{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 1832,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.39035615921020506,
      "epoch": 0.04381161007667032,
      "grad_norm": 19.073434829711914,
      "learning_rate": 2e-05,
      "loss": 4.3018,
      "mean_token_accuracy": 0.5682853847742081,
      "num_tokens": 29485.0,
      "step": 10
    },
    {
      "entropy": 0.7484827220439911,
      "epoch": 0.08762322015334063,
      "grad_norm": 3.970487356185913,
      "learning_rate": 2e-05,
      "loss": 1.9588,
      "mean_token_accuracy": 0.6895880490541458,
      "num_tokens": 58977.0,
      "step": 20
    },
    {
      "entropy": 1.1801635682582856,
      "epoch": 0.13143483023001096,
      "grad_norm": 2.532576322555542,
      "learning_rate": 2e-05,
      "loss": 1.431,
      "mean_token_accuracy": 0.7343599200248718,
      "num_tokens": 88402.0,
      "step": 30
    },
    {
      "entropy": 1.1498825669288635,
      "epoch": 0.17524644030668127,
      "grad_norm": 1.982933521270752,
      "learning_rate": 2e-05,
      "loss": 1.1355,
      "mean_token_accuracy": 0.7867384225130081,
      "num_tokens": 117810.0,
      "step": 40
    },
    {
      "entropy": 0.8834485694766044,
      "epoch": 0.21905805038335158,
      "grad_norm": 1.8390016555786133,
      "learning_rate": 2e-05,
      "loss": 0.8543,
      "mean_token_accuracy": 0.8341205582022667,
      "num_tokens": 147327.0,
      "step": 50
    },
    {
      "entropy": 0.5984191231429576,
      "epoch": 0.2628696604600219,
      "grad_norm": 1.747591257095337,
      "learning_rate": 2e-05,
      "loss": 0.5852,
      "mean_token_accuracy": 0.8857637628912925,
      "num_tokens": 176791.0,
      "step": 60
    },
    {
      "entropy": 0.3483880817890167,
      "epoch": 0.3066812705366922,
      "grad_norm": 1.5721803903579712,
      "learning_rate": 2e-05,
      "loss": 0.3545,
      "mean_token_accuracy": 0.9292098119854927,
      "num_tokens": 206271.0,
      "step": 70
    },
    {
      "entropy": 0.20057316161692143,
      "epoch": 0.35049288061336253,
      "grad_norm": 1.2334290742874146,
      "learning_rate": 2e-05,
      "loss": 0.1877,
      "mean_token_accuracy": 0.9710112065076828,
      "num_tokens": 235692.0,
      "step": 80
    },
    {
      "entropy": 0.12095212489366532,
      "epoch": 0.39430449069003287,
      "grad_norm": 0.8091076016426086,
      "learning_rate": 2e-05,
      "loss": 0.1073,
      "mean_token_accuracy": 0.9890251606702805,
      "num_tokens": 265133.0,
      "step": 90
    },
    {
      "entropy": 0.10107735879719257,
      "epoch": 0.43811610076670315,
      "grad_norm": 0.6297779679298401,
      "learning_rate": 2e-05,
      "loss": 0.085,
      "mean_token_accