": 1025
    },
    {
      "entropy": 6.599410104751587,
      "epoch": 1.0383257690368128,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999882377861355,
      "loss": 6.259,
      "mean_token_accuracy": 0.11469498798251151,
      "num_tokens": 2277037.0,
      "step": 1030
    },
    {
      "entropy": 6.523988580703735,
      "epoch": 1.043368633383762,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999838322534433,
      "loss": 6.2056,
      "mean_token_accuracy": 0.10915817320346832,
      "num_tokens": 2289829.0,
      "step": 1035
    },
    {
      "entropy": 6.515505409240722,
      "epoch": 1.048411497730711,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999787274658107,
      "loss": 6.1,
      "mean_token_accuracy": 0.11675480157136917,
      "num_tokens": 2300340.0,
      "step": 1040
    },
    {
      "entropy": 6.47110800743103,
      "epoch": 1.0534543620776602,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999729234391032,
      "loss": 6.1279,
      "mean_token_accuracy": 0.1232002280652523,
      "num_tokens": 2310009.0,
      "step": 1045
    },
    {
      "entropy": 6.43739595413208,
      "epoch": 1.058497226424609,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049996642019136,
      "loss": 6.1581,
      "mean_token_accuracy": 0.11546300202608109,
      "num_tokens": 2319290.0,
      "step": 1050
    },
    {
      "entropy": 6.552343559265137,
      "epoch": 1.0635400907715582,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999592177427933,
      "loss": 6.1181,
      "mean_token_accuracy": 0.12222910076379775,
      "num_tokens": 2329170.0,
      "step": 1055
    },
    {
      "entropy": 6.407629442214966,
      "epoch": 1.0685829551185073,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999513161157885,
      "loss": 6.119,
      "mean_token_accuracy": 0.11926354318857194,
      "num_tokens": 2339617.0,
      "step": 1060
    },
    {
      "entropy": 6.497067499160766,
      "epoch": 1.0736258194654564,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999427153349041,
      "loss": 6.1083,
      "mean_token_accuracy": 0.12506095916032792,
      "num_tokens": 2349548.0,
      "step": 1065
    },
    {
      "entropy": 6.351389837265015,
      "epoch": 1.0786686838124055,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999334154268714,
      "loss": 6.1019,
      "mean_token_accuracy": 0.11952552795410157,
      "num_tokens": 2359116.0,
      "step": 1070
    },
    {
      "entropy": 6.541378593444824,
      "epoch": 1.0837115481593544,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999234164205949,
      "loss": 6.1649,
      "mean_token_accuracy": 0.12546638995409012,
      "num_tokens": 2369412.0,
      "step": 1075
    },
    {
      "entropy": 6.466941499710083,
      "epoch": 1.0887544125063036,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999127183471517,
      "loss": 6.222,
      "mean_token_accuracy": 0.11774109154939652,
      "num_tokens": 2381276.0,
      "step": 1080
    },
    {
      "entropy": 6.495827722549438,
      "epoch": 1.0937972768532527,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999013212397915,
      "loss": 6.1256,
      "mean_token_accuracy": 0.12179253101348878,
      "num_tokens": 2392135.0,
      "step": 1085
    },
    {
      "entropy": 6.338582515716553,
      "epoch": 1.0988401412002018,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998892251339368,
      "loss": 6.0403,
      "mean_token_accuracy": 0.13191248327493668,
      "num_tokens": 2402957.0,
      "step": 1090
    },
    {
      "entropy": 6.49658522605896,
      "epoch": 1.1038830055471507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998764300671827,
      "loss": 6.1791,
      "mean_token_accuracy": 0.12375712767243385,
      "num_tokens": 2413386.0,
      "step": 1095
    },
    {
      "entropy": 6.396690225601196,
      "epoch": 1.1089258698940998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998629360792965,
      "loss": 6.1582,
      "mean_token_accuracy": 0.11660370007157325,
      "num_tokens": 2424441.0,
      "step": 1100
    },
    {
      "entropy": 6.422626638412476,
      "epoch": 1.113968734241049,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998487432122179,
      "loss": 6.1819,
      "mean_token_accuracy": 0.11399172693490982,
      "num_tokens": 2436144.0,
      "step": 1105
    },
    {
      "entropy": 6.488900423049927,
      "epoch": 1.119011598587998,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998338515100585,
      "loss": 6.0988,
      "mean_token_accuracy": 0.1264061599969864,
      "num_tokens": 2447393.0,
      "step": 1110
    },
    {
      "entropy": 6.347085332870483,
      "epoch": 1.1240544629349472,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998182610191022,
      "loss": 6.0551,
      "mean_token_accuracy": 0.12492778077721596,
      "num_tokens": 2458632.0,
      "step": 1115
    },
    {
      "entropy": 6.4905647277832035,
      "epoch": 1.129097327281896,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998019717878047,
      "loss": 6.135,
      "mean_token_accuracy": 0.11796133294701576,
      "num_tokens": 2469331.0,
      "step": 1120
    },
    {
      "entropy": 6.457821035385132,
      "epoch": 1.1341401916288452,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499784983866793,
      "loss": 6.1834,
      "mean_token_accuracy": 0.11562436521053314,
      "num_tokens": 2479501.0,
      "step": 1125
    },
    {
      "entropy": 6.3564197540283205,
      "epoch": 1.1391830559757943,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997672973088661,
      "loss": 6.052,
      "mean_token_accuracy": 0.126077089458704,
      "num_tokens": 2490087.0,
      "step": 1130
    },
    {
      "entropy": 6.441509485244751,
      "epoch": 1.1442259203227434,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997489121689944,
      "loss": 6.1185,
      "mean_token_accuracy": 0.1169286049902439,
      "num_tokens": 2501130.0,
      "step": 1135
    },
    {
      "entropy": 6.476033449172974,
      "epoch": 1.1492687846696923,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997298285043189,
      "loss": 6.2252,
      "mean_token_accuracy": 0.11470448672771454,
      "num_tokens": 2513999.0,
      "step": 1140
    },
    {
      "entropy": 6.4647002696990965,
      "epoch": 1.1543116490166414,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997100463741527,
      "loss": 6.1458,
      "mean_token_accuracy": 0.11966953650116921,
      "num_tokens": 2525579.0,
      "step": 1145
    },
    {
      "entropy": 6.375444412231445,
      "epoch": 1.1593545133635905,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996895658399785,
      "loss": 6.1188,
      "mean_token_accuracy": 0.12094511762261391,
      "num_tokens": 2536091.0,
      "step": 1150
    },
    {
      "entropy": 6.43728551864624,
      "epoch": 1.1643973777105396,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996683869654507,
      "loss": 6.1577,
      "mean_token_accuracy": 0.12777142524719237,
      "num_tokens": 2546073.0,
      "step": 1155
    },
    {
      "entropy": 6.448345422744751,
      "epoch": 1.1694402420574885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996465098163934,
      "loss": 6.108,
      "mean_token_accuracy": 0.11881096065044403,
      "num_tokens": 2556626.0,
      "step": 1160
    },
    {
      "entropy": 6.405810403823852,
      "epoch": 1.1744831064044376,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996239344608015,
      "loss": 6.1156,
      "mean_token_accuracy": 0.12832853496074675,
      "num_tokens": 2567304.0,
      "step": 1165
    },
    {
      "entropy": 6.377554607391358,
      "epoch": 1.1795259707513868,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996006609688398,
      "loss": 6.001,
      "mean_token_accuracy": 0.1183304913341999,
      "num_tokens": 2577242.0,
      "step": 1170
    },
    {
      "entropy": 6.341175317764282,
      "epoch": 1.1845688350983359,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995766894128426,
      "loss": 6.0978,
      "mean_token_accuracy": 0.11528865247964859,
      "num_tokens": 2588330.0,
      "step": 1175
    },
    {
      "entropy": 6.347274208068848,
      "epoch": 1.189611699445285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995520198673143,
      "loss": 6.1208,
      "mean_token_accuracy": 0.1222249336540699,
      "num_tokens": 2598471.0,
      "step": 1180
    },
    {
      "entropy": 6.476546859741211,
      "epoch": 1.194654563792234,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995266524089283,
      "loss": 6.1917,
      "mean_token_accuracy": 0.11832692325115204,
      "num_tokens": 2610238.0,
      "step": 1185
    },
    {
      "entropy": 6.360067367553711,
      "epoch": 1.199697428139183,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995005871165271,
      "loss": 6.0669,
      "mean_token_accuracy": 0.11930153965950012,
      "num_tokens": 2619870.0,
      "step": 1190
    },
    {
      "entropy": 6.458434772491455,
      "epoch": 1.2047402924861321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994738240711225,
      "loss": 6.1989,
      "mean_token_accuracy": 0.11891994774341583,
      "num_tokens": 2631541.0,
      "step": 1195
    },
    {
      "entropy": 6.346074151992798,
      "epoch": 1.2097831568330812,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994463633558945,
      "loss": 6.0434,
      "mean_token_accuracy": 0.12345149591565133,
      "num_tokens": 2641896.0,
      "step": 1200
    },
    {
      "entropy": 6.380703020095825,
      "epoch": 1.2148260211800301,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994182050561917,
      "loss": 6.1134,
      "mean_token_accuracy": 0.11929800137877464,
      "num_tokens": 2654208.0,
      "step": 1205
    },
    {
      "entropy": 6.355344963073731,
      "epoch": 1.2198688855269793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993893492595306,
      "loss": 6.0965,
      "mean_token_accuracy": 0.12326192557811737,
      "num_tokens": 2665115.0,
      "step": 1210
    },
    {
      "entropy": 6.437494277954102,
      "epoch": 1.2249117498739284,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993597960555959,
      "loss": 6.1801,
      "mean_token_accuracy": 0.11685346364974976,
      "num_tokens": 2676093.0,
      "step": 1215
    },
    {
      "entropy": 6.38605375289917,
      "epoch": 1.2299546142208775,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993295455362395,
      "loss": 6.1218,
      "mean_token_accuracy": 0.12748127356171607,
      "num_tokens": 2687866.0,
      "step": 1220
    },
    {
      "entropy": 6.403005504608155,
      "epoch": 1.2349974785678266,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004992985977954808,
      "loss": 6.1404,
      "mean_token_accuracy": 0.12012984082102776,
      "num_tokens": 2699814.0,
      "step": 1225
    },
    {
      "entropy": 6.36786527633667,
      "epoch": 1.2400403429147757,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499266952929506,
      "loss": 6.1101,
      "mean_token_accuracy": 0.13033016249537469,
      "num_tokens": 2710384.0,
      "step": 1230
    },
    {
      "entropy": 6.3866376876831055,
      "epoch": 1.2450832072617246,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992346110366682,
      "loss": 6.0859,
      "mean_token_accuracy": 0.12090532258152961,
      "num_tokens": 2721442.0,
      "step": 1235
    },
    {
      "entropy": 6.362185049057007,
      "epoch": 1.2501260716086737,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992015722174865,
      "loss": 6.1027,
      "mean_token_accuracy": 0.11984567195177079,
      "num_tokens": 2733605.0,
      "step": 1240
    },
    {
      "entropy": 6.381150722503662,
      "epoch": 1.2551689359556228,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991678365746466,
      "loss": 6.1376,
      "mean_token_accuracy": 0.12807665169239044,
      "num_tokens": 2744853.0,
      "step": 1245
    },
    {
      "entropy": 6.278868913650513,
      "epoch": 1.2602118003025717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991334042129996,
      "loss": 6.0394,
      "mean_token_accuracy": 0.12930434346199035,
      "num_tokens": 2754477.0,
      "step": 1250
    },
    {
      "entropy": 6.3861668586730955,
      "epoch": 1.2652546646495209,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990982752395618,
      "loss": 6.062,
      "mean_token_accuracy": 0.12282699942588807,
      "num_tokens": 2765895.0,
      "step": 1255
    },
    {
      "entropy": 6.3185412883758545,
      "epoch": 1.27029752899647,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990624497635152,
      "loss": 6.0955,
      "mean_token_accuracy": 0.12333361357450486,
      "num_tokens": 2776897.0,
      "step": 1260
    },
    {
      "entropy": 6.4460615634918215,
      "epoch": 1.275340393343419,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499025927896206,
      "loss": 6.1191,
      "mean_token_accuracy": 0.12341732308268547,
      "num_tokens": 2788666.0,
      "step": 1265
    },
    {
      "entropy": 6.285901021957398,
      "epoch": 1.2803832576903682,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989887097511451,
      "loss": 6.0122,
      "mean_token_accuracy": 0.12529174312949182,
      "num_tokens": 2799167.0,
      "step": 1270
    },
    {
      "entropy": 6.281661653518677,
      "epoch": 1.2854261220373173,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989507954440074,
      "loss": 6.0129,
      "mean_token_accuracy": 0.11882785484194755,
      "num_tokens": 2809774.0,
      "step": 1275
    },
    {
      "entropy": 6.318780517578125,
      "epoch": 1.2904689863842662,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989121850926313,
      "loss": 6.1337,
      "mean_token_accuracy": 0.11986329406499863,
      "num_tokens": 2820515.0,
      "step": 1280
    },
    {
      "entropy": 6.336706733703613,
      "epoch": 1.2955118507312153,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988728788170188,
      "loss": 6.0786,
      "mean_token_accuracy": 0.1216197930276394,
      "num_tokens": 2832163.0,
      "step": 1285
    },
    {
      "entropy": 6.311708307266235,
      "epoch": 1.3005547150781644,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988328767393346,
      "loss": 6.0644,
      "mean_token_accuracy": 0.1252337910234928,
      "num_tokens": 2843717.0,
      "step": 1290
    },
    {
      "entropy": 6.23866548538208,
      "epoch": 1.3055975794251133,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987921789839061,
      "loss": 6.0187,
      "mean_token_accuracy": 0.12154641002416611,
      "num_tokens": 2854489.0,
      "step": 1295
    },
    {
      "entropy": 6.304507398605347,
      "epoch": 1.3106404437720625,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987507856772229,
      "loss": 6.0674,
      "mean_token_accuracy": 0.12463115528225899,
      "num_tokens": 2865317.0,
      "step": 1300
    },
    {
      "entropy": 6.30941219329834,
      "epoch": 1.3156833081190116,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987086969479364,
      "loss": 5.9961,
      "mean_token_accuracy": 0.12587646394968033,
      "num_tokens": 2877466.0,
      "step": 1305
    },
    {
      "entropy": 6.2870680809021,
      "epoch": 1.3207261724659607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004986659129268592,
      "loss": 6.018,
      "mean_token_accuracy": 0.1252691850066185,
      "num_tokens": 2887698.0,
      "step": 1310
    },
    {
      "entropy": 6.190779304504394,
      "epoch": 1.3257690368129098,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004986224337469652,
      "loss": 5.9728,
      "mean_token_accuracy": 0.13109618574380874,
      "num_tokens": 2899339.0,
      "step": 1315
    },
    {
      "entropy": 6.250626802444458,
      "epoch": 1.330811901159859,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985782595433886,
      "loss": 6.0246,
      "mean_token_accuracy": 0.12930653765797614,
      "num_tokens": 2908340.0,
      "step": 1320
    },
    {
      "entropy": 6.36276421546936,
      "epoch": 1.3358547655068078,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985333904534237,
      "loss": 6.1153,
      "mean_token_accuracy": 0.12473672702908516,
      "num_tokens": 2920432.0,
      "step": 1325
    },
    {
      "entropy": 6.318589591979981,
      "epoch": 1.340897629853757,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498487826616525,
      "loss": 6.0603,
      "mean_token_accuracy": 0.12252447828650474,
      "num_tokens": 2930385.0,
      "step": 1330
    },
    {
      "entropy": 6.31836199760437,
      "epoch": 1.345940494200706,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004984415681743056,
      "loss": 6.0133,
      "mean_token_accuracy": 0.1288218967616558,
      "num_tokens": 2940760.0,
      "step": 1335
    },
    {
      "entropy": 6.281600284576416,
      "epoch": 1.350983358547655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004983946152705382,
      "loss": 6.0344,
      "mean_token_accuracy": 0.13234889060258864,
      "num_tokens": 2952441.0,
      "step": 1340
    },
    {
      "entropy": 6.320213270187378,
      "epoch": 1.356026222894604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983469680511531,
      "loss": 6.0584,
      "mean_token_accuracy": 0.125762165337801,
      "num_tokens": 2962766.0,
      "step": 1345
    },
    {
      "entropy": 6.2049356460571286,
      "epoch": 1.3610690872415532,
      "grad_norm": 1.0,
      "learning_rate": 0.0004982986266642393,
      "loss": 5.9448,
      "mean_token_accuracy": 0.13519371226429938,
      "num_tokens": 2974239.0,
      "step": 1350
    },
    {
      "entropy": 6.304392290115357,
      "epoch": 1.3661119515885023,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982495912600426,
      "loss": 5.9836,
      "mean_token_accuracy": 0.12600109577178956,
      "num_tokens": 2984497.0,
      "step": 1355
    },
    {
      "entropy": 6.244618892669678,
      "epoch": 1.3711548159354514,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981998619909664,
      "loss": 5.9772,
      "mean_token_accuracy": 0.12882400155067444,
      "num_tokens": 2995367.0,
      "step": 1360
    },
    {
      "entropy": 6.267851448059082,
      "epoch": 1.3761976802824005,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004981494390115704,
      "loss": 6.0067,
      "mean_token_accuracy": 0.1295331373810768,
      "num_tokens": 3007687.0,
      "step": 1365
    },
    {
      "entropy": 6.281758832931518,
      "epoch": 1.3812405446293494,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980983224785704,
      "loss": 6.0025,
      "mean_token_accuracy": 0.1259416989982128,
      "num_tokens": 3017505.0,
      "step": 1370
    },
    {
      "entropy": 6.1590625762939455,
      "epoch": 1.3862834089762985,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004980465125508376,
      "loss": 5.9409,
      "mean_token_accuracy": 0.12704395055770873,
      "num_tokens": 3028271.0,
      "step": 1375
    },
    {
      "entropy": 6.286492538452149,
      "epoch": 1.3913262733232477,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004979940093893986,
      "loss": 5.9927,
      "mean_token_accuracy": 0.12936830893158913,
      "num_tokens": 3040440.0,
      "step": 1380
    },
    {
      "entropy": 6.132359313964844,
      "epoch": 1.3963691376701965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979408131574345,
      "loss": 5.9758,
      "mean_token_accuracy": 0.12391475141048432,
      "num_tokens": 3050999.0,
      "step": 1385
    },
    {
      "entropy": 6.337554454803467,
      "epoch": 1.4014120020171457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978869240202802,
      "loss": 6.027,
      "mean_token_accuracy": 0.12895868942141533,
      "num_tokens": 3063374.0,
      "step": 1390
    },
    {
      "entropy": 6.267091894149781,
      "epoch": 1.4064548663640948,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978323421454246,
      "loss": 6.0622,
      "mean_token_accuracy": 0.12893708497285844,
      "num_tokens": 3075464.0,
      "step": 1395
    },
    {
      "entropy": 6.185108709335327,
      "epoch": 1.411497730711044,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977770677025092,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1268825463950634,
      "num_tokens": 3086398.0,
      "step": 1400
    },
    {
      "entropy": 6.22553391456604,
      "epoch": 1.416540595057993,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977211008633284,
      "loss": 6.0521,
      "mean_token_accuracy": 0.11881259009242058,
      "num_tokens": 3097777.0,
      "step": 1405
    },
    {
      "entropy": 6.272607231140137,
      "epoch": 1.4215834594049421,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004976644418018285,
      "loss": 6.0208,
      "mean_token_accuracy": 0.13160417675971986,
      "num_tokens": 3108711.0,
      "step": 1410
    },
    {
      "entropy": 6.2519410133361815,
      "epoch": 1.426626323751891,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497607090694107,
      "loss": 6.0407,
      "mean_token_accuracy": 0.12776092514395715,
      "num_tokens": 3119589.0,
      "step": 1415
    },
    {
      "entropy": 6.209818506240845,
      "epoch": 1.4316691880988401,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004975490477184126,
      "loss": 5.973,
      "mean_token_accuracy": 0.13335097581148148,
      "num_tokens": 3131744.0,
      "step": 1420
    },
    {
      "entropy": 6.2790721416473385,
      "epoch": 1.4367120524457893,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497490313055144,
      "loss": 6.0089,
      "mean_token_accuracy": 0.12579527944326402,
      "num_tokens": 3143526.0,
      "step": 1425
    },
    {
      "entropy": 6.256898641586304,
      "epoch": 1.4417549167927382,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004974308868868501,
      "loss": 5.9551,
      "mean_token_accuracy": 0.12504913732409478,
      "num_tokens": 3154561.0,
      "step": 1430
    },
    {
      "entropy": 6.238351345062256,
      "epoch": 1.4467977811396873,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973707693982287,
      "loss": 6.1168,
      "mean_token_accuracy": 0.12133650034666062,
      "num_tokens": 3166052.0,
      "step": 1435
    },
    {
      "entropy": 6.357213068008423,
      "epoch": 1.4518406454866364,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004973099607761264,
      "loss": 6.0734,
      "mean_token_accuracy": 0.12326012775301934,
      "num_tokens": 3179335.0,
      "step": 1440
    },
    {
      "entropy": 6.196079301834106,
      "epoch": 1.4568835098335855,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004972484612095378,
      "loss": 5.9556,
      "mean_token_accuracy": 0.12705831453204156,
      "num_tokens": 3192129.0,
      "step": 1445
    },
    {
      "entropy": 6.309902286529541,
      "epoch": 1.4619263741805346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497186270889605,
      "loss": 6.0646,
      "mean_token_accuracy": 0.12445836812257767,
      "num_tokens": 3202792.0,
      "step": 1450
    },
    {
      "entropy": 6.250305843353272,
      "epoch": 1.4669692385274837,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004971233900096169,
      "loss": 6.0093,
      "mean_token_accuracy": 0.12253658324480057,
      "num_tokens": 3213707.0,
      "step": 1455
    },
    {
      "entropy": 6.325912141799927,
      "epoch": 1.4720121028744326,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004970598187650085,
      "loss": 5.9955,
      "mean_token_accuracy": 0.13006001710891724,
      "num_tokens": 3224531.0,
      "step": 1460
    },
    {
      "entropy": 6.210040140151977,
      "epoch": 1.4770549672213817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496995557353361,
      "loss": 5.9749,
      "mean_token_accuracy": 0.1309881828725338,
      "num_tokens": 3234656.0,
      "step": 1465
    },
    {
      "entropy": 6.212216138839722,
      "epoch": 1.4820978315683309,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004969306059744004,
      "loss": 6.0054,
      "mean_token_accuracy": 0.13170277923345566,
      "num_tokens": 3246745.0,
      "step": 1470
    },
    {
      "entropy": 6.162950563430786,
      "epoch": 1.4871406959152798,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004968649648299968,
      "loss": 5.9935,
      "mean_token_accuracy": 0.12871679663658142,
      "num_tokens": 3257277.0,
      "step": 1475
    },
    {
      "entropy": 6.304571056365967,
      "epoch": 1.4921835602622289,
      "grad_norm": 1.125,
      "learning_rate": 0.0004967986341241646,
      "loss": 5.9564,
      "mean_token_accuracy": 0.13108997642993928,
      "num_tokens": 3267704.0,
      "step": 1480
    },
    {
      "entropy": 6.211066675186157,
      "epoch": 1.497226424609178,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004967316140630611,
      "loss": 6.0263,
      "mean_token_accuracy": 0.1284073054790497,
      "num_tokens": 3279772.0,
      "step": 1485
    },
    {
      "entropy": 6.19977822303772,
      "epoch": 1.502269288956127,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004966639048549862,
      "loss": 5.9248,
      "mean_token_accuracy": 0.13373203277587892,
      "num_tokens": 3290536.0,
      "step": 1490
    },
    {
      "entropy": 6.163003826141358,
      "epoch": 1.5073121533030762,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965955067103816,
      "loss": 6.0015,
      "mean_token_accuracy": 0.13202472701668738,
      "num_tokens": 3302051.0,
      "step": 1495
    },
    {
      "entropy": 6.216978979110718,
      "epoch": 1.5123550176500253,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004965264198418305,
      "loss": 5.9805,
      "mean_token_accuracy": 0.11812185272574424,
      "num_tokens": 3314324.0,
      "step": 1500
    },
    {
      "epoch": 1.5123550176500253,
      "eval_entropy": 5.988947841047331,
      "eval_loss": 6.126096725463867,
      "eval_mean_token_accuracy": 0.12609033783276877,
      "eval_num_tokens": 3314324.0,
      "eval_runtime": 2.0434,
      "eval_samples_per_second": 1535.706,
      "eval_steps_per_second": 192.33,
      "step": 1500
    },
    {
      "entropy": 6.162470149993896,
      "epoch": 1.5173978819969742,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964566444640564,
      "loss": 5.9764,
      "mean_token_accuracy": 0.12956623807549478,
      "num_tokens": 3326246.0,
      "step": 1505
    },
    {
      "entropy": 6.1691357612609865,
      "epoch": 1.5224407463439233,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004963861807939227,
      "loss": 5.9697,
      "mean_token_accuracy": 0.13050643503665924,
      "num_tokens": 3336308.0,
      "step": 1510
    },
    {
      "entropy": 6.113428115844727,
      "epoch": 1.5274836106908725,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004963150290504321,
      "loss": 5.9096,
      "mean_token_accuracy": 0.1319293461740017,
      "num_tokens": 3347596.0,
      "step": 1515
    },
    {
      "entropy": 6.130649662017822,
      "epoch": 1.5325264750378214,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962431894547259,
      "loss": 5.8863,
      "mean_token_accuracy": 0.13102136179804802,
      "num_tokens": 3358370.0,
      "step": 1520
    },
    {
      "entropy": 6.282006025314331,
      "epoch": 1.5375693393847705,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004961706622300834,
      "loss": 5.9189,
      "mean_token_accuracy": 0.12766351029276848,
      "num_tokens": 3368630.0,
      "step": 1525
    },
    {
      "entropy": 6.164436769485474,
      "epoch": 1.5426122037317196,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004960974476019206,
      "loss": 5.9346,
      "mean_token_accuracy": 0.13147821873426438,
      "num_tokens": 3379952.0,
      "step": 1530
    },
    {
      "entropy": 6.137597465515137,
      "epoch": 1.5476550680786687,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960235457977905,
      "loss": 5.8684,
      "mean_token_accuracy": 0.13159159570932388,
      "num_tokens": 3390097.0,
      "step": 1535
    },
    {
      "entropy": 6.087500667572021,
      "epoch": 1.5526979324256178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004959489570473814,
      "loss": 5.8595,
      "mean_token_accuracy": 0.1370345577597618,
      "num_tokens": 3399778.0,
      "step": 1540
    },
    {
      "entropy": 6.216370344161987,
      "epoch": 1.557740796772567,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000495873681582517,
      "loss": 6.0103,
      "mean_token_accuracy": 0.12765116170048713,
      "num_tokens": 3410213.0,
      "step": 1545
    },
    {
      "entropy": 6.172270250320435,
      "epoch": 1.5627836611195158,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957977196371552,
      "loss": 6.0277,
      "mean_token_accuracy": 0.13068305626511573,
      "num_tokens": 3422071.0,
      "step": 1550
    },
    {
      "entropy": 6.241772222518921,
      "epoch": 1.567826525466465,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004957210714473873,
      "loss": 5.9931,
      "mean_token_accuracy": 0.13293333649635314,
      "num_tokens": 3432102.0,
      "step": 1555
    },
    {
      "entropy": 6.157260370254517,
      "epoch": 1.572869389813414,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004956437372514378,
      "loss": 5.962,
      "mean_token_accuracy": 0.12790244817733765,
      "num_tokens": 3442821.0,
      "step": 1560
    },
    {
      "entropy": 6.1026246547698975,
      "epoch": 1.577912254160363,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004955657172896631,
      "loss": 5.9231,
      "mean_token_accuracy": 0.1324974425137043,
      "num_tokens": 3454378.0,
      "step": 1565
    },
    {
      "entropy": 6.163491344451904,
      "epoch": 1.582955118507312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954870118045511,
      "loss": 5.9412,
      "mean_token_accuracy": 0.13418091386556624,
      "num_tokens": 3463933.0,
      "step": 1570
    },
    {
      "entropy": 6.149846458435059,
      "epoch": 1.5879979828542612,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004954076210407203,
      "loss": 5.8842,
      "mean_token_accuracy": 0.12891194969415665,
      "num_tokens": 3473993.0,
      "step": 1575
    },
    {
      "entropy": 6.184996604919434,
      "epoch": 1.5930408472012103,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004953275452449188,
      "loss": 6.0255,
      "mean_token_accuracy": 0.12238797843456269,
      "num_tokens": 3484882.0,
      "step": 1580
    },
    {
      "entropy": 6.172972679138184,
      "epoch": 1.5980837115481594,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004952467846660243,
      "loss": 5.9429,
      "mean_token_accuracy": 0.13000317960977553,
      "num_tokens": 3496503.0,
      "step": 1585
    },
    {
      "entropy": 6.168394422531128,
      "epoch": 1.6031265758951085,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004951653395550423,
      "loss": 6.0277,
      "mean_token_accuracy": 0.13080177530646325,
      "num_tokens": 3508696.0,
      "step": 1590
    },
    {
      "entropy": 6.255574321746826,
      "epoch": 1.6081694402420574,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004950832101651062,
      "loss": 5.9399,
      "mean_token_accuracy": 0.13680166453123094,
      "num_tokens": 3519542.0,
      "step": 1595
    },
    {
      "entropy": 6.162024974822998,
      "epoch": 1.6132123045890066,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950003967514762,
      "loss": 5.9039,
      "mean_token_accuracy": 0.13319480940699577,
      "num_tokens": 3531576.0,
      "step": 1600
    },
    {
      "entropy": 6.168517971038819,
      "epoch": 1.6182551689359557,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000494916899571538,
      "loss": 5.8714,
      "mean_token_accuracy": 0.13608561530709268,
      "num_tokens": 3542816.0,
      "step": 1605
    },
    {
      "entropy": 6.182320785522461,
      "epoch": 1.6232980332829046,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948327188848029,
      "loss": 6.0066,
      "mean_token_accuracy": 0.12684836611151695,
      "num_tokens": 3554100.0,
      "step": 1610
    },
    {
      "entropy": 6.230264568328858,
      "epoch": 1.6283408976298537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004947478549529065,
      "loss": 5.8562,
      "mean_token_accuracy": 0.12772923931479455,
      "num_tokens": 3565048.0,
      "step": 1615
    },
    {
      "entropy": 6.198247385025025,
      "epoch": 1.6333837619768028,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004946623080396077,
      "loss": 5.9378,
      "mean_token_accuracy": 0.1336819685995579,
      "num_tokens": 3575587.0,
      "step": 1620
    },
    {
      "entropy": 6.148627471923828,
      "epoch": 1.638426626323752,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945760784107884,
      "loss": 6.0094,
      "mean_token_accuracy": 0.12861310243606566,
      "num_tokens": 3587059.0,
      "step": 1625
    },
    {
      "entropy": 6.1220705032348635,
      "epoch": 1.643469490670701,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004944891663344522,
      "loss": 5.8406,
      "mean_token_accuracy": 0.13324351757764816,
      "num_tokens": 3598492.0,
      "step": 1630
    },
    {
      "entropy": 6.100050592422486,
      "epoch": 1.6485123550176501,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004944015720807239,
      "loss": 5.9125,
      "mean_token_accuracy": 0.1337079659104347,
      "num_tokens": 3609513.0,
      "step": 1635
    },
    {
      "entropy": 6.179864168167114,
      "epoch": 1.653555219364599,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004943132959218484,
      "loss": 5.9377,
      "mean_token_accuracy": 0.13239555582404136,
      "num_tokens": 3621327.0,
      "step": 1640
    },
    {
      "entropy": 6.07295184135437,
      "epoch": 1.6585980837115482,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049422433813219,
      "loss": 5.8455,
      "mean_token_accuracy": 0.1334436446428299,
      "num_tokens": 3632136.0,
      "step": 1645
    },
    {
      "entropy": 6.140309047698975,
      "epoch": 1.6636409480584973,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941346989882318,
      "loss": 5.8253,
      "mean_token_accuracy": 0.13711320981383324,
      "num_tokens": 3642271.0,
      "step": 1650
    },
    {
      "entropy": 5.994101095199585,
      "epoch": 1.6686838124054462,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004940443787685739,
      "loss": 5.9454,
      "mean_token_accuracy": 0.12989369183778762,
      "num_tokens": 3652355.0,
      "step": 1655
    },
    {
      "entropy": 6.283844423294068,
      "epoch": 1.6737266767523953,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004939533777539341,
      "loss": 5.9346,
      "mean_token_accuracy": 0.12859864830970763,
      "num_tokens": 3663714.0,
      "step": 1660
    },
    {
      "entropy": 6.0334240913391115,
      "epoch": 1.6787695410993444,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004938616962271453,
      "loss": 5.8808,
      "mean_token_accuracy": 0.12998360991477967,
      "num_tokens": 3674675.0,
      "step": 1665
    },
    {
      "entropy": 6.12380805015564,
      "epoch": 1.6838124054462935,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004937693344731561,
      "loss": 5.954,
      "mean_token_accuracy": 0.1350781202316284,
      "num_tokens": 3686403.0,
      "step": 1670
    },
    {
      "entropy": 6.118779897689819,
      "epoch": 1.6888552697932426,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936762927790286,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1421683058142662,
      "num_tokens": 3697259.0,
      "step": 1675
    },
    {
      "entropy": 6.124578905105591,
      "epoch": 1.6938981341401917,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935825714339389,
      "loss": 5.9011,
      "mean_token_accuracy": 0.1348792150616646,
      "num_tokens": 3708536.0,
      "step": 1680
    },
    {
      "entropy": 6.220402526855469,
      "epoch": 1.6989409984871406,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934881707291748,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1309715911746025,
      "num_tokens": 3719181.0,
      "step": 1685
    },
    {
      "entropy": 6.012847232818603,
      "epoch": 1.7039838628340898,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933930909581362,
      "loss": 5.7549,
      "mean_token_accuracy": 0.13449717462062835,
      "num_tokens": 3729325.0,
      "step": 1690
    },
    {
      "entropy": 6.101225805282593,
      "epoch": 1.7090267271810389,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004932973324163328,
      "loss": 5.9008,
      "mean_token_accuracy": 0.13119059279561043,
      "num_tokens": 3740462.0,
      "step": 1695
    },
    {
      "entropy": 6.210686779022216,
      "epoch": 1.7140695915279878,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004932008954013846,
      "loss": 5.9776,
      "mean_token_accuracy": 0.13249426260590552,
      "num_tokens": 3752523.0,
      "step": 1700
    },
    {
      "entropy": 6.021712589263916,
      "epoch": 1.7191124558749369,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004931037802130201,
      "loss": 5.8401,
      "mean_token_accuracy": 0.13165827840566635,
      "num_tokens": 3764120.0,
      "step": 1705
    },
    {
      "entropy": 6.134450483322143,
      "epoch": 1.724155320221886,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004930059871530755,
      "loss": 5.9728,
      "mean_token_accuracy": 0.13011515960097314,
      "num_tokens": 3776733.0,
      "step": 1710
    },
    {
      "entropy": 6.178735971450806,
      "epoch": 1.7291981845688351,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004929075165254938,
      "loss": 5.9398,
      "mean_token_accuracy": 0.126145376265049,
      "num_tokens": 3787799.0,
      "step": 1715
    },
    {
      "entropy": 6.030863428115845,
      "epoch": 1.7342410489157842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004928083686363241,
      "loss": 5.863,
      "mean_token_accuracy": 0.13384964913129807,
      "num_tokens": 3797995.0,
      "step": 1720
    },
    {
      "entropy": 6.203894710540771,
      "epoch": 1.7392839132627333,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004927085437937204,
      "loss": 5.9367,
      "mean_token_accuracy": 0.13228645250201226,
      "num_tokens": 3809580.0,
      "step": 1725
    },
    {
      "entropy": 6.146491956710816,
      "epoch": 1.7443267776096822,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004926080423079404,
      "loss": 5.9601,
      "mean_token_accuracy": 0.13396261408925056,
      "num_tokens": 3820150.0,
      "step": 1730
    },
    {
      "entropy": 6.114956378936768,
      "epoch": 1.7493696419566314,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492506864491345,
      "loss": 5.8223,
      "mean_token_accuracy": 0.1336710497736931,
      "num_tokens": 3830391.0,
      "step": 1735
    },
    {
      "entropy": 5.977906370162964,
      "epoch": 1.7544125063035805,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924050106583976,
      "loss": 5.8382,
      "mean_token_accuracy": 0.13849121034145356,
      "num_tokens": 3842773.0,
      "step": 1740
    },
    {
      "entropy": 6.199583864212036,
      "epoch": 1.7594553706505294,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923024811256618,
      "loss": 5.8795,
      "mean_token_accuracy": 0.13239177986979483,
      "num_tokens": 3853646.0,
      "step": 1745
    },
    {
      "entropy": 6.085389232635498,
      "epoch": 1.7644982349974785,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000492199276211802,
      "loss": 5.8426,
      "mean_token_accuracy": 0.13615378588438035,
      "num_tokens": 3865319.0,
      "step": 1750
    },
    {
      "entropy": 6.044885206222534,
      "epoch": 1.7695410993444276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004920953962375813,
      "loss": 5.8977,
      "mean_token_accuracy": 0.13141547441482543,
      "num_tokens": 3877107.0,
      "step": 1755
    },
    {
      "entropy": 6.093702268600464,
      "epoch": 1.7745839636913767,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004919908415258612,
      "loss": 5.9368,
      "mean_token_accuracy": 0.13592029139399528,
      "num_tokens": 3889334.0,
      "step": 1760
    },
    {
      "entropy": 6.16692967414856,
      "epoch": 1.7796268280383258,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004918856124016002,
      "loss": 5.8309,
      "mean_token_accuracy": 0.13962946012616156,
      "num_tokens": 3899482.0,
      "step": 1765
    },
    {
      "entropy": 6.017588901519775,
      "epoch": 1.784669692385275,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004917797091918526,
      "loss": 5.7533,
      "mean_token_accuracy": 0.14028869047760964,
      "num_tokens": 3910216.0,
      "step": 1770
    },
    {
      "entropy": 6.108081436157226,
      "epoch": 1.789712556732224,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004916731322257683,
      "loss": 5.9182,
      "mean_token_accuracy": 0.13611244261264802,
      "num_tokens": 3922246.0,
      "step": 1775
    },
    {
      "entropy": 6.097801351547242,
      "epoch": 1.794755421079173,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004915658818345908,
      "loss": 5.8098,
      "mean_token_accuracy": 0.14234220013022422,
      "num_tokens": 3933688.0,
      "step": 1780
    },
    {
      "entropy": 6.0966685771942135,
      "epoch": 1.799798285426122,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914579583516568,
      "loss": 5.9188,
      "mean_token_accuracy": 0.1313386917114258,
      "num_tokens": 3944168.0,
      "step": 1785
    },
    {
      "entropy": 6.024723100662231,
      "epoch": 1.804841149773071,
      "grad_norm": 1.046875,
      "learning_rate": 0.000491349362112395,
      "loss": 5.8244,
      "mean_token_accuracy": 0.13950928971171378,
      "num_tokens": 3954346.0,
      "step": 1790
    },
    {
      "entropy": 6.133016967773438,
      "epoch": 1.80988401412002,
      "grad_norm": 0.890625,
      "learning_rate": 0.000491240093454325,
      "loss": 5.9441,
      "mean_token_accuracy": 0.13316622376441956,
      "num_tokens": 3966375.0,
      "step": 1795
    },
    {
      "entropy": 6.1298604011535645,
      "epoch": 1.8149268784669692,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911301527170563,
      "loss": 5.91,
      "mean_token_accuracy": 0.13569873794913292,
      "num_tokens": 3977030.0,
      "step": 1800
    },
    {
      "entropy": 6.000961494445801,
      "epoch": 1.8199697428139183,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004910195402422873,
      "loss": 5.8183,
      "mean_token_accuracy": 0.13428226485848427,
      "num_tokens": 3988064.0,
      "step": 1805
    },
    {
      "entropy": 6.15083794593811,
      "epoch": 1.8250126071608674,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909082563738042,
      "loss": 5.8778,
      "mean_token_accuracy": 0.13448215946555137,
      "num_tokens": 3997994.0,
      "step": 1810
    },
    {
      "entropy": 6.131000089645386,
      "epoch": 1.8300554715078166,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004907963014574797,
      "loss": 5.9245,
      "mean_token_accuracy": 0.13084746673703193,
      "num_tokens": 4008645.0,
      "step": 1815
    },
    {
      "entropy": 6.088234567642212,
      "epoch": 1.8350983358547657,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004906836758412722,
      "loss": 5.8739,
      "mean_token_accuracy": 0.1310124434530735,
      "num_tokens": 4020387.0,
      "step": 1820
    },
    {
      "entropy": 5.93744158744812,
      "epoch": 1.8401412002017146,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004905703798752248,
      "loss": 5.7445,
      "mean_token_accuracy": 0.138221687823534,
      "num_tokens": 4032720.0,
      "step": 1825
    },
    {
      "entropy": 6.048615217208862,
      "epoch": 1.8451840645486637,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000490456413911464,
      "loss": 5.8111,
      "mean_token_accuracy": 0.1402272567152977,
      "num_tokens": 4043438.0,
      "step": 1830
    },
    {
      "entropy": 6.0049676418304445,
      "epoch": 1.8502269288956126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004903417783041987,
      "loss": 5.8002,
      "mean_token_accuracy": 0.14097705483436584,
      "num_tokens": 4054467.0,
      "step": 1835
    },
    {
      "entropy": 6.081655836105346,
      "epoch": 1.8552697932425617,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004902264734097188,
      "loss": 5.797,
      "mean_token_accuracy": 0.13625596538186074,
      "num_tokens": 4066418.0,
      "step": 1840
    },
    {
      "entropy": 6.121115112304688,
      "epoch": 1.8603126575895108,
      "grad_norm": 0.875,
      "learning_rate": 0.0004901104995863948,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1304987095296383,
      "num_tokens": 4078770.0,
      "step": 1845
    },
    {
      "entropy": 6.036437129974365,
      "epoch": 1.86535552193646,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004899938571946759,
      "loss": 5.9277,
      "mean_token_accuracy": 0.1331052951514721,
      "num_tokens": 4089998.0,
      "step": 1850
    },
    {
      "entropy": 6.108745002746582,
      "epoch": 1.870398386283409,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004898765465970894,
      "loss": 5.84,
      "mean_token_accuracy": 0.140093395113945,
      "num_tokens": 4101409.0,
      "step": 1855
    },
    {
      "entropy": 6.08279275894165,
      "epoch": 1.8754412506303582,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004897585681582393,
      "loss": 5.9562,
      "mean_token_accuracy": 0.13204566687345504,
      "num_tokens": 4113598.0,
      "step": 1860
    },
    {
      "entropy": 6.0311401844024655,
      "epoch": 1.8804841149773073,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004896399222448053,
      "loss": 5.8903,
      "mean_token_accuracy": 0.13502624183893203,
      "num_tokens": 4124639.0,
      "step": 1865
    },
    {
      "entropy": 6.1842851638793945,
      "epoch": 1.8855269793242562,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004895206092255417,
      "loss": 5.9528,
      "mean_token_accuracy": 0.1261625297367573,
      "num_tokens": 4136280.0,
      "step": 1870
    },
    {
      "entropy": 6.0394891738891605,
      "epoch": 1.8905698436712053,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894006294712761,
      "loss": 5.798,
      "mean_token_accuracy": 0.13385596573352815,
      "num_tokens": 4148251.0,
      "step": 1875
    },
    {
      "entropy": 6.066561222076416,
      "epoch": 1.8956127080181542,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004892799833549083,
      "loss": 5.8793,
      "mean_token_accuracy": 0.13789762407541276,
      "num_tokens": 4159298.0,
      "step": 1880
    },
    {
      "entropy": 6.094835710525513,
      "epoch": 1.9006555723651033,
      "grad_norm": 1.0,
      "learning_rate": 0.0004891586712514091,
      "loss": 5.7976,
      "mean_token_accuracy": 0.13934613019227982,
      "num_tokens": 4169974.0,
      "step": 1885
    },
    {
      "entropy": 6.004667186737061,
      "epoch": 1.9056984367120524,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004890366935378196,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1426852449774742,
      "num_tokens": 4181442.0,
      "step": 1890
    },
    {
      "entropy": 6.072988843917846,
      "epoch": 1.9107413010590015,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004889140505932492,
      "loss": 5.8431,
      "mean_token_accuracy": 0.1416805826127529,
      "num_tokens": 4191535.0,
      "step": 1895
    },
    {
      "entropy": 6.028358507156372,
      "epoch": 1.9157841654059506,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887907427988749,
      "loss": 5.8658,
      "mean_token_accuracy": 0.13532739132642746,
      "num_tokens": 4201741.0,
      "step": 1900
    },
    {
      "entropy": 6.012132596969605,
      "epoch": 1.9208270297528998,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004886667705379403,
      "loss": 5.7607,
      "mean_token_accuracy": 0.1425751641392708,
      "num_tokens": 4211762.0,
      "step": 1905
    },
    {
      "entropy": 6.053383636474609,
      "epoch": 1.9258698940998489,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004885421341957541,
      "loss": 5.8402,
      "mean_token_accuracy": 0.13849018067121505,
      "num_tokens": 4222322.0,
      "step": 1910
    },
    {
      "entropy": 6.139491748809815,
      "epoch": 1.9309127584467978,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004884168341596888,
      "loss": 5.9232,
      "mean_token_accuracy": 0.13450294807553292,
      "num_tokens": 4232166.0,
      "step": 1915
    },
    {
      "entropy": 6.027013111114502,
      "epoch": 1.9359556227937469,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004882908708191797,
      "loss": 5.8756,
      "mean_token_accuracy": 0.13291192501783372,
      "num_tokens": 4243651.0,
      "step": 1920
    },
    {
      "entropy": 6.078602075576782,
      "epoch": 1.9409984871406958,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048816424456572403,
      "loss": 5.8051,
      "mean_token_accuracy": 0.14046954810619355,
      "num_tokens": 4255137.0,
      "step": 1925
    },
    {
      "entropy": 5.943156766891479,
      "epoch": 1.946041351487645,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048803695579287905,
      "loss": 5.804,
      "mean_token_accuracy": 0.13446772545576097,
      "num_tokens": 4264904.0,
      "step": 1930
    },
    {
      "entropy": 5.996735715866089,
      "epoch": 1.951084215834594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048790900489626104,
      "loss": 5.7651,
      "mean_token_accuracy": 0.1391440138220787,
      "num_tokens": 4275305.0,
      "step": 1935
    },
    {
      "entropy": 6.009812355041504,
      "epoch": 1.9561270801815431,
      "grad_norm": 1.0,
      "learning_rate": 0.0004877803922735445,
      "loss": 5.8517,
      "mean_token_accuracy": 0.13518264666199684,
      "num_tokens": 4287038.0,
      "step": 1940
    },
    {
      "entropy": 6.1983202457427975,
      "epoch": 1.9611699445284922,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004876511183244603,
      "loss": 5.9598,
      "mean_token_accuracy": 0.12747678980231286,
      "num_tokens": 4298668.0,
      "step": 1945
    },
    {
      "entropy": 5.964114427566528,
      "epoch": 1.9662128088754414,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048752118345079496,
      "loss": 5.7317,
      "mean_token_accuracy": 0.1404605582356453,
      "num_tokens": 4309429.0,
      "step": 1950
    },
    {
      "entropy": 6.004175186157227,
      "epoch": 1.9712556732223905,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048739058805638905,
      "loss": 5.8295,
      "mean_token_accuracy": 0.13574060425162315,
      "num_tokens": 4321137.0,
      "step": 1955
    },
    {
      "entropy": 6.061343431472778,
      "epoch": 1.9762985375693394,
      "grad_norm": 1.0,
      "learning_rate": 0.000487259332547136,
      "loss": 5.8552,
      "mean_token_accuracy": 0.1286994807422161,
      "num_tokens": 4331249.0,
      "step": 1960
    },
    {
      "entropy": 6.057663679122925,
      "epoch": 1.9813414019162885,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000487127417330981,
      "loss": 5.8436,
      "mean_token_accuracy": 0.1318496935069561,
      "num_tokens": 4342154.0,
      "step": 1965
    },
    {
      "entropy": 6.129978799819947,
      "epoch": 1.9863842662632374,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004869948428179196,
      "loss": 5.878,
      "mean_token_accuracy": 0.13126063123345375,
      "num_tokens": 4353906.0,
      "step": 1970
    },
    {
      "entropy": 6.023782205581665,
      "epoch": 1.9914271306101865,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004868616094199964,
      "loss": 5.9263,
      "mean_token_accuracy": 0.13177506774663925,
      "num_tokens": 4366416.0,
      "step": 1975
    },
    {
      "entropy": 6.075288438796997,
      "epoch": 1.9964699949571356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000486727717551304,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1384887605905533,
      "num_tokens": 4376743.0,
      "step": 1980
    },
    {
      "entropy": 5.952099906073676,
      "epoch": 2.00100857286939,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004865931676279813,
      "loss": 5.6221,
      "mean_token_accuracy": 0.15006235241889954,
      "num_tokens": 4385379.0,
      "step": 1985
    },
    {
      "entropy": 6.009042406082154,
      "epoch": 2.006051437216339,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004864579600682125,
      "loss": 5.5673,
      "mean_token_accuracy": 0.14534694105386733,
      "num_tokens": 4396872.0,
      "step": 1990
    },
    {
      "entropy": 6.029089117050171,
      "epoch": 2.011094301563288,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048632209529222607,
      "loss": 5.493,
      "mean_token_accuracy": 0.15098107904195784,
      "num_tokens": 4407075.0,
      "step": 1995
    },
    {
      "entropy": 5.954484272003174,
      "epoch": 2.016137165910237,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048618557372229253,
      "loss": 5.528,
      "mean_token_accuracy": 0.15284537971019746,
      "num_tokens": 4418020.0,
      "step": 2000
    },
    {
      "epoch": 2.016137165910237,
      "eval_entropy": 5.739924000116401,
      "eval_loss": 5.955663204193115,
      "eval_mean_token_accuracy": 0.13522450197440675,
      "eval_num_tokens": 4418020.0,
      "eval_runtime": 2.0884,
      "eval_samples_per_second": 1502.568,
      "eval_steps_per_second": 188.18,
      "step": 2000
    },
    {
      "entropy": 5.945985460281372,
      "epoch": 2.0211800302571863,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004860483957827243,
      "loss": 5.5488,
      "mean_token_accuracy": 0.15342542231082917,
      "num_tokens": 4429740.0,
      "step": 2005
    },
    {
      "entropy": 6.039470911026001,
      "epoch": 2.026222894604135,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004859105618998735,
      "loss": 5.5347,
      "mean_token_accuracy": 0.14306792616844177,
      "num_tokens": 4440183.0,
      "step": 2010
    },
    {
      "entropy": 5.972193956375122,
      "epoch": 2.031265758951084,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004857720725021311,
      "loss": 5.5174,
      "mean_token_accuracy": 0.14989553838968278,
      "num_tokens": 4451750.0,
      "step": 2015
    },
    {
      "entropy": 5.816289949417114,
      "epoch": 2.036308623298033,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048563292801992534,
      "loss": 5.4737,
      "mean_token_accuracy": 0.15668680369853974,
      "num_tokens": 4462071.0,
      "step": 2020
    },
    {
      "entropy": 6.050288486480713,
      "epoch": 2.0413514876449823,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004854931288857203,
      "loss": 5.6568,
      "mean_token_accuracy": 0.1456625282764435,
      "num_tokens": 4474241.0,
      "step": 2025
    },
    {
      "entropy": 5.960632467269898,
      "epoch": 2.0463943519919314,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048535267553401505,
      "loss": 5.4922,
      "mean_token_accuracy": 0.1506148397922516,
      "num_tokens": 4487083.0,
      "step": 2030
    },
    {
      "entropy": 5.86674542427063,
      "epoch": 2.0514372163388805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048521156840134193,
      "loss": 5.5406,
      "mean_token_accuracy": 0.15699613094329834,
      "num_tokens": 4497181.0,
      "step": 2035
    },
    {
      "entropy": 6.016703796386719,
      "epoch": 2.0564800806858297,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000485069807926265,
      "loss": 5.5274,
      "mean_token_accuracy": 0.15277422219514847,
      "num_tokens": 4508250.0,
      "step": 2040
    },
    {
      "entropy": 5.8712366104125975,
      "epoch": 2.0615229450327788,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004849273945493792,
      "loss": 5.5229,
      "mean_token_accuracy": 0.15639246255159378,
      "num_tokens": 4519778.0,
      "step": 2045
    },
    {
      "entropy": 5.941642713546753,
      "epoch": 2.066565809379728,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004847843287133084,
      "loss": 5.5055,
      "mean_token_accuracy": 0.14586257934570312,
      "num_tokens": 4530879.0,
      "step": 2050
    },
    {
      "entropy": 5.944451856613159,
      "epoch": 2.0716086737266766,
      "grad_norm": 0.875,
      "learning_rate": 0.0004846406108627048,
      "loss": 5.5719,
      "mean_token_accuracy": 0.15614590048789978,
      "num_tokens": 4542560.0,
      "step": 2055
    },
    {
      "entropy": 5.872511577606201,
      "epoch": 2.0766515380736257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048449624144424665,
      "loss": 5.4693,
      "mean_token_accuracy": 0.1582667663693428,
      "num_tokens": 4552213.0,
      "step": 2060
    },
    {
      "entropy": 5.8945520401000975,
      "epoch": 2.081694402420575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048435122090663744,
      "loss": 5.4888,
      "mean_token_accuracy": 0.1566730335354805,
      "num_tokens": 4562798.0,
      "step": 2065
    },
    {
      "entropy": 5.927812004089356,
      "epoch": 2.086737266767524,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048420554970060424,
      "loss": 5.5391,
      "mean_token_accuracy": 0.14681653976440429,
      "num_tokens": 4574534.0,
      "step": 2070
    },
    {
      "entropy": 5.931223726272583,
      "epoch": 2.091780131114473,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048405922827889677,
      "loss": 5.5042,
      "mean_token_accuracy": 0.14926443621516228,
      "num_tokens": 4586239.0,
      "step": 2075
    },
    {
      "entropy": 5.888582849502564,
      "epoch": 2.096822995461422,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048391225709628515,
      "loss": 5.5507,
      "mean_token_accuracy": 0.1506632573902607,
      "num_tokens": 4598018.0,
      "step": 2080
    },
    {
      "entropy": 5.933531332015991,
      "epoch": 2.1018658598083713,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004837646366095593,
      "loss": 5.5267,
      "mean_token_accuracy": 0.14935630857944487,
      "num_tokens": 4608985.0,
      "step": 2085
    },
    {
      "entropy": 5.903959321975708,
      "epoch": 2.1069087241553204,
      "grad_norm": 1.125,
      "learning_rate": 0.00048361636727752716,
      "loss": 5.5818,
      "mean_token_accuracy": 0.15361014753580093,
      "num_tokens": 4621103.0,
      "step": 2090
    },
    {
      "entropy": 5.8726660251617435,
      "epoch": 2.1119515885022695,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004834674495610131,
      "loss": 5.5588,
      "mean_token_accuracy": 0.15249116122722625,
      "num_tokens": 4632244.0,
      "step": 2095
    },
    {
      "entropy": 5.908159255981445,
      "epoch": 2.116994452849218,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048331788392285693,
      "loss": 5.4765,
      "mean_token_accuracy": 0.15572449415922165,
      "num_tokens": 4642387.0,
      "step": 2100
    },
    {
      "entropy": 5.934552431106567,
      "epoch": 2.1220373171961673,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048316767082791205,
      "loss": 5.6106,
      "mean_token_accuracy": 0.1467052735388279,
      "num_tokens": 4655181.0,
      "step": 2105
    },
    {
      "entropy": 5.8572265625,
      "epoch": 2.1270801815431164,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048301681074304427,
      "loss": 5.5122,
      "mean_token_accuracy": 0.14930952489376068,
      "num_tokens": 4665664.0,
      "step": 2110
    },
    {
      "entropy": 5.829651737213135,
      "epoch": 2.1321230458900655,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048286530413713016,
      "loss": 5.4759,
      "mean_token_accuracy": 0.15683304220438005,
      "num_tokens": 4677683.0,
      "step": 2115
    },
    {
      "entropy": 5.921738862991333,
      "epoch": 2.1371659102370146,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048271315148105574,
      "loss": 5.5304,
      "mean_token_accuracy": 0.14627472013235093,
      "num_tokens": 4688891.0,
      "step": 2120
    },
    {
      "entropy": 5.849819469451904,
      "epoch": 2.1422087745839637,
      "grad_norm": 1.046875,
      "learning_rate": 0.000482560353247715,
      "loss": 5.5568,
      "mean_token_accuracy": 0.15030487179756163,
      "num_tokens": 4700765.0,
      "step": 2125
    },
    {
      "entropy": 5.8970269680023195,
      "epoch": 2.147251638930913,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004824069099120084,
      "loss": 5.576,
      "mean_token_accuracy": 0.14962146580219268,
      "num_tokens": 4712004.0,
      "step": 2130
    },
    {
      "entropy": 5.896734142303467,
      "epoch": 2.152294503277862,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004822528219508413,
      "loss": 5.487,
      "mean_token_accuracy": 0.15196678787469864,
      "num_tokens": 4722427.0,
      "step": 2135
    },
    {
      "entropy": 5.9260717868804935,
      "epoch": 2.157337367624811,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004820980898431227,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1583951383829117,
      "num_tokens": 4733815.0,
      "step": 2140
    },
    {
      "entropy": 5.862832736968994,
      "epoch": 2.1623802319717598,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004819427140697635,
      "loss": 5.5166,
      "mean_token_accuracy": 0.151727132499218,
      "num_tokens": 4745070.0,
      "step": 2145
    },
    {
      "entropy": 5.931176424026489,
      "epoch": 2.167423096318709,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004817866951136753,
      "loss": 5.5859,
      "mean_token_accuracy": 0.14487538561224939,
      "num_tokens": 4756066.0,
      "step": 2150
    },
    {
      "entropy": 5.905827474594116,
      "epoch": 2.172465960665658,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048163003345976847,
      "loss": 5.5925,
      "mean_token_accuracy": 0.14235491454601287,
      "num_tokens": 4767153.0,
      "step": 2155
    },
    {
      "entropy": 5.907755947113037,
      "epoch": 2.177508825012607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004814727295949512,
      "loss": 5.5016,
      "mean_token_accuracy": 0.15076689422130585,
      "num_tokens": 4777554.0,
      "step": 2160
    },
    {
      "entropy": 5.863094186782837,
      "epoch": 2.1825516893595562,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048131478400812735,
      "loss": 5.5516,
      "mean_token_accuracy": 0.1508598580956459,
      "num_tokens": 4788881.0,
      "step": 2165
    },
    {
      "entropy": 5.963677263259887,
      "epoch": 2.1875945537065054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004811561971901956,
      "loss": 5.5276,
      "mean_token_accuracy": 0.14635513722896576,
      "num_tokens": 4798680.0,
      "step": 2170
    },
    {
      "entropy": 5.903114986419678,
      "epoch": 2.1926374180534545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004809969696340474,
      "loss": 5.5199,
      "mean_token_accuracy": 0.1514839932322502,
      "num_tokens": 4808096.0,
      "step": 2175
    },
    {
      "entropy": 5.803523302078247,
      "epoch": 2.1976802824004036,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048083710183456554,
      "loss": 5.4785,
      "mean_token_accuracy": 0.15707470923662187,
      "num_tokens": 4818778.0,
      "step": 2180
    },
    {
      "entropy": 5.820753574371338,
      "epoch": 2.2027231467473527,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048067659428862306,
      "loss": 5.4739,
      "mean_token_accuracy": 0.1569182261824608,
      "num_tokens": 4828586.0,
      "step": 2185
    },
    {
      "entropy": 5.870742511749268,
      "epoch": 2.2077660110943014,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048051544749508096,
      "loss": 5.5666,
      "mean_token_accuracy": 0.14620909839868546,
      "num_tokens": 4839419.0,
      "step": 2190
    },
    {
      "entropy": 5.873201608657837,
      "epoch": 2.2128088754412505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004803536619547872,
      "loss": 5.503,
      "mean_token_accuracy": 0.15512274950742722,
      "num_tokens": 4849700.0,
      "step": 2195
    },
    {
      "entropy": 5.891473436355591,
      "epoch": 2.2178517397881996,
      "grad_norm": 1.03125,
      "learning_rate": 0.000480191238170575,
      "loss": 5.6601,
      "mean_token_accuracy": 0.14363080710172654,
      "num_tokens": 4860261.0,
      "step": 2200
    },
    {
      "entropy": 5.899570369720459,
      "epoch": 2.2228946041351487,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004800281766472612,
      "loss": 5.5751,
      "mean_token_accuracy": 0.14639226496219634,
      "num_tokens": 4872277.0,
      "step": 2205
    },
    {
      "entropy": 5.893243741989136,
      "epoch": 2.227937468482098,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047986447789164486,
      "loss": 5.6251,
      "mean_token_accuracy": 0.15411222875118255,
      "num_tokens": 4883512.0,
      "step": 2210
    },
    {
      "entropy": 5.878543472290039,
      "epoch": 2.232980332829047,
      "grad_norm": 1.0,
      "learning_rate": 0.0004797001424125054,
      "loss": 5.549,
      "mean_token_accuracy": 0.15132871270179749,
      "num_tokens": 4896197.0,
      "step": 2215
    },
    {
      "entropy": 5.9062926292419435,
      "epoch": 2.238023197175996,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047953517072060124,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1517067864537239,
      "num_tokens": 4907397.0,
      "step": 2220
    },
    {
      "entropy": 5.830627298355102,
      "epoch": 2.243066061522945,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004793695633286684,
      "loss": 5.5807,
      "mean_token_accuracy": 0.15095050781965255,
      "num_tokens": 4919008.0,
      "step": 2225
    },
    {
      "entropy": 5.853465175628662,
      "epoch": 2.2481089258698943,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047920332075141825,
      "loss": 5.5096,
      "mean_token_accuracy": 0.14976307600736619,
      "num_tokens": 4930877.0,
      "step": 2230
    },
    {
      "entropy": 5.817238712310791,
      "epoch": 2.253151790216843,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004790364435055367,
      "loss": 5.425,
      "mean_token_accuracy": 0.16180467009544372,
      "num_tokens": 4942274.0,
      "step": 2235
    },
    {
      "entropy": 5.863884592056275,
      "epoch": 2.258194654563792,
      "grad_norm": 1.0,
      "learning_rate": 0.000478868932109682,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1519279509782791,
      "num_tokens": 4953740.0,
      "step": 2240
    },
    {
      "entropy": 5.873799562454224,
      "epoch": 2.263237518910741,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004787007870844835,
      "loss": 5.5517,
      "mean_token_accuracy": 0.15242820233106613,
      "num_tokens": 4964760.0,
      "step": 2245
    },
    {
      "entropy": 5.898807573318481,
      "epoch": 2.2682803832576903,
      "grad_norm": 1.0,
      "learning_rate": 0.00047853200895253976,
      "loss": 5.6184,
      "mean_token_accuracy": 0.14816488102078437,
      "num_tokens": 4976404.0,
      "step": 2250
    },
    {
      "entropy": 5.847094631195068,
      "epoch": 2.2733232476046394,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047836259823841716,
      "loss": 5.5016,
      "mean_token_accuracy": 0.14656250029802323,
      "num_tokens": 4987552.0,
      "step": 2255
    },
    {
      "entropy": 5.873461771011352,
      "epoch": 2.2783661119515886,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047819255546864797,
      "loss": 5.5156,
      "mean_token_accuracy": 0.15206850171089173,
      "num_tokens": 4998912.0,
      "step": 2260
    },
    {
      "entropy": 5.86563892364502,
      "epoch": 2.2834089762985377,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004780218811717293,
      "loss": 5.4861,
      "mean_token_accuracy": 0.15683870911598205,
      "num_tokens": 5009556.0,
      "step": 2265
    },
    {
      "entropy": 5.8067628860473635,
      "epoch": 2.288451840645487,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047785057587812057,
      "loss": 5.5311,
      "mean_token_accuracy": 0.15605937987565993,
      "num_tokens": 5021569.0,
      "step": 2270
    },
    {
      "entropy": 5.900359725952148,
      "epoch": 2.293494704992436,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047767864012024266,
      "loss": 5.6064,
      "mean_token_accuracy": 0.1454856812953949,
      "num_tokens": 5032994.0,
      "step": 2275
    },
    {
      "entropy": 5.924926233291626,
      "epoch": 2.2985375693393846,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047750607443247585,
      "loss": 5.5447,
      "mean_token_accuracy": 0.14936016798019408,
      "num_tokens": 5044681.0,
      "step": 2280
    },
    {
      "entropy": 5.762581253051758,
      "epoch": 2.3035804336863337,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047733287935115837,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16046452075242995,
      "num_tokens": 5055509.0,
      "step": 2285
    },
    {
      "entropy": 5.763984966278076,
      "epoch": 2.308623298033283,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004771590554145844,
      "loss": 5.4838,
      "mean_token_accuracy": 0.15912238359451295,
      "num_tokens": 5066903.0,
      "step": 2290
    },
    {
      "entropy": 5.854858827590943,
      "epoch": 2.313666162380232,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047698460316300285,
      "loss": 5.5341,
      "mean_token_accuracy": 0.14713411778211594,
      "num_tokens": 5078411.0,
      "step": 2295
    },
    {
      "entropy": 5.831680345535278,
      "epoch": 2.318709026727181,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004768095231386153,
      "loss": 5.5526,
      "mean_token_accuracy": 0.14783076643943788,
      "num_tokens": 5089811.0,
      "step": 2300
    },
    {
      "entropy": 5.905092048645019,
      "epoch": 2.32375189107413,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004766338158855744,
      "loss": 5.5318,
      "mean_token_accuracy": 0.15881457775831223,
      "num_tokens": 5101196.0,
      "step": 2305
    },
    {
      "entropy": 5.782911729812622,
      "epoch": 2.3287947554210793,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004764574819499826,
      "loss": 5.5179,
      "mean_token_accuracy": 0.15974755138158797,
      "num_tokens": 5112379.0,
      "step": 2310
    },
    {
      "entropy": 5.952156448364258,
      "epoch": 2.3338376197680284,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047628052187988955,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1402716465294361,
      "num_tokens": 5125065.0,
      "step": 2315
    },
    {
      "entropy": 5.938916254043579,
      "epoch": 2.338880484114977,
      "grad_norm": 1.0,
      "learning_rate": 0.0004761029362252914,
      "loss": 5.611,
      "mean_token_accuracy": 0.15130804926157,
      "num_tokens": 5137471.0,
      "step": 2320
    },
    {
      "entropy": 5.8899760246276855,
      "epoch": 2.343923348461926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047592472553812856,
      "loss": 5.54,
      "mean_token_accuracy": 0.15442633181810378,
      "num_tokens": 5147682.0,
      "step": 2325
    },
    {
      "entropy": 5.835421705245972,
      "epoch": 2.3489662128088753,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004757458903722837,
      "loss": 5.514,
      "mean_token_accuracy": 0.14723399579524993,
      "num_tokens": 5156143.0,
      "step": 2330
    },
    {
      "entropy": 5.86648907661438,
      "epoch": 2.3540090771558244,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047556643128358075,
      "loss": 5.5784,
      "mean_token_accuracy": 0.14724974930286408,
      "num_tokens": 5167462.0,
      "step": 2335
    },
    {
      "entropy": 5.814883089065551,
      "epoch": 2.3590519415027735,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047538634882978277,
      "loss": 5.495,
      "mean_token_accuracy": 0.15153746753931047,
      "num_tokens": 5178030.0,
      "step": 2340
    },
    {
      "entropy": 5.757813596725464,
      "epoch": 2.3640948058497226,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047520564357059013,
      "loss": 5.499,
      "mean_token_accuracy": 0.15266751050949096,
      "num_tokens": 5188296.0,
      "step": 2345
    },
    {
      "entropy": 5.773411130905151,
      "epoch": 2.3691376701966718,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004750243160676389,
      "loss": 5.5133,
      "mean_token_accuracy": 0.15319937467575073,
      "num_tokens": 5199058.0,
      "step": 2350
    },
    {
      "entropy": 5.8733357906341555,
      "epoch": 2.374180534543621,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004748423668844992,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1519238084554672,
      "num_tokens": 5209314.0,
      "step": 2355
    },
    {
      "entropy": 5.862629032135009,
      "epoch": 2.37922339889057,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004746597965866733,
      "loss": 5.5843,
      "mean_token_accuracy": 0.14848749935626984,
      "num_tokens": 5220248.0,
      "step": 2360
    },
    {
      "entropy": 5.890047597885132,
      "epoch": 2.384266263237519,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004744766057415938,
      "loss": 5.5677,
      "mean_token_accuracy": 0.1431344673037529,
      "num_tokens": 5230874.0,
      "step": 2365
    },
    {
      "entropy": 5.85345344543457,
      "epoch": 2.389309127584468,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004742927949186222,
      "loss": 5.5003,
      "mean_token_accuracy": 0.15552936047315596,
      "num_tokens": 5243328.0,
      "step": 2370
    },
    {
      "entropy": 5.741024541854858,
      "epoch": 2.394351991931417,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047410836468904664,
      "loss": 5.4638,
      "mean_token_accuracy": 0.1511007145047188,
      "num_tokens": 5254357.0,
      "step": 2375
    },
    {
      "entropy": 5.789620447158813,
      "epoch": 2.399394856278366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047392331562608073,
      "loss": 5.4534,
      "mean_token_accuracy": 0.15577590167522432,
      "num_tokens": 5264492.0,
      "step": 2380
    },
    {
      "entropy": 5.7129381656646725,
      "epoch": 2.404437720625315,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004737376483048609,
      "loss": 5.4273,
      "mean_token_accuracy": 0.15612087547779083,
      "num_tokens": 5274816.0,
      "step": 2385
    },
    {
      "entropy": 5.830111742019653,
      "epoch": 2.4094805849722642,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004735513633024458,
      "loss": 5.5253,
      "mean_token_accuracy": 0.15363801121711732,
      "num_tokens": 5286855.0,
      "step": 2390
    },
    {
      "entropy": 5.920066022872925,
      "epoch": 2.4145234493192134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004733644611978134,
      "loss": 5.6269,
      "mean_token_accuracy": 0.13929646238684654,
      "num_tokens": 5299594.0,
      "step": 2395
    },
    {
      "entropy": 5.762104845046997,
      "epoch": 2.4195663136661625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004731769425718597,
      "loss": 5.4584,
      "mean_token_accuracy": 0.15275404006242752,
      "num_tokens": 5310078.0,
      "step": 2400
    },
    {
      "entropy": 5.798458290100098,
      "epoch": 2.4246091780131116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000472988808007397,
      "loss": 5.4842,
      "mean_token_accuracy": 0.15803922563791276,
      "num_tokens": 5319544.0,
      "step": 2405
    },
    {
      "entropy": 5.8020671844482425,
      "epoch": 2.4296520423600603,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00047280005808915176,
      "loss": 5.5491,
      "mean_token_accuracy": 0.14798999726772308,
      "num_tokens": 5331025.0,
      "step": 2410
    },
    {
      "entropy": 5.743836212158203,
      "epoch": 2.4346949067070094,
      "grad_norm": 1.125,
      "learning_rate": 0.0004726106934037631,
      "loss": 5.5195,
      "mean_token_accuracy": 0.15750828236341477,
      "num_tokens": 5343086.0,
      "step": 2415
    },
    {
      "entropy": 5.946476697921753,
      "epoch": 2.4397377710539585,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004724207145397809,
      "loss": 5.5844,
      "mean_token_accuracy": 0.1544722720980644,
      "num_tokens": 5353598.0,
      "step": 2420
    },
    {
      "entropy": 5.801823663711548,
      "epoch": 2.4447806354009076,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047223012208766384,
      "loss": 5.4544,
      "mean_token_accuracy": 0.1517728313803673,
      "num_tokens": 5363607.0,
      "step": 2425
    },
    {
      "entropy": 5.745185041427613,
      "epoch": 2.4498234997478567,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004720389166397777,
      "loss": 5.5181,
      "mean_token_accuracy": 0.15420024693012238,
      "num_tokens": 5375198.0,
      "step": 2430
    },
    {
      "entropy": 5.788627862930298,
      "epoch": 2.454866364094806,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047184709879039325,
      "loss": 5.5002,
      "mean_token_accuracy": 0.1526012524962425,
      "num_tokens": 5386058.0,
      "step": 2435
    },
    {
      "entropy": 5.888590955734253,
      "epoch": 2.459909228441755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047165466913568497,
      "loss": 5.5293,
      "mean_token_accuracy": 0.15335189551115036,
      "num_tokens": 5395720.0,
      "step": 2440
    },
    {
      "entropy": 5.8117420196533205,
      "epoch": 2.464952092788704,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004714616282737285,
      "loss": 5.5663,
      "mean_token_accuracy": 0.15131355226039886,
      "num_tokens": 5406252.0,
      "step": 2445
    },
    {
      "entropy": 5.824741077423096,
      "epoch": 2.469994957135653,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004712679768044995,
      "loss": 5.5712,
      "mean_token_accuracy": 0.15128226727247238,
      "num_tokens": 5418125.0,
      "step": 2450
    },
    {
      "entropy": 5.867142248153686,
      "epoch": 2.4750378214826023,
      "grad_norm": 1.125,
      "learning_rate": 0.000471073715329871,
      "loss": 5.5349,
      "mean_token_accuracy": 0.1528469055891037,
      "num_tokens": 5428095.0,
      "step": 2455
    },
    {
      "entropy": 5.851259517669678,
      "epoch": 2.4800806858295514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004708788444536123,
      "loss": 5.5422,
      "mean_token_accuracy": 0.14842627048492432,
      "num_tokens": 5438359.0,
      "step": 2460
    },
    {
      "entropy": 5.791734457015991,
      "epoch": 2.4851235501765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004706833647813865,
      "loss": 5.4803,
      "mean_token_accuracy": 0.1604042187333107,
      "num_tokens": 5448559.0,
      "step": 2465
    },
    {
      "entropy": 5.730999326705932,
      "epoch": 2.4901664145234492,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047048727692074893,
      "loss": 5.4955,
      "mean_token_accuracy": 0.15414907783269882,
      "num_tokens": 5459739.0,
      "step": 2470
    },
    {
      "entropy": 5.774705123901367,
      "epoch": 2.4952092788703983,
      "grad_norm": 1.0,
      "learning_rate": 0.0004702905814811452,
      "loss": 5.4687,
      "mean_token_accuracy": 0.14784765392541885,
      "num_tokens": 5471160.0,
      "step": 2475
    },
    {
      "entropy": 5.737928628921509,
      "epoch": 2.5002521432173475,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047009327907390926,
      "loss": 5.5336,
      "mean_token_accuracy": 0.14980174079537392,
      "num_tokens": 5482344.0,
      "step": 2480
    },
    {
      "entropy": 5.863484191894531,
      "epoch": 2.5052950075642966,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004698953703122616,
      "loss": 5.5357,
      "mean_token_accuracy": 0.15072937160730362,
      "num_tokens": 5493111.0,
      "step": 2485
    },
    {
      "entropy": 5.6815407276153564,
      "epoch": 2.5103378719112457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046969685581130715,
      "loss": 5.4669,
      "mean_token_accuracy": 0.15772203207015992,
      "num_tokens": 5503877.0,
      "step": 2490
    },
    {
      "entropy": 5.846375513076782,
      "epoch": 2.515380736258195,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046949773618803353,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1496770590543747,
      "num_tokens": 5514067.0,
      "step": 2495
    },
    {
      "entropy": 5.855765962600708,
      "epoch": 2.5204236006051435,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046929801206130925,
      "loss": 5.5303,
      "mean_token_accuracy": 0.15081025362014772,
      "num_tokens": 5525534.0,
      "step": 2500
    },
    {
      "epoch": 2.5204236006051435,
      "eval_entropy": 5.665463418450974,
      "eval_loss": 5.860804080963135,
      "eval_mean_token_accuracy": 0.14090687732280968,
      "eval_num_tokens": 5525534.0,
      "eval_runtime": 2.0444,
      "eval_samples_per_second": 1534.955,
      "eval_steps_per_second": 192.236,
      "step": 2500
    },
    {
      "entropy": 5.806359052658081,
      "epoch": 2.5254664649520926,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046909768405188135,
      "loss": 5.5048,
      "mean_token_accuracy": 0.15887946933507918,
      "num_tokens": 5535525.0,
      "step": 2505
    },
    {
      "entropy": 5.754853534698486,
      "epoch": 2.5305093292990417,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046889675278237404,
      "loss": 5.4963,
      "mean_token_accuracy": 0.15863076150417327,
      "num_tokens": 5546129.0,
      "step": 2510
    },
    {
      "entropy": 5.880631685256958,
      "epoch": 2.535552193645991,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004686952188772863,
      "loss": 5.643,
      "mean_token_accuracy": 0.1497264802455902,
      "num_tokens": 5557548.0,
      "step": 2515
    },
    {
      "entropy": 5.889855480194091,
      "epoch": 2.54059505799294,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046849308296299005,
      "loss": 5.5396,
      "mean_token_accuracy": 0.1546356976032257,
      "num_tokens": 5567367.0,
      "step": 2520
    },
    {
      "entropy": 5.80710654258728,
      "epoch": 2.545637922339889,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004682903456677285,
      "loss": 5.4579,
      "mean_token_accuracy": 0.15481625944375993,
      "num_tokens": 5577489.0,
      "step": 2525
    },
    {
      "entropy": 5.760212707519531,
      "epoch": 2.550680786686838,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046808700762161385,
      "loss": 5.512,
      "mean_token_accuracy": 0.1453149624168873,
      "num_tokens": 5587877.0,
      "step": 2530
    },
    {
      "entropy": 5.767353677749634,
      "epoch": 2.5557236510337873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004678830694566253,
      "loss": 5.502,
      "mean_token_accuracy": 0.15954444408416749,
      "num_tokens": 5597863.0,
      "step": 2535
    },
    {
      "entropy": 5.798882102966308,
      "epoch": 2.5607665153807364,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004676785318066075,
      "loss": 5.5212,
      "mean_token_accuracy": 0.15344623178243638,
      "num_tokens": 5608141.0,
      "step": 2540
    },
    {
      "entropy": 5.790043830871582,
      "epoch": 2.5658093797276855,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004674733953072682,
      "loss": 5.5958,
      "mean_token_accuracy": 0.14917798191308976,
      "num_tokens": 5619660.0,
      "step": 2545
    },
    {
      "entropy": 5.935689163208008,
      "epoch": 2.5708522440746346,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004672676605961763,
      "loss": 5.5701,
      "mean_token_accuracy": 0.14701586365699768,
      "num_tokens": 5630271.0,
      "step": 2550
    },
    {
      "entropy": 5.801031875610351,
      "epoch": 2.5758951084215833,
      "grad_norm": 1.0,
      "learning_rate": 0.00046706132831276007,
      "loss": 5.5756,
      "mean_token_accuracy": 0.1508823052048683,
      "num_tokens": 5642594.0,
      "step": 2555
    },
    {
      "entropy": 5.76602954864502,
      "epoch": 2.5809379727685324,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046685439909830486,
      "loss": 5.5023,
      "mean_token_accuracy": 0.15386482924222947,
      "num_tokens": 5653361.0,
      "step": 2560
    },
    {
      "entropy": 5.757552146911621,
      "epoch": 2.5859808371154815,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004666468735959516,
      "loss": 5.4214,
      "mean_token_accuracy": 0.15881368070840834,
      "num_tokens": 5664167.0,
      "step": 2565
    },
    {
      "entropy": 5.774722766876221,
      "epoch": 2.5910237014624307,
      "grad_norm": 1.125,
      "learning_rate": 0.0004664387524506943,
      "loss": 5.5175,
      "mean_token_accuracy": 0.15454470217227936,
      "num_tokens": 5675903.0,
      "step": 2570
    },
    {
      "entropy": 5.815897417068482,
      "epoch": 2.59606656580938,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004662300363093784,
      "loss": 5.4963,
      "mean_token_accuracy": 0.15150748044252396,
      "num_tokens": 5686319.0,
      "step": 2575
    },
    {
      "entropy": 5.903947401046753,
      "epoch": 2.601109430156329,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004660207258206983,
      "loss": 5.5975,
      "mean_token_accuracy": 0.1484207347035408,
      "num_tokens": 5696000.0,
      "step": 2580
    },
    {
      "entropy": 5.775115823745727,
      "epoch": 2.606152294503278,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046581082163519585,
      "loss": 5.5039,
      "mean_token_accuracy": 0.1515410214662552,
      "num_tokens": 5708145.0,
      "step": 2585
    },
    {
      "entropy": 5.697235536575318,
      "epoch": 2.6111951588502267,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004656003244052582,
      "loss": 5.4467,
      "mean_token_accuracy": 0.16312831789255142,
      "num_tokens": 5718053.0,
      "step": 2590
    },
    {
      "entropy": 5.809553384780884,
      "epoch": 2.616238023197176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004653892347851155,
      "loss": 5.5329,
      "mean_token_accuracy": 0.15416440665721892,
      "num_tokens": 5728723.0,
      "step": 2595
    },
    {
      "entropy": 5.7317040920257565,
      "epoch": 2.621280887544125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004651775534308392,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1595622032880783,
      "num_tokens": 5739414.0,
      "step": 2600
    },
    {
      "entropy": 5.846518087387085,
      "epoch": 2.626323751891074,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649652810003398,
      "loss": 5.5945,
      "mean_token_accuracy": 0.14265543445944787,
      "num_tokens": 5751247.0,
      "step": 2605
    },
    {
      "entropy": 5.754087924957275,
      "epoch": 2.631366616238023,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046475241815336504,
      "loss": 5.4567,
      "mean_token_accuracy": 0.157156303524971,
      "num_tokens": 5761791.0,
      "step": 2610
    },
    {
      "entropy": 5.701113224029541,
      "epoch": 2.6364094805849723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004645389655514974,
      "loss": 5.4455,
      "mean_token_accuracy": 0.15011154115200043,
      "num_tokens": 5773597.0,
      "step": 2615
    },
    {
      "entropy": 5.800614356994629,
      "epoch": 2.6414523449319214,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004643249238581525,
      "loss": 5.4736,
      "mean_token_accuracy": 0.15215098410844802,
      "num_tokens": 5782963.0,
      "step": 2620
    },
    {
      "entropy": 5.682878828048706,
      "epoch": 2.6464952092788705,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004641102937385769,
      "loss": 5.4995,
      "mean_token_accuracy": 0.15128629058599471,
      "num_tokens": 5793474.0,
      "step": 2625
    },
    {
      "entropy": 5.812402820587158,
      "epoch": 2.6515380736258196,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004638950758598461,
      "loss": 5.4867,
      "mean_token_accuracy": 0.15513542294502258,
      "num_tokens": 5804018.0,
      "step": 2630
    },
    {
      "entropy": 5.836827611923217,
      "epoch": 2.6565809379727687,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046367927089086203,
      "loss": 5.5222,
      "mean_token_accuracy": 0.150009623169899,
      "num_tokens": 5815897.0,
      "step": 2635
    },
    {
      "entropy": 5.790867662429809,
      "epoch": 2.661623802319718,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046346287950235163,
      "loss": 5.5416,
      "mean_token_accuracy": 0.15370291024446486,
      "num_tokens": 5825883.0,
      "step": 2640
    },
    {
      "entropy": 5.817716598510742,
      "epoch": 2.6666666666666665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004632459023668642,
      "loss": 5.5153,
      "mean_token_accuracy": 0.1593001514673233,
      "num_tokens": 5836668.0,
      "step": 2645
    },
    {
      "entropy": 5.830860710144043,
      "epoch": 2.6717095310136156,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004630283401587699,
      "loss": 5.6442,
      "mean_token_accuracy": 0.14761790335178376,
      "num_tokens": 5848497.0,
      "step": 2650
    },
    {
      "entropy": 5.747366666793823,
      "epoch": 2.6767523953605648,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004628101935542568,
      "loss": 5.4604,
      "mean_token_accuracy": 0.15803054869174957,
      "num_tokens": 5859231.0,
      "step": 2655
    },
    {
      "entropy": 5.792496824264527,
      "epoch": 2.681795259707514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046259146323132986,
      "loss": 5.595,
      "mean_token_accuracy": 0.1492522418498993,
      "num_tokens": 5871190.0,
      "step": 2660
    },
    {
      "entropy": 5.80987024307251,
      "epoch": 2.686838124054463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046237214986980776,
      "loss": 5.4762,
      "mean_token_accuracy": 0.153760689496994,
      "num_tokens": 5881788.0,
      "step": 2665
    },
    {
      "entropy": 5.759541654586792,
      "epoch": 2.691880988401412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004621522541513216,
      "loss": 5.4486,
      "mean_token_accuracy": 0.15549346655607224,
      "num_tokens": 5891289.0,
      "step": 2670
    },
    {
      "entropy": 5.719479513168335,
      "epoch": 2.696923852748361,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004619317767593123,
      "loss": 5.4831,
      "mean_token_accuracy": 0.15185485780239105,
      "num_tokens": 5903594.0,
      "step": 2675
    },
    {
      "entropy": 5.8441746711730955,
      "epoch": 2.70196671709531,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046171071837902873,
      "loss": 5.5976,
      "mean_token_accuracy": 0.14605054557323455,
      "num_tokens": 5914592.0,
      "step": 2680
    },
    {
      "entropy": 5.7381562232971195,
      "epoch": 2.707009581442259,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004614890796975254,
      "loss": 5.4318,
      "mean_token_accuracy": 0.1508565589785576,
      "num_tokens": 5927137.0,
      "step": 2685
    },
    {
      "entropy": 5.737528085708618,
      "epoch": 2.712052445789208,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004612668614036605,
      "loss": 5.541,
      "mean_token_accuracy": 0.15300887674093247,
      "num_tokens": 5938730.0,
      "step": 2690
    },
    {
      "entropy": 5.816342782974243,
      "epoch": 2.7170953101361572,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004610440641880934,
      "loss": 5.5557,
      "mean_token_accuracy": 0.14880598187446595,
      "num_tokens": 5948912.0,
      "step": 2695
    },
    {
      "entropy": 5.703631258010864,
      "epoch": 2.7221381744831064,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046082068874328324,
      "loss": 5.479,
      "mean_token_accuracy": 0.15688103288412095,
      "num_tokens": 5960253.0,
      "step": 2700
    },
    {
      "entropy": 5.805550909042358,
      "epoch": 2.7271810388300555,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046059673576348596,
      "loss": 5.5666,
      "mean_token_accuracy": 0.14729509204626084,
      "num_tokens": 5972615.0,
      "step": 2705
    },
    {
      "entropy": 5.851122283935547,
      "epoch": 2.7322239031770046,
      "grad_norm": 1.0,
      "learning_rate": 0.00046037220594475264,
      "loss": 5.557,
      "mean_token_accuracy": 0.15245793312788009,
      "num_tokens": 5985023.0,
      "step": 2710
    },
    {
      "entropy": 5.730431699752808,
      "epoch": 2.7372667675239537,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000460147099984927,
      "loss": 5.4472,
      "mean_token_accuracy": 0.16028818488121033,
      "num_tokens": 5996445.0,
      "step": 2715
    },
    {
      "entropy": 5.774463891983032,
      "epoch": 2.742309631870903,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004599214185836438,
      "loss": 5.4948,
      "mean_token_accuracy": 0.15225832313299179,
      "num_tokens": 6007385.0,
      "step": 2720
    },
    {
      "entropy": 5.750139379501343,
      "epoch": 2.747352496217852,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045969516244232574,
      "loss": 5.5376,
      "mean_token_accuracy": 0.15341005325317383,
      "num_tokens": 6018174.0,
      "step": 2725
    },
    {
      "entropy": 5.689708185195923,
      "epoch": 2.752395360564801,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00045946833226418235,
      "loss": 5.4543,
      "mean_token_accuracy": 0.1482044279575348,
      "num_tokens": 6030049.0,
      "step": 2730
    },
    {
      "entropy": 5.750078916549683,
      "epoch": 2.7574382249117497,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004592409287542069,
      "loss": 5.505,
      "mean_token_accuracy": 0.15137905031442642,
      "num_tokens": 6040155.0,
      "step": 2735
    },
    {
      "entropy": 5.6978637218475345,
      "epoch": 2.762481089258699,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004590129526191748,
      "loss": 5.5021,
      "mean_token_accuracy": 0.15914395600557327,
      "num_tokens": 6050569.0,
      "step": 2740
    },
    {
      "entropy": 5.786010313034057,
      "epoch": 2.767523953605648,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004587844045676413,
      "loss": 5.4437,
      "mean_token_accuracy": 0.15803942680358887,
      "num_tokens": 6060871.0,
      "step": 2745
    },
    {
      "entropy": 5.7462421417236325,
      "epoch": 2.772566817952597,
      "grad_norm": 1.0,
      "learning_rate": 0.00045855528530993874,
      "loss": 5.4651,
      "mean_token_accuracy": 0.15985400527715682,
      "num_tokens": 6073045.0,
      "step": 2750
    },
    {
      "entropy": 5.768692970275879,
      "epoch": 2.777609682299546,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045832559555817525,
      "loss": 5.502,
      "mean_token_accuracy": 0.1526694893836975,
      "num_tokens": 6084798.0,
      "step": 2755
    },
    {
      "entropy": 5.760065793991089,
      "epoch": 2.7826525466464953,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045809533602623185,
      "loss": 5.4369,
      "mean_token_accuracy": 0.15314445197582244,
      "num_tokens": 6095831.0,
      "step": 2760
    },
    {
      "entropy": 5.707836771011353,
      "epoch": 2.7876954109934444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004578645074297603,
      "loss": 5.4213,
      "mean_token_accuracy": 0.1573302611708641,
      "num_tokens": 6106958.0,
      "step": 2765
    },
    {
      "entropy": 5.7063134670257565,
      "epoch": 2.792738275340393,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045763311048618144,
      "loss": 5.451,
      "mean_token_accuracy": 0.16227541416883468,
      "num_tokens": 6118165.0,
      "step": 2770
    },
    {
      "entropy": 5.692743158340454,
      "epoch": 2.797781139687342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004574011459146822,
      "loss": 5.4479,
      "mean_token_accuracy": 0.15757308900356293,
      "num_tokens": 6129297.0,
      "step": 2775
    },
    {
      "entropy": 5.75752363204956,
      "epoch": 2.8028240040342913,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045716861443621386,
      "loss": 5.4855,
      "mean_token_accuracy": 0.15124510377645492,
      "num_tokens": 6140071.0,
      "step": 2780
    },
    {
      "entropy": 5.732733631134034,
      "epoch": 2.8078668683812404,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045693551677348967,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1521101713180542,
      "num_tokens": 6150940.0,
      "step": 2785
    },
    {
      "entropy": 5.756269931793213,
      "epoch": 2.8129097327281896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004567018536509823,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1500523254275322,
      "num_tokens": 6161966.0,
      "step": 2790
    },
    {
      "entropy": 5.838177490234375,
      "epoch": 2.8179525970751387,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045646762579492245,
      "loss": 5.5243,
      "mean_token_accuracy": 0.14861284494400023,
      "num_tokens": 6174455.0,
      "step": 2795
    },
    {
      "entropy": 5.722758102416992,
      "epoch": 2.822995461422088,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045623283393329546,
      "loss": 5.4502,
      "mean_token_accuracy": 0.15393734872341155,
      "num_tokens": 6184789.0,
      "step": 2800
    },
    {
      "entropy": 5.725300741195679,
      "epoch": 2.828038325769037,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045599747879584,
      "loss": 5.5423,
      "mean_token_accuracy": 0.14950788617134095,
      "num_tokens": 6195704.0,
      "step": 2805
    },
    {
      "entropy": 5.7946454048156735,
      "epoch": 2.833081190115986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004557615611140451,
      "loss": 5.424,
      "mean_token_accuracy": 0.15766871273517608,
      "num_tokens": 6206290.0,
      "step": 2810
    },
    {
      "entropy": 5.773143482208252,
      "epoch": 2.838124054462935,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004555250816211485,
      "loss": 5.5375,
      "mean_token_accuracy": 0.1489001050591469,
      "num_tokens": 6216660.0,
      "step": 2815
    },
    {
      "entropy": 5.746375370025635,
      "epoch": 2.8431669188098843,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00045528804105213395,
      "loss": 5.4751,
      "mean_token_accuracy": 0.1593405783176422,
      "num_tokens": 6228005.0,
      "step": 2820
    },
    {
      "entropy": 5.703634405136109,
      "epoch": 2.848209783156833,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004550504401437289,
      "loss": 5.4385,
      "mean_token_accuracy": 0.15495084077119828,
      "num_tokens": 6239447.0,
      "step": 2825
    },
    {
      "entropy": 5.752251195907593,
      "epoch": 2.853252647503782,
      "grad_norm": 0.875,
      "learning_rate": 0.00045481227963440235,
      "loss": 5.5516,
      "mean_token_accuracy": 0.14331979677081108,
      "num_tokens": 6252312.0,
      "step": 2830
    },
    {
      "entropy": 5.766775178909302,
      "epoch": 2.858295511850731,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045457356026436295,
      "loss": 5.4855,
      "mean_token_accuracy": 0.15106730610132219,
      "num_tokens": 6264131.0,
      "step": 2835
    },
    {
      "entropy": 5.663240242004394,
      "epoch": 2.8633383761976803,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00045433428277555577,
      "loss": 5.4265,
      "mean_token_accuracy": 0.15811824798583984,
      "num_tokens": 6275161.0,
      "step": 2840
    },
    {
      "entropy": 5.689582061767578,
      "epoch": 2.8683812405446294,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004540944479116609,
      "loss": 5.4832,
      "mean_token_accuracy": 0.1488562524318695,
      "num_tokens": 6285655.0,
      "step": 2845
    },
    {
      "entropy": 5.768545293807984,
      "epoch": 2.8734241048915785,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004538540564180906,
      "loss": 5.572,
      "mean_token_accuracy": 0.1562221720814705,
      "num_tokens": 6296532.0,
      "step": 2850
    },
    {
      "entropy": 5.878363561630249,
      "epoch": 2.8784669692385276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004536131090419871,
      "loss": 5.5974,
      "mean_token_accuracy": 0.15142830461263657,
      "num_tokens": 6308229.0,
      "step": 2855
    },
    {
      "entropy": 5.747320890426636,
      "epoch": 2.8835098335854763,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004533716065322205,
      "loss": 5.5356,
      "mean_token_accuracy": 0.15433450639247895,
      "num_tokens": 6321257.0,
      "step": 2860
    },
    {
      "entropy": 5.785921573638916,
      "epoch": 2.8885526979324254,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004531295496393861,
      "loss": 5.4623,
      "mean_token_accuracy": 0.15531488209962846,
      "num_tokens": 6332040.0,
      "step": 2865
    },
    {
      "entropy": 5.636278247833252,
      "epoch": 2.8935955622793745,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004528869391158023,
      "loss": 5.3599,
      "mean_token_accuracy": 0.16037604510784148,
      "num_tokens": 6342925.0,
      "step": 2870
    },
    {
      "entropy": 5.764258241653442,
      "epoch": 2.8986384266263236,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00045264377571550834,
      "loss": 5.5312,
      "mean_token_accuracy": 0.15192896723747254,
      "num_tokens": 6354263.0,
      "step": 2875
    },
    {
      "entropy": 5.8196563720703125,
      "epoch": 2.9036812909732728,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004524000601942615,
      "loss": 5.5274,
      "mean_token_accuracy": 0.14990588426589965,
      "num_tokens": 6365497.0,
      "step": 2880
    },
    {
      "entropy": 5.790423011779785,
      "epoch": 2.908724155320222,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004521557933095354,
      "loss": 5.5746,
      "mean_token_accuracy": 0.1506519302725792,
      "num_tokens": 6376501.0,
      "step": 2885
    },
    {
      "entropy": 5.795026397705078,
      "epoch": 2.913767019667171,
      "grad_norm": 1.015625,
      "learning_rate": 0.000451910975820517,
      "loss": 5.5151,
      "mean_token_accuracy": 0.1521720841526985,
      "num_tokens": 6387632.0,
      "step": 2890
    },
    {
      "entropy": 5.6698846340179445,
      "epoch": 2.91880988401412,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045166560848810476,
      "loss": 5.5013,
      "mean_token_accuracy": 0.15847814083099365,
      "num_tokens": 6398813.0,
      "step": 2895
    },
    {
      "entropy": 5.837898063659668,
      "epoch": 2.9238527483610692,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045141969207490606,
      "loss": 5.4186,
      "mean_token_accuracy": 0.15635257065296174,
      "num_tokens": 6409532.0,
      "step": 2900
    },
    {
      "entropy": 5.730175876617432,
      "epoch": 2.9288956127080183,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004511732273452347,
      "loss": 5.5496,
      "mean_token_accuracy": 0.14952748268842697,
      "num_tokens": 6420755.0,
      "step": 2905
    },
    {
      "entropy": 5.820220947265625,
      "epoch": 2.9339384770549675,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004509262150651089,
      "loss": 5.5818,
      "mean_token_accuracy": 0.15502915233373643,
      "num_tokens": 6430662.0,
      "step": 2910
    },
    {
      "entropy": 5.7733540534973145,
      "epoch": 2.938981341401916,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045067865600224833,
      "loss": 5.4634,
      "mean_token_accuracy": 0.1565277934074402,
      "num_tokens": 6441230.0,
      "step": 2915
    },
    {
      "entropy": 5.7728832244873045,
      "epoch": 2.9440242057488653,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004504305509260724,
      "loss": 5.6008,
      "mean_token_accuracy": 0.1425420969724655,
      "num_tokens": 6452574.0,
      "step": 2920
    },
    {
      "entropy": 5.74591097831726,
      "epoch": 2.9490670700958144,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004501819006076974,
      "loss": 5.449,
      "mean_token_accuracy": 0.15784769207239152,
      "num_tokens": 6462562.0,
      "step": 2925
    },
    {
      "entropy": 5.704241037368774,
      "epoch": 2.9541099344427635,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044993270581993413,
      "loss": 5.4592,
      "mean_token_accuracy": 0.15327904224395753,
      "num_tokens": 6473594.0,
      "step": 2930
    },
    {
      "entropy": 5.738255357742309,
      "epoch": 2.9591527987897126,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044968296733728587,
      "loss": 5.4819,
      "mean_token_accuracy": 0.15456565022468566,
      "num_tokens": 6484129.0,
      "step": 2935
    },
    {
      "entropy": 5.773097801208496,
      "epoch": 2.9641956631366617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004494326859359456,
      "loss": 5.4635,
      "mean_token_accuracy": 0.15108635127544404,
      "num_tokens": 6495584.0,
      "step": 2940
    },
    {
      "entropy": 5.746885061264038,
      "epoch": 2.969238527483611,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044918186239379363,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1557958319783211,
      "num_tokens": 6507872.0,
      "step": 2945
    },
    {
      "entropy": 5.699970388412476,
      "epoch": 2.9742813918305595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004489304974903953,
      "loss": 5.4214,
      "mean_token_accuracy": 0.15810350328683853,
      "num_tokens": 6519172.0,
      "step": 2950
    },
    {
      "entropy": 5.6845227718353275,
      "epoch": 2.9793242561775086,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004486785920069985,
      "loss": 5.5146,
      "mean_token_accuracy": 0.15763776153326034,
      "num_tokens": 6530151.0,
      "step": 2955
    },
    {
      "entropy": 5.798524999618531,
      "epoch": 2.9843671205244577,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004484261467265313,
      "loss": 5.5,
      "mean_token_accuracy": 0.154867172986269,
      "num_tokens": 6541881.0,
      "step": 2960
    },
    {
      "entropy": 5.7200994968414305,
      "epoch": 2.989409984871407,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00044817316243359956,
      "loss": 5.5202,
      "mean_token_accuracy": 0.15170194953680038,
      "num_tokens": 6553390.0,
      "step": 2965
    },
    {
      "entropy": 5.742916917800903,
      "epoch": 2.994452849218356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044791963991448427,
      "loss": 5.3511,
      "mean_token_accuracy": 0.1625334694981575,
      "num_tokens": 6563754.0,
      "step": 2970
    },
    {
      "entropy": 5.6947300910949705,
      "epoch": 2.999495713565305,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044766557995713915,
      "loss": 5.4304,
      "mean_token_accuracy": 0.15076594054698944,
      "num_tokens": 6574124.0,
      "step": 2975
    },
    {
      "entropy": 5.7346918318006725,
      "epoch": 3.0040342914775593,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004474109833511885,
      "loss": 5.2542,
      "mean_token_accuracy": 0.16592459049489763,
      "num_tokens": 6582795.0,
      "step": 2980
    },
    {
      "entropy": 5.706083011627197,
      "epoch": 3.0090771558245084,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004471558508879244,
      "loss": 5.1793,
      "mean_token_accuracy": 0.17631951868534088,
      "num_tokens": 6593649.0,
      "step": 2985
    },
    {
      "entropy": 5.684178686141967,
      "epoch": 3.0141200201714575,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004469001833603045,
      "loss": 5.194,
      "mean_token_accuracy": 0.17615290582180024,
      "num_tokens": 6606370.0,
      "step": 2990
    },
    {
      "entropy": 5.630094337463379,
      "epoch": 3.0191628845184066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004466439815629493,
      "loss": 5.1446,
      "mean_token_accuracy": 0.17401258796453475,
      "num_tokens": 6616599.0,
      "step": 2995
    },
    {
      "entropy": 5.772274303436279,
      "epoch": 3.0242057488653553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044638724629214,
      "loss": 5.2127,
      "mean_token_accuracy": 0.16511795073747634,
      "num_tokens": 6627707.0,
      "step": 3000
    },
    {
      "epoch": 3.0242057488653553,
      "eval_entropy": 5.510967042913267,
      "eval_loss": 5.7936272621154785,
      "eval_mean_token_accuracy": 0.14498784324822536,
      "eval_num_tokens": 6627707.0,
      "eval_runtime": 2.0597,
      "eval_samples_per_second": 1523.507,
      "eval_steps_per_second": 190.803,
      "step": 3000
    },
    {
      "entropy": 5.664925575256348,
      "epoch": 3.0292486132123044,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044612997834581557,
      "loss": 5.1272,
      "mean_token_accuracy": 0.17839677929878234,
      "num_tokens": 6639001.0,
      "step": 3005
    },
    {
      "entropy": 5.683994770050049,
      "epoch": 3.0342914775592535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044587217852357086,
      "loss": 5.2086,
      "mean_token_accuracy": 0.1659701272845268,
      "num_tokens": 6649318.0,
      "step": 3010
    },
    {
      "entropy": 5.662193679809571,
      "epoch": 3.0393343419062027,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044561384762665363,
      "loss": 5.236,
      "mean_token_accuracy": 0.16719333380460738,
      "num_tokens": 6660584.0,
      "step": 3015
    },
    {
      "entropy": 5.613649082183838,
      "epoch": 3.044377206253152,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004453549864579623,
      "loss": 5.0714,
      "mean_token_accuracy": 0.18089322596788407,
      "num_tokens": 6671361.0,
      "step": 3020
    },
    {
      "entropy": 5.699713706970215,
      "epoch": 3.049420070600101,
      "grad_norm": 1.0,
      "learning_rate": 0.0004450955958220433,
      "loss": 5.3063,
      "mean_token_accuracy": 0.16489263623952866,
      "num_tokens": 6681896.0,
      "step": 3025
    },
    {
      "entropy": 5.668648290634155,
      "epoch": 3.05446293494705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044483567652508864,
      "loss": 5.1611,
      "mean_token_accuracy": 0.17029865086078644,
      "num_tokens": 6693360.0,
      "step": 3030
    },
    {
      "entropy": 5.653792142868042,
      "epoch": 3.059505799293999,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004445752293749337,
      "loss": 5.2329,
      "mean_token_accuracy": 0.16625278890132905,
      "num_tokens": 6704921.0,
      "step": 3035
    },
    {
      "entropy": 5.695637035369873,
      "epoch": 3.0645486636409482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004443142551810541,
      "loss": 5.1967,
      "mean_token_accuracy": 0.1730018064379692,
      "num_tokens": 6715659.0,
      "step": 3040
    },
    {
      "entropy": 5.588192987442016,
      "epoch": 3.069591527987897,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044405275475456365,
      "loss": 5.0988,
      "mean_token_accuracy": 0.18201044499874114,
      "num_tokens": 6726431.0,
      "step": 3045
    },
    {
      "entropy": 5.634479188919068,
      "epoch": 3.074634392334846,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004437907289082116,
      "loss": 5.1876,
      "mean_token_accuracy": 0.17533796280622482,
      "num_tokens": 6738002.0,
      "step": 3050
    },
    {
      "entropy": 5.681352806091309,
      "epoch": 3.079677256681795,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044352817845638053,
      "loss": 5.1705,
      "mean_token_accuracy": 0.1775447517633438,
      "num_tokens": 6748509.0,
      "step": 3055
    },
    {
      "entropy": 5.671092796325683,
      "epoch": 3.0847201210287443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004432651042150833,
      "loss": 5.214,
      "mean_token_accuracy": 0.16949847042560579,
      "num_tokens": 6759258.0,
      "step": 3060
    },
    {
      "entropy": 5.6300904750823975,
      "epoch": 3.0897629853756934,
      "grad_norm": 1.125,
      "learning_rate": 0.00044300150700196054,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17419342398643495,
      "num_tokens": 6771263.0,
      "step": 3065
    },
    {
      "entropy": 5.660618734359741,
      "epoch": 3.0948058497226425,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004427373876362786,
      "loss": 5.1866,
      "mean_token_accuracy": 0.17338812053203584,
      "num_tokens": 6779965.0,
      "step": 3070
    },
    {
      "entropy": 5.6531593799591064,
      "epoch": 3.0998487140695916,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004424727469389265,
      "loss": 5.1603,
      "mean_token_accuracy": 0.1707604169845581,
      "num_tokens": 6791521.0,
      "step": 3075
    },
    {
      "entropy": 5.688219833374023,
      "epoch": 3.1048915784165407,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004422075857324137,
      "loss": 5.2287,
      "mean_token_accuracy": 0.16988472938537597,
      "num_tokens": 6803537.0,
      "step": 3080
    },
    {
      "entropy": 5.687441444396972,
      "epoch": 3.10993444276349,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004419419048408673,
      "loss": 5.2478,
      "mean_token_accuracy": 0.1688054010272026,
      "num_tokens": 6813811.0,
      "step": 3085
    },
    {
      "entropy": 5.645955801010132,
      "epoch": 3.1149773071104385,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004416757050900297,
      "loss": 5.15,
      "mean_token_accuracy": 0.17355225533246993,
      "num_tokens": 6825422.0,
      "step": 3090
    },
    {
      "entropy": 5.76006293296814,
      "epoch": 3.1200201714573876,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004414089873072557,
      "loss": 5.2999,
      "mean_token_accuracy": 0.1641416370868683,
      "num_tokens": 6837086.0,
      "step": 3095
    },
    {
      "entropy": 5.633365058898926,
      "epoch": 3.1250630358043368,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004411417523215105,
      "loss": 5.1372,
      "mean_token_accuracy": 0.17556397914886473,
      "num_tokens": 6847195.0,
      "step": 3100
    },
    {
      "entropy": 5.606680631637573,
      "epoch": 3.130105900151286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004408740009633665,
      "loss": 5.1515,
      "mean_token_accuracy": 0.175725756585598,
      "num_tokens": 6857492.0,
      "step": 3105
    },
    {
      "entropy": 5.6575929641723635,
      "epoch": 3.135148764498235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044060573406500117,
      "loss": 5.3348,
      "mean_token_accuracy": 0.16419571042060851,
      "num_tokens": 6869190.0,
      "step": 3110
    },
    {
      "entropy": 5.669050645828247,
      "epoch": 3.140191628845184,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004403369524601941,
      "loss": 5.1277,
      "mean_token_accuracy": 0.17212760746479033,
      "num_tokens": 6880728.0,
      "step": 3115
    },
    {
      "entropy": 5.5693145275115965,
      "epoch": 3.145234493192133,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004400676569843249,
      "loss": 5.1683,
      "mean_token_accuracy": 0.1651963084936142,
      "num_tokens": 6892043.0,
      "step": 3120
    },
    {
      "entropy": 5.636732482910157,
      "epoch": 3.1502773575390823,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043979784847436994,
      "loss": 5.1965,
      "mean_token_accuracy": 0.17691275626420974,
      "num_tokens": 6903913.0,
      "step": 3125
    },
    {
      "entropy": 5.5700384140014645,
      "epoch": 3.1553202218860315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004395275277689004,
      "loss": 5.1594,
      "mean_token_accuracy": 0.17174604237079621,
      "num_tokens": 6913951.0,
      "step": 3130
    },
    {
      "entropy": 5.58952841758728,
      "epoch": 3.16036308623298,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004392566957080793,
      "loss": 5.1743,
      "mean_token_accuracy": 0.1700451597571373,
      "num_tokens": 6924762.0,
      "step": 3135
    },
    {
      "entropy": 5.605324411392212,
      "epoch": 3.1654059505799292,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043898535313365884,
      "loss": 5.1602,
      "mean_token_accuracy": 0.17511910647153855,
      "num_tokens": 6935845.0,
      "step": 3140
    },
    {
      "entropy": 5.612947082519531,
      "epoch": 3.1704488149268784,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000438713500888978,
      "loss": 5.1926,
      "mean_token_accuracy": 0.17189945429563522,
      "num_tokens": 6947338.0,
      "step": 3145
    },
    {
      "entropy": 5.678471946716309,
      "epoch": 3.1754916792738275,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004384411398189599,
      "loss": 5.2251,
      "mean_token_accuracy": 0.17044371515512466,
      "num_tokens": 6958084.0,
      "step": 3150
    },
    {
      "entropy": 5.631912612915039,
      "epoch": 3.1805345436207766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000438168270770109,
      "loss": 5.1531,
      "mean_token_accuracy": 0.17372213900089264,
      "num_tokens": 6968063.0,
      "step": 3155
    },
    {
      "entropy": 5.625943660736084,
      "epoch": 3.1855774079677257,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004378948945905085,
      "loss": 5.158,
      "mean_token_accuracy": 0.1735137492418289,
      "num_tokens": 6978424.0,
      "step": 3160
    },
    {
      "entropy": 5.531322908401489,
      "epoch": 3.190620272314675,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00043762101212981805,
      "loss": 5.1474,
      "mean_token_accuracy": 0.17346943616867067,
      "num_tokens": 6991067.0,
      "step": 3165
    },
    {
      "entropy": 5.581675863265991,
      "epoch": 3.195663136661624,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043734662423927054,
      "loss": 5.2304,
      "mean_token_accuracy": 0.17353299409151077,
      "num_tokens": 7000642.0,
      "step": 3170
    },
    {
      "entropy": 5.732257843017578,
      "epoch": 3.200706001008573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004370717317716699,
      "loss": 5.2629,
      "mean_token_accuracy": 0.17225463092327117,
      "num_tokens": 7011305.0,
      "step": 3175
    },
    {
      "entropy": 5.615920305252075,
      "epoch": 3.205748865355522,
      "grad_norm": 1.125,
      "learning_rate": 0.0004367963355813882,
      "loss": 5.2052,
      "mean_token_accuracy": 0.164471235871315,
      "num_tokens": 7021452.0,
      "step": 3180
    },
    {
      "entropy": 5.636391925811767,
      "epoch": 3.210791729702471,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043652043652436323,
      "loss": 5.2752,
      "mean_token_accuracy": 0.16358589082956315,
      "num_tokens": 7033189.0,
      "step": 3185
    },
    {
      "entropy": 5.669807577133179,
      "epoch": 3.21583459404942,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004362440354580956,
      "loss": 5.2261,
      "mean_token_accuracy": 0.17085227221250535,
      "num_tokens": 7044153.0,
      "step": 3190
    },
    {
      "entropy": 5.646782636642456,
      "epoch": 3.220877458396369,
      "grad_norm": 1.125,
      "learning_rate": 0.0004359671332416462,
      "loss": 5.2676,
      "mean_token_accuracy": 0.1627621978521347,
      "num_tokens": 7056679.0,
      "step": 3195
    },
    {
      "entropy": 5.641334152221679,
      "epoch": 3.225920322743318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004356897307356335,
      "loss": 5.1341,
      "mean_token_accuracy": 0.17200587689876556,
      "num_tokens": 7067428.0,
      "step": 3200
    },
    {
      "entropy": 5.642390489578247,
      "epoch": 3.2309631870902673,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004354118288022309,
      "loss": 5.2571,
      "mean_token_accuracy": 0.1659141808748245,
      "num_tokens": 7079196.0,
      "step": 3205
    },
    {
      "entropy": 5.6666196346282955,
      "epoch": 3.2360060514372164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043513342830516405,
      "loss": 5.1482,
      "mean_token_accuracy": 0.1730696216225624,
      "num_tokens": 7088936.0,
      "step": 3210
    },
    {
      "entropy": 5.607312726974487,
      "epoch": 3.2410489157841655,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000434854530109708,
      "loss": 5.2409,
      "mean_token_accuracy": 0.1737443044781685,
      "num_tokens": 7101504.0,
      "step": 3215
    },
    {
      "entropy": 5.657965803146363,
      "epoch": 3.2460917801311147,
      "grad_norm": 0.984375,
      "learning_rate": 0.00043457513508268487,
      "loss": 5.2169,
      "mean_token_accuracy": 0.16955721378326416,
      "num_tokens": 7114121.0,
      "step": 3220
    },
    {
      "entropy": 5.565788412094117,
      "epoch": 3.2511346444780633,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043429524409246083,
      "loss": 5.12,
      "mean_token_accuracy": 0.17057781666517258,
      "num_tokens": 7123321.0,
      "step": 3225
    },
    {
      "entropy": 5.595669555664062,
      "epoch": 3.2561775088250124,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004340148580089436,
      "loss": 5.2029,
      "mean_token_accuracy": 0.171379691362381,
      "num_tokens": 7134145.0,
      "step": 3230
    },
    {
      "entropy": 5.720705318450928,
      "epoch": 3.2612203731719616,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004337339777035794,
      "loss": 5.244,
      "mean_token_accuracy": 0.16833797991275787,
      "num_tokens": 7144727.0,
      "step": 3235
    },
    {
      "entropy": 5.523499059677124,
      "epoch": 3.2662632375189107,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000433452604049351,
      "loss": 5.1393,
      "mean_token_accuracy": 0.17887951135635377,
      "num_tokens": 7155531.0,
      "step": 3240
    },
    {
      "entropy": 5.575527286529541,
      "epoch": 3.27130610186586,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00043317073792077394,
      "loss": 5.1808,
      "mean_token_accuracy": 0.17313862442970276,
      "num_tokens": 7167234.0,
      "step": 3245
    },
    {
      "entropy": 5.6531819820404055,
      "epoch": 3.276348966212809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004328883801938947,
      "loss": 5.2603,
      "mean_token_accuracy": 0.1627066120505333,
      "num_tokens": 7178556.0,
      "step": 3250
    },
    {
      "entropy": 5.509610319137574,
      "epoch": 3.281391830559758,
      "grad_norm": 1.125,
      "learning_rate": 0.00043260553174628776,
      "loss": 5.1176,
      "mean_token_accuracy": 0.17404565513134002,
      "num_tokens": 7189074.0,
      "step": 3255
    },
    {
      "entropy": 5.6646552085876465,
      "epoch": 3.286434694906707,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004323221934570524,
      "loss": 5.2639,
      "mean_token_accuracy": 0.17479537129402162,
      "num_tokens": 7200567.0,
      "step": 3260
    },
    {
      "entropy": 5.68840389251709,
      "epoch": 3.2914775592536563,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043203836620681074,
      "loss": 5.3077,
      "mean_token_accuracy": 0.16148419082164764,
      "num_tokens": 7211012.0,
      "step": 3265
    },
    {
      "entropy": 5.559213352203369,
      "epoch": 3.2965204236006054,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043175405087770435,
      "loss": 5.1787,
      "mean_token_accuracy": 0.17627494931221008,
      "num_tokens": 7222271.0,
      "step": 3270
    },
    {
      "entropy": 5.624978828430176,
      "epoch": 3.301563287947554,
      "grad_norm": 0.90625,
      "learning_rate": 0.00043146924835339186,
      "loss": 5.1468,
      "mean_token_accuracy": 0.17507985532283782,
      "num_tokens": 7235067.0,
      "step": 3275
    },
    {
      "entropy": 5.580532646179199,
      "epoch": 3.306606152294503,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043118395951904605,
      "loss": 5.2285,
      "mean_token_accuracy": 0.16788760274648667,
      "num_tokens": 7246159.0,
      "step": 3280
    },
    {
      "entropy": 5.742359924316406,
      "epoch": 3.3116490166414523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004308981852613512,
      "loss": 5.2461,
      "mean_token_accuracy": 0.16732081025838852,
      "num_tokens": 7255944.0,
      "step": 3285
    },
    {
      "entropy": 5.638572692871094,
      "epoch": 3.3166918809884014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004306119264685003,
      "loss": 5.1807,
      "mean_token_accuracy": 0.17448685467243194,
      "num_tokens": 7267273.0,
      "step": 3290
    },
    {
      "entropy": 5.572154808044433,
      "epoch": 3.3217347453353505,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004303251840301923,
      "loss": 5.1861,
      "mean_token_accuracy": 0.1709363728761673,
      "num_tokens": 7277622.0,
      "step": 3295
    },
    {
      "entropy": 5.584944152832032,
      "epoch": 3.3267776096822996,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00043003795883762933,
      "loss": 5.2352,
      "mean_token_accuracy": 0.1699904501438141,
      "num_tokens": 7288851.0,
      "step": 3300
    },
    {
      "entropy": 5.687366771697998,
      "epoch": 3.3318204740292487,
      "grad_norm": 0.96875,
      "learning_rate": 0.00042975025178351386,
      "loss": 5.2894,
      "mean_token_accuracy": 0.16496133506298066,
      "num_tokens": 7302175.0,
      "step": 3305
    },
    {
      "entropy": 5.6211694240570065,
      "epoch": 3.336863338376198,
      "grad_norm": 1.1875,
      "learning_rate": 0.000429462063762046,
      "loss": 5.1503,
      "mean_token_accuracy": 0.17872511297464372,
      "num_tokens": 7312693.0,
      "step": 3310
    },
    {
      "entropy": 5.544012928009034,
      "epoch": 3.3419062027231465,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042917339566892095,
      "loss": 5.1869,
      "mean_token_accuracy": 0.16638617664575578,
      "num_tokens": 7323790.0,
      "step": 3315
    },
    {
      "entropy": 5.581661939620972,
      "epoch": 3.3469490670700957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004288842484013257,
      "loss": 5.1713,
      "mean_token_accuracy": 0.17432422935962677,
      "num_tokens": 7333797.0,
      "step": 3320
    },
    {
      "entropy": 5.596593189239502,
      "epoch": 3.3519919314170448,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004285946228579365,
      "loss": 5.2103,
      "mean_token_accuracy": 0.17532770931720734,
      "num_tokens": 7344674.0,
      "step": 3325
    },
    {
      "entropy": 5.618811130523682,
      "epoch": 3.357034795763994,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004283045199389164,
      "loss": 5.2323,
      "mean_token_accuracy": 0.1707398474216461,
      "num_tokens": 7355978.0,
      "step": 3330
    },
    {
      "entropy": 5.717554807662964,
      "epoch": 3.362077660110943,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042801394054591197,
      "loss": 5.3068,
      "mean_token_accuracy": 0.1689215824007988,
      "num_tokens": 7367054.0,
      "step": 3335
    },
    {
      "entropy": 5.671964311599732,
      "epoch": 3.367120524457892,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042772288558205057,
      "loss": 5.2725,
      "mean_token_accuracy": 0.16591570675373077,
      "num_tokens": 7379588.0,
      "step": 3340
    },
    {
      "entropy": 5.568938875198365,
      "epoch": 3.3721633888048412,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004274313559519377,
      "loss": 5.237,
      "mean_token_accuracy": 0.16997685879468918,
      "num_tokens": 7392411.0,
      "step": 3345
    },
    {
      "entropy": 5.575924348831177,
      "epoch": 3.3772062531517903,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004271393525616544,
      "loss": 5.1302,
      "mean_token_accuracy": 0.17118777483701705,
      "num_tokens": 7402801.0,
      "step": 3350
    },
    {
      "entropy": 5.5714741230010985,
      "epoch": 3.3822491174987395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042684687631875377,
      "loss": 5.2075,
      "mean_token_accuracy": 0.1664086252450943,
      "num_tokens": 7413084.0,
      "step": 3355
    },
    {
      "entropy": 5.610323286056518,
      "epoch": 3.3872919818456886,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042655392813225864,
      "loss": 5.1504,
      "mean_token_accuracy": 0.1750144362449646,
      "num_tokens": 7424004.0,
      "step": 3360
    },
    {
      "entropy": 5.5400786876678465,
      "epoch": 3.3923348461926373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042626050891265884,
      "loss": 5.1649,
      "mean_token_accuracy": 0.17817920744419097,
      "num_tokens": 7434461.0,
      "step": 3365
    },
    {
      "entropy": 5.652910804748535,
      "epoch": 3.3973777105395864,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004259666195719081,
      "loss": 5.2382,
      "mean_token_accuracy": 0.1693527951836586,
      "num_tokens": 7445552.0,
      "step": 3370
    },
    {
      "entropy": 5.591069173812866,
      "epoch": 3.4024205748865355,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042567226102342134,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17021757811307908,
      "num_tokens": 7457073.0,
      "step": 3375
    },
    {
      "entropy": 5.589842796325684,
      "epoch": 3.4074634392334846,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004253774341820715,
      "loss": 5.2095,
      "mean_token_accuracy": 0.17243159413337708,
      "num_tokens": 7468197.0,
      "step": 3380
    },
    {
      "entropy": 5.647180271148682,
      "epoch": 3.4125063035804337,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004250821399641874,
      "loss": 5.1662,
      "mean_token_accuracy": 0.17413502484560012,
      "num_tokens": 7480751.0,
      "step": 3385
    },
    {
      "entropy": 5.564364671707153,
      "epoch": 3.417549167927383,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004247863792875502,
      "loss": 5.187,
      "mean_token_accuracy": 0.17214548587799072,
      "num_tokens": 7492259.0,
      "step": 3390
    },
    {
      "entropy": 5.6444212913513185,
      "epoch": 3.422592032274332,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042449015307139076,
      "loss": 5.2736,
      "mean_token_accuracy": 0.162250117957592,
      "num_tokens": 7504317.0,
      "step": 3395
    },
    {
      "entropy": 5.583958148956299,
      "epoch": 3.427634896621281,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042419346223638725,
      "loss": 5.1769,
      "mean_token_accuracy": 0.1668773114681244,
      "num_tokens": 7516039.0,
      "step": 3400
    },
    {
      "entropy": 5.608598232269287,
      "epoch": 3.4326777609682297,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004238963077046616,
      "loss": 5.2359,
      "mean_token_accuracy": 0.168964421749115,
      "num_tokens": 7526095.0,
      "step": 3405
    },
    {
      "entropy": 5.540010595321656,
      "epoch": 3.437720625315179,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004235986903997767,
      "loss": 5.194,
      "mean_token_accuracy": 0.16597900539636612,
      "num_tokens": 7537320.0,
      "step": 3410
    },
    {
      "entropy": 5.6050502300262455,
      "epoch": 3.442763489662128,
      "grad_norm": 1.125,
      "learning_rate": 0.00042330061124673437,
      "loss": 5.2559,
      "mean_token_accuracy": 0.17067249417304992,
      "num_tokens": 7548100.0,
      "step": 3415
    },
    {
      "entropy": 5.66834077835083,
      "epoch": 3.447806354009077,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042300207117197116,
      "loss": 5.2646,
      "mean_token_accuracy": 0.16529277414083482,
      "num_tokens": 7560483.0,
      "step": 3420
    },
    {
      "entropy": 5.605900573730469,
      "epoch": 3.452849218356026,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042270307110335657,
      "loss": 5.1734,
      "mean_token_accuracy": 0.17358876764774323,
      "num_tokens": 7571256.0,
      "step": 3425
    },
    {
      "entropy": 5.636985492706299,
      "epoch": 3.4578920827029753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004224036119701899,
      "loss": 5.1679,
      "mean_token_accuracy": 0.1722208335995674,
      "num_tokens": 7581716.0,
      "step": 3430
    },
    {
      "entropy": 5.608111047744751,
      "epoch": 3.4629349470499244,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042210369470319694,
      "loss": 5.1991,
      "mean_token_accuracy": 0.17423703968524934,
      "num_tokens": 7591932.0,
      "step": 3435
    },
    {
      "entropy": 5.5758538246154785,
      "epoch": 3.4679778113968736,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004218033202345274,
      "loss": 5.1868,
      "mean_token_accuracy": 0.17040564715862275,
      "num_tokens": 7602842.0,
      "step": 3440
    },
    {
      "entropy": 5.565664863586425,
      "epoch": 3.4730206757438227,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004215024894977522,
      "loss": 5.1703,
      "mean_token_accuracy": 0.16753354370594026,
      "num_tokens": 7614075.0,
      "step": 3445
    },
    {
      "entropy": 5.679861736297608,
      "epoch": 3.478063540090772,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004212012034278601,
      "loss": 5.3377,
      "mean_token_accuracy": 0.16588482558727263,
      "num_tokens": 7625642.0,
      "step": 3450
    },
    {
      "entropy": 5.650643348693848,
      "epoch": 3.4831064044377205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042089946296125513,
      "loss": 5.1984,
      "mean_token_accuracy": 0.1726632907986641,
      "num_tokens": 7636447.0,
      "step": 3455
    },
    {
      "entropy": 5.525870323181152,
      "epoch": 3.4881492687846696,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042059726903575373,
      "loss": 5.1295,
      "mean_token_accuracy": 0.17201301008462905,
      "num_tokens": 7648684.0,
      "step": 3460
    },
    {
      "entropy": 5.521808719635009,
      "epoch": 3.4931921331316187,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004202946225905815,
      "loss": 5.1955,
      "mean_token_accuracy": 0.17038234025239946,
      "num_tokens": 7660864.0,
      "step": 3465
    },
    {
      "entropy": 5.575794696807861,
      "epoch": 3.498234997478568,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004199915245663706,
      "loss": 5.2251,
      "mean_token_accuracy": 0.1730599045753479,
      "num_tokens": 7671939.0,
      "step": 3470
    },
    {
      "entropy": 5.714139986038208,
      "epoch": 3.503277861825517,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041968797590515673,
      "loss": 5.231,
      "mean_token_accuracy": 0.17252027094364167,
      "num_tokens": 7681536.0,
      "step": 3475
    },
    {
      "entropy": 5.6585310935974125,
      "epoch": 3.508320726172466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004193839775503761,
      "loss": 5.1757,
      "mean_token_accuracy": 0.17951995730400086,
      "num_tokens": 7693674.0,
      "step": 3480
    },
    {
      "entropy": 5.530286884307861,
      "epoch": 3.513363590519415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041907953044686253,
      "loss": 5.2093,
      "mean_token_accuracy": 0.1805584505200386,
      "num_tokens": 7704533.0,
      "step": 3485
    },
    {
      "entropy": 5.563656044006348,
      "epoch": 3.518406454866364,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004187746355408448,
      "loss": 5.2544,
      "mean_token_accuracy": 0.17241516709327698,
      "num_tokens": 7715448.0,
      "step": 3490
    },
    {
      "entropy": 5.632718753814697,
      "epoch": 3.523449319213313,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004184692937799431,
      "loss": 5.2317,
      "mean_token_accuracy": 0.1686718389391899,
      "num_tokens": 7726808.0,
      "step": 3495
    },
    {
      "entropy": 5.780042743682861,
      "epoch": 3.528492183560262,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041816350611316693,
      "loss": 5.2301,
      "mean_token_accuracy": 0.16806208491325378,
      "num_tokens": 7737899.0,
      "step": 3500
    },
    {
      "epoch": 3.528492183560262,
      "eval_entropy": 5.465983019530318,
      "eval_loss": 5.754375457763672,
      "eval_mean_token_accuracy": 0.14750550607688553,
      "eval_num_tokens": 7737899.0,
      "eval_runtime": 2.0857,
      "eval_samples_per_second": 1504.498,
      "eval_steps_per_second": 188.422,
      "step": 3500
    },
    {
      "entropy": 5.567254877090454,
      "epoch": 3.533535047907211,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00041785727349091117,
      "loss": 5.1732,
      "mean_token_accuracy": 0.17707798033952712,
      "num_tokens": 7749268.0,
      "step": 3505
    },
    {
      "entropy": 5.587961959838867,
      "epoch": 3.5385779122541603,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000417550596864954,
      "loss": 5.2433,
      "mean_token_accuracy": 0.16885677278041838,
      "num_tokens": 7760312.0,
      "step": 3510
    },
    {
      "entropy": 5.626178169250489,
      "epoch": 3.5436207766011094,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041724347718845344,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1722480148077011,
      "num_tokens": 7771130.0,
      "step": 3515
    },
    {
      "entropy": 5.656566190719604,
      "epoch": 3.5486636409480585,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041693591541594455,
      "loss": 5.2153,
      "mean_token_accuracy": 0.16637210100889205,
      "num_tokens": 7783068.0,
      "step": 3520
    },
    {
      "entropy": 5.5883924007415775,
      "epoch": 3.5537065052950076,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004166279125033363,
      "loss": 5.3053,
      "mean_token_accuracy": 0.16569079011678695,
      "num_tokens": 7794314.0,
      "step": 3525
    },
    {
      "entropy": 5.681946277618408,
      "epoch": 3.5587493696419568,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041631946940790893,
      "loss": 5.2567,
      "mean_token_accuracy": 0.1711440071463585,
      "num_tokens": 7806888.0,
      "step": 3530
    },
    {
      "entropy": 5.549884796142578,
      "epoch": 3.563792233988906,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041601058708831077,
      "loss": 5.2437,
      "mean_token_accuracy": 0.1701399266719818,
      "num_tokens": 7818690.0,
      "step": 3535
    },
    {
      "entropy": 5.603638315200806,
      "epoch": 3.568835098335855,
      "grad_norm": 1.140625,
      "learning_rate": 0.000415701266504555,
      "loss": 5.1724,
      "mean_token_accuracy": 0.16600555181503296,
      "num_tokens": 7829321.0,
      "step": 3540
    },
    {
      "entropy": 5.597897672653199,
      "epoch": 3.5738779626828037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004153915086180173,
      "loss": 5.2272,
      "mean_token_accuracy": 0.17103207409381865,
      "num_tokens": 7840223.0,
      "step": 3545
    },
    {
      "entropy": 5.553087282180786,
      "epoch": 3.578920827029753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004150813143914323,
      "loss": 5.2207,
      "mean_token_accuracy": 0.16141388714313507,
      "num_tokens": 7850999.0,
      "step": 3550
    },
    {
      "entropy": 5.659564781188965,
      "epoch": 3.583963691376702,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004147706847888908,
      "loss": 5.2823,
      "mean_token_accuracy": 0.16547922492027284,
      "num_tokens": 7861935.0,
      "step": 3555
    },
    {
      "entropy": 5.57454023361206,
      "epoch": 3.589006555723651,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004144596207758367,
      "loss": 5.2568,
      "mean_token_accuracy": 0.1736156165599823,
      "num_tokens": 7872960.0,
      "step": 3560
    },
    {
      "entropy": 5.618500232696533,
      "epoch": 3.5940494200706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004141481233190642,
      "loss": 5.3037,
      "mean_token_accuracy": 0.16824372857809067,
      "num_tokens": 7883881.0,
      "step": 3565
    },
    {
      "entropy": 5.656716585159302,
      "epoch": 3.5990922844175492,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004138361933867145,
      "loss": 5.293,
      "mean_token_accuracy": 0.16565231382846832,
      "num_tokens": 7895106.0,
      "step": 3570
    },
    {
      "entropy": 5.587721586227417,
      "epoch": 3.6041351487644984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004135238319482731,
      "loss": 5.1767,
      "mean_token_accuracy": 0.17160560935735703,
      "num_tokens": 7906020.0,
      "step": 3575
    },
    {
      "entropy": 5.560271549224853,
      "epoch": 3.609178013111447,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004132110399745666,
      "loss": 5.2761,
      "mean_token_accuracy": 0.16673479825258256,
      "num_tokens": 7918959.0,
      "step": 3580
    },
    {
      "entropy": 5.706091690063476,
      "epoch": 3.614220877458396,
      "grad_norm": 1.125,
      "learning_rate": 0.0004128978184377596,
      "loss": 5.2143,
      "mean_token_accuracy": 0.17027741372585298,
      "num_tokens": 7929804.0,
      "step": 3585
    },
    {
      "entropy": 5.572180938720703,
      "epoch": 3.6192637418053453,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004125841683113518,
      "loss": 5.2258,
      "mean_token_accuracy": 0.16989876478910446,
      "num_tokens": 7941558.0,
      "step": 3590
    },
    {
      "entropy": 5.539093637466431,
      "epoch": 3.6243066061522944,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041227009057017523,
      "loss": 5.2533,
      "mean_token_accuracy": 0.16980014890432357,
      "num_tokens": 7951737.0,
      "step": 3595
    },
    {
      "entropy": 5.530291366577148,
      "epoch": 3.6293494704992435,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004119555861903906,
      "loss": 5.1773,
      "mean_token_accuracy": 0.18252773135900496,
      "num_tokens": 7963164.0,
      "step": 3600
    },
    {
      "entropy": 5.636479663848877,
      "epoch": 3.6343923348461926,
      "grad_norm": 1.25,
      "learning_rate": 0.0004116406561494847,
      "loss": 5.2281,
      "mean_token_accuracy": 0.16936488151550294,
      "num_tokens": 7973275.0,
      "step": 3605
    },
    {
      "entropy": 5.5932845115661625,
      "epoch": 3.6394351991931417,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041132530142626756,
      "loss": 5.2041,
      "mean_token_accuracy": 0.17626294046640395,
      "num_tokens": 7984568.0,
      "step": 3610
    },
    {
      "entropy": 5.605865240097046,
      "epoch": 3.644478063540091,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041100952300086884,
      "loss": 5.2184,
      "mean_token_accuracy": 0.16594548374414445,
      "num_tokens": 7994263.0,
      "step": 3615
    },
    {
      "entropy": 5.621344995498657,
      "epoch": 3.64952092788704,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041069332185473514,
      "loss": 5.2053,
      "mean_token_accuracy": 0.1700098693370819,
      "num_tokens": 8004881.0,
      "step": 3620
    },
    {
      "entropy": 5.568978357315063,
      "epoch": 3.654563792233989,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000410376698970627,
      "loss": 5.2018,
      "mean_token_accuracy": 0.17287467867136003,
      "num_tokens": 8015955.0,
      "step": 3625
    },
    {
      "entropy": 5.571693420410156,
      "epoch": 3.659606656580938,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041005965533261554,
      "loss": 5.197,
      "mean_token_accuracy": 0.1720926970243454,
      "num_tokens": 8027010.0,
      "step": 3630
    },
    {
      "entropy": 5.534357213973999,
      "epoch": 3.6646495209278873,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040974219192607973,
      "loss": 5.1549,
      "mean_token_accuracy": 0.17025922983884811,
      "num_tokens": 8038564.0,
      "step": 3635
    },
    {
      "entropy": 5.5699748516082765,
      "epoch": 3.669692385274836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040942430973770323,
      "loss": 5.2184,
      "mean_token_accuracy": 0.17340129613876343,
      "num_tokens": 8049975.0,
      "step": 3640
    },
    {
      "entropy": 5.624978351593017,
      "epoch": 3.674735249621785,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004091060097554712,
      "loss": 5.2384,
      "mean_token_accuracy": 0.1729784280061722,
      "num_tokens": 8061042.0,
      "step": 3645
    },
    {
      "entropy": 5.620555353164673,
      "epoch": 3.679778113968734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004087872929686673,
      "loss": 5.1644,
      "mean_token_accuracy": 0.17539747357368468,
      "num_tokens": 8071777.0,
      "step": 3650
    },
    {
      "entropy": 5.501795244216919,
      "epoch": 3.6848209783156833,
      "grad_norm": 0.984375,
      "learning_rate": 0.00040846816036787056,
      "loss": 5.251,
      "mean_token_accuracy": 0.16808881163597106,
      "num_tokens": 8082387.0,
      "step": 3655
    },
    {
      "entropy": 5.672415685653687,
      "epoch": 3.6898638426626325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004081486129449527,
      "loss": 5.1772,
      "mean_token_accuracy": 0.17548073828220367,
      "num_tokens": 8091818.0,
      "step": 3660
    },
    {
      "entropy": 5.661386299133301,
      "epoch": 3.6949067070095816,
      "grad_norm": 0.953125,
      "learning_rate": 0.00040782865169307423,
      "loss": 5.2208,
      "mean_token_accuracy": 0.17344626188278198,
      "num_tokens": 8103377.0,
      "step": 3665
    },
    {
      "entropy": 5.564823722839355,
      "epoch": 3.6999495713565302,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004075082776066822,
      "loss": 5.2655,
      "mean_token_accuracy": 0.16864969134330748,
      "num_tokens": 8114723.0,
      "step": 3670
    },
    {
      "entropy": 5.5291748046875,
      "epoch": 3.7049924357034794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004071874916815068,
      "loss": 5.1209,
      "mean_token_accuracy": 0.17161226719617845,
      "num_tokens": 8124530.0,
      "step": 3675
    },
    {
      "entropy": 5.508653402328491,
      "epoch": 3.7100353000504285,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040686629491455787,
      "loss": 5.2047,
      "mean_token_accuracy": 0.17398662567138673,
      "num_tokens": 8134894.0,
      "step": 3680
    },
    {
      "entropy": 5.566262292861938,
      "epoch": 3.7150781643973776,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040654468830412247,
      "loss": 5.1686,
      "mean_token_accuracy": 0.17678223699331283,
      "num_tokens": 8143499.0,
      "step": 3685
    },
    {
      "entropy": 5.540221309661865,
      "epoch": 3.7201210287443267,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040622267284976134,
      "loss": 5.2061,
      "mean_token_accuracy": 0.167320916056633,
      "num_tokens": 8154556.0,
      "step": 3690
    },
    {
      "entropy": 5.57552285194397,
      "epoch": 3.725163893091276,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040590024955230596,
      "loss": 5.3059,
      "mean_token_accuracy": 0.16665130108594894,
      "num_tokens": 8166047.0,
      "step": 3695
    },
    {
      "entropy": 5.60207314491272,
      "epoch": 3.730206757438225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004055774194138554,
      "loss": 5.2199,
      "mean_token_accuracy": 0.1716768190264702,
      "num_tokens": 8177068.0,
      "step": 3700
    },
    {
      "entropy": 5.573896265029907,
      "epoch": 3.735249621785174,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004052541834377731,
      "loss": 5.1976,
      "mean_token_accuracy": 0.17590048164129257,
      "num_tokens": 8187316.0,
      "step": 3705
    },
    {
      "entropy": 5.628233432769775,
      "epoch": 3.740292486132123,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040493054262868397,
      "loss": 5.3263,
      "mean_token_accuracy": 0.16767822355031967,
      "num_tokens": 8198642.0,
      "step": 3710
    },
    {
      "entropy": 5.578885126113891,
      "epoch": 3.7453353504790723,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040460649799247116,
      "loss": 5.1745,
      "mean_token_accuracy": 0.17076120525598526,
      "num_tokens": 8208476.0,
      "step": 3715
    },
    {
      "entropy": 5.568337678909302,
      "epoch": 3.7503782148260214,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004042820505362728,
      "loss": 5.2521,
      "mean_token_accuracy": 0.1745902717113495,
      "num_tokens": 8219590.0,
      "step": 3720
    },
    {
      "entropy": 5.541711616516113,
      "epoch": 3.7554210791729705,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004039572012684792,
      "loss": 5.1665,
      "mean_token_accuracy": 0.1824551299214363,
      "num_tokens": 8230707.0,
      "step": 3725
    },
    {
      "entropy": 5.612268257141113,
      "epoch": 3.760463943519919,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00040363195119872916,
      "loss": 5.2237,
      "mean_token_accuracy": 0.16631621420383452,
      "num_tokens": 8241774.0,
      "step": 3730
    },
    {
      "entropy": 5.667590618133545,
      "epoch": 3.7655068078668683,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040330630133790757,
      "loss": 5.3393,
      "mean_token_accuracy": 0.16305677890777587,
      "num_tokens": 8254314.0,
      "step": 3735
    },
    {
      "entropy": 5.592640209197998,
      "epoch": 3.7705496722138174,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040298025269814165,
      "loss": 5.1922,
      "mean_token_accuracy": 0.16968763321638108,
      "num_tokens": 8265650.0,
      "step": 3740
    },
    {
      "entropy": 5.554779815673828,
      "epoch": 3.7755925365607665,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004026538062927981,
      "loss": 5.1541,
      "mean_token_accuracy": 0.1729245662689209,
      "num_tokens": 8276808.0,
      "step": 3745
    },
    {
      "entropy": 5.484794569015503,
      "epoch": 3.7806354009077157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040232696313647985,
      "loss": 5.1158,
      "mean_token_accuracy": 0.17406528890132905,
      "num_tokens": 8287867.0,
      "step": 3750
    },
    {
      "entropy": 5.57227258682251,
      "epoch": 3.7856782652546648,
      "grad_norm": 1.125,
      "learning_rate": 0.0004019997242450231,
      "loss": 5.1586,
      "mean_token_accuracy": 0.17769554257392883,
      "num_tokens": 8298707.0,
      "step": 3755
    },
    {
      "entropy": 5.5232086181640625,
      "epoch": 3.7907211296016134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004016720906354937,
      "loss": 5.1196,
      "mean_token_accuracy": 0.1758482873439789,
      "num_tokens": 8309363.0,
      "step": 3760
    },
    {
      "entropy": 5.630809116363525,
      "epoch": 3.7957639939485626,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004013440633261845,
      "loss": 5.2763,
      "mean_token_accuracy": 0.1665689617395401,
      "num_tokens": 8319376.0,
      "step": 3765
    },
    {
      "entropy": 5.5739246845245365,
      "epoch": 3.8008068582955117,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040101564333661213,
      "loss": 5.1957,
      "mean_token_accuracy": 0.1741108775138855,
      "num_tokens": 8330842.0,
      "step": 3770
    },
    {
      "entropy": 5.514135837554932,
      "epoch": 3.805849722642461,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040068683168751316,
      "loss": 5.1545,
      "mean_token_accuracy": 0.17713779509067534,
      "num_tokens": 8341588.0,
      "step": 3775
    },
    {
      "entropy": 5.5442602157592775,
      "epoch": 3.81089258698941,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040035762940084205,
      "loss": 5.2048,
      "mean_token_accuracy": 0.170051971077919,
      "num_tokens": 8351664.0,
      "step": 3780
    },
    {
      "entropy": 5.636847925186157,
      "epoch": 3.815935451336359,
      "grad_norm": 1.03125,
      "learning_rate": 0.000400028037499767,
      "loss": 5.2895,
      "mean_token_accuracy": 0.16166006922721862,
      "num_tokens": 8362803.0,
      "step": 3785
    },
    {
      "entropy": 5.521243333816528,
      "epoch": 3.820978315683308,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003996980570086673,
      "loss": 5.2005,
      "mean_token_accuracy": 0.17746953070163726,
      "num_tokens": 8373791.0,
      "step": 3790
    },
    {
      "entropy": 5.596409702301026,
      "epoch": 3.8260211800302573,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039936768895312987,
      "loss": 5.2076,
      "mean_token_accuracy": 0.1722688153386116,
      "num_tokens": 8385738.0,
      "step": 3795
    },
    {
      "entropy": 5.569927978515625,
      "epoch": 3.8310640443772064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039903693435994626,
      "loss": 5.1994,
      "mean_token_accuracy": 0.17024473696947098,
      "num_tokens": 8396924.0,
      "step": 3800
    },
    {
      "entropy": 5.580840587615967,
      "epoch": 3.8361069087241555,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039870579425710937,
      "loss": 5.2847,
      "mean_token_accuracy": 0.1627711072564125,
      "num_tokens": 8409040.0,
      "step": 3805
    },
    {
      "entropy": 5.4725579738616945,
      "epoch": 3.8411497730711046,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003983742696738103,
      "loss": 5.128,
      "mean_token_accuracy": 0.17557265758514404,
      "num_tokens": 8419630.0,
      "step": 3810
    },
    {
      "entropy": 5.624800014495849,
      "epoch": 3.8461926374180537,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003980423616404351,
      "loss": 5.268,
      "mean_token_accuracy": 0.17147323489189148,
      "num_tokens": 8430036.0,
      "step": 3815
    },
    {
      "entropy": 5.57819504737854,
      "epoch": 3.8512355017650024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003977100711885616,
      "loss": 5.1559,
      "mean_token_accuracy": 0.17113681137561798,
      "num_tokens": 8439829.0,
      "step": 3820
    },
    {
      "entropy": 5.528414011001587,
      "epoch": 3.8562783661119515,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039737739935095614,
      "loss": 5.1705,
      "mean_token_accuracy": 0.167098268866539,
      "num_tokens": 8450492.0,
      "step": 3825
    },
    {
      "entropy": 5.5317644596099855,
      "epoch": 3.8613212304589006,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0003970443471615705,
      "loss": 5.229,
      "mean_token_accuracy": 0.16513402611017228,
      "num_tokens": 8462200.0,
      "step": 3830
    },
    {
      "entropy": 5.599163627624511,
      "epoch": 3.8663640948058497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003967109156555386,
      "loss": 5.1873,
      "mean_token_accuracy": 0.1717960089445114,
      "num_tokens": 8473393.0,
      "step": 3835
    },
    {
      "entropy": 5.498919200897217,
      "epoch": 3.871406959152799,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039637710586917324,
      "loss": 5.1683,
      "mean_token_accuracy": 0.1705564886331558,
      "num_tokens": 8484646.0,
      "step": 3840
    },
    {
      "entropy": 5.567472743988037,
      "epoch": 3.876449823499748,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039604291883996287,
      "loss": 5.2247,
      "mean_token_accuracy": 0.17050785571336746,
      "num_tokens": 8495371.0,
      "step": 3845
    },
    {
      "entropy": 5.553992557525635,
      "epoch": 3.8814926878466967,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039570835560656854,
      "loss": 5.2359,
      "mean_token_accuracy": 0.17486534565687178,
      "num_tokens": 8507976.0,
      "step": 3850
    },
    {
      "entropy": 5.563290071487427,
      "epoch": 3.8865355521936458,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003953734172088205,
      "loss": 5.2093,
      "mean_token_accuracy": 0.17615207582712172,
      "num_tokens": 8519120.0,
      "step": 3855
    },
    {
      "entropy": 5.5632445335388185,
      "epoch": 3.891578416540595,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039503810468771494,
      "loss": 5.2,
      "mean_token_accuracy": 0.1745639830827713,
      "num_tokens": 8529727.0,
      "step": 3860
    },
    {
      "entropy": 5.53722915649414,
      "epoch": 3.896621280887544,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003947024190854111,
      "loss": 5.1703,
      "mean_token_accuracy": 0.16986439228057862,
      "num_tokens": 8540202.0,
      "step": 3865
    },
    {
      "entropy": 5.578685855865478,
      "epoch": 3.901664145234493,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039436636144522737,
      "loss": 5.1756,
      "mean_token_accuracy": 0.17763300836086274,
      "num_tokens": 8550810.0,
      "step": 3870
    },
    {
      "entropy": 5.516595458984375,
      "epoch": 3.9067070095814422,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039402993281163875,
      "loss": 5.1812,
      "mean_token_accuracy": 0.1703437387943268,
      "num_tokens": 8561120.0,
      "step": 3875
    },
    {
      "entropy": 5.560618877410889,
      "epoch": 3.9117498739283914,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003936931342302732,
      "loss": 5.225,
      "mean_token_accuracy": 0.1671963796019554,
      "num_tokens": 8572153.0,
      "step": 3880
    },
    {
      "entropy": 5.558005237579346,
      "epoch": 3.9167927382753405,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003933559667479084,
      "loss": 5.2009,
      "mean_token_accuracy": 0.17049952894449233,
      "num_tokens": 8583740.0,
      "step": 3885
    },
    {
      "entropy": 5.565677213668823,
      "epoch": 3.9218356026222896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039301843141246876,
      "loss": 5.207,
      "mean_token_accuracy": 0.16934604644775392,
      "num_tokens": 8596017.0,
      "step": 3890
    },
    {
      "entropy": 5.583962774276733,
      "epoch": 3.9268784669692387,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039268052927302185,
      "loss": 5.2086,
      "mean_token_accuracy": 0.17622604221105576,
      "num_tokens": 8608121.0,
      "step": 3895
    },
    {
      "entropy": 5.587419509887695,
      "epoch": 3.931921331316188,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0003923422613797753,
      "loss": 5.234,
      "mean_token_accuracy": 0.17340211272239686,
      "num_tokens": 8621446.0,
      "step": 3900
    },
    {
      "entropy": 5.570633602142334,
      "epoch": 3.936964195663137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003920036287840734,
      "loss": 5.2304,
      "mean_token_accuracy": 0.17638086825609206,
      "num_tokens": 8632088.0,
      "step": 3905
    },
    {
      "entropy": 5.526527357101441,
      "epoch": 3.9420070600100856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039166463253839434,
      "loss": 5.0955,
      "mean_token_accuracy": 0.18359021097421646,
      "num_tokens": 8642546.0,
      "step": 3910
    },
    {
      "entropy": 5.541931676864624,
      "epoch": 3.9470499243570347,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003913252736963462,
      "loss": 5.2137,
      "mean_token_accuracy": 0.16715478152036667,
      "num_tokens": 8652869.0,
      "step": 3915
    },
    {
      "entropy": 5.580309057235718,
      "epoch": 3.952092788703984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000390985553312664,
      "loss": 5.1424,
      "mean_token_accuracy": 0.1727869004011154,
      "num_tokens": 8662834.0,
      "step": 3920
    },
    {
      "entropy": 5.575259733200073,
      "epoch": 3.957135653050933,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003906454724432065,
      "loss": 5.2779,
      "mean_token_accuracy": 0.16993938982486725,
      "num_tokens": 8673372.0,
      "step": 3925
    },
    {
      "entropy": 5.552485084533691,
      "epoch": 3.962178517397882,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039030503214495296,
      "loss": 5.1982,
      "mean_token_accuracy": 0.17314982265233994,
      "num_tokens": 8683660.0,
      "step": 3930
    },
    {
      "entropy": 5.578641510009765,
      "epoch": 3.967221381744831,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003899642334759996,
      "loss": 5.2246,
      "mean_token_accuracy": 0.1707220196723938,
      "num_tokens": 8694044.0,
      "step": 3935
    },
    {
      "entropy": 5.581694316864014,
      "epoch": 3.97226424609178,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003896230774955567,
      "loss": 5.242,
      "mean_token_accuracy": 0.1673800379037857,
      "num_tokens": 8705014.0,
      "step": 3940
    },
    {
      "entropy": 5.587334012985229,
      "epoch": 3.977307110438729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038928156526394476,
      "loss": 5.1819,
      "mean_token_accuracy": 0.16846218705177307,
      "num_tokens": 8716007.0,
      "step": 3945
    },
    {
      "entropy": 5.557935953140259,
      "epoch": 3.982349974785678,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003889396978425918,
      "loss": 5.2104,
      "mean_token_accuracy": 0.17732370048761367,
      "num_tokens": 8728390.0,
      "step": 3950
    },
    {
      "entropy": 5.51874041557312,
      "epoch": 3.987392839132627,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003885974762940295,
      "loss": 5.1601,
      "mean_token_accuracy": 0.17531049251556396,
      "num_tokens": 8738806.0,
      "step": 3955
    },
    {
      "entropy": 5.555565595626831,
      "epoch": 3.9924357034795763,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003882549016818905,
      "loss": 5.2108,
      "mean_token_accuracy": 0.17215800583362578,
      "num_tokens": 8749667.0,
      "step": 3960
    },
    {
      "entropy": 5.507802677154541,
      "epoch": 3.9974785678265254,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038791197507090456,
      "loss": 5.2164,
      "mean_token_accuracy": 0.17192330360412597,
      "num_tokens": 8761380.0,
      "step": 3965
    },
    {
      "entropy": 5.543631394704183,
      "epoch": 4.00201714573878,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038756869752689527,
      "loss": 5.0852,
      "mean_token_accuracy": 0.1823349826865726,
      "num_tokens": 8770921.0,
      "step": 3970
    },
    {
      "entropy": 5.501462984085083,
      "epoch": 4.007060010085729,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003872250701167775,
      "loss": 4.8604,
      "mean_token_accuracy": 0.189979450404644,
      "num_tokens": 8782706.0,
      "step": 3975
    },
    {
      "entropy": 5.524435567855835,
      "epoch": 4.012102874432678,
      "grad_norm": 0.9375,
      "learning_rate": 0.00038688109390855297,
      "loss": 4.982,
      "mean_token_accuracy": 0.19535945653915404,
      "num_tokens": 8794600.0,
      "step": 3980
    },
    {
      "entropy": 5.523233938217163,
      "epoch": 4.017145738779627,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003865367699713077,
      "loss": 4.873,
      "mean_token_accuracy": 0.18957179933786392,
      "num_tokens": 8804149.0,
      "step": 3985
    },
    {
      "entropy": 5.529042720794678,
      "epoch": 4.022188603126576,
      "grad_norm": 1.015625,
      "learning_rate": 0.00038619209937520836,
      "loss": 4.8892,
      "mean_token_accuracy": 0.19928188621997833,
      "num_tokens": 8814723.0,
      "step": 3990
    },
    {
      "entropy": 5.537768363952637,
      "epoch": 4.027231467473525,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038584708319149924,
      "loss": 4.9651,
      "mean_token_accuracy": 0.18986125141382218,
      "num_tokens": 8827258.0,
      "step": 3995
    },
    {
      "entropy": 5.505543279647827,
      "epoch": 4.032274331820474,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038550172249249856,
      "loss": 4.9351,
      "mean_token_accuracy": 0.19257259964942933,
      "num_tokens": 8837528.0,
      "step": 4000
    },
    {
      "epoch": 4.032274331820474,
      "eval_entropy": 5.332700965968707,
      "eval_loss": 5.718493461608887,
      "eval_mean_token_accuracy": 0.15012926902176466,
      "eval_num_tokens": 8837528.0,
      "eval_runtime": 2.0906,
      "eval_samples_per_second": 1500.996,
      "eval_steps_per_second": 187.983,
      "step": 4000
    }
  ],
  "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": 2189081738649600.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}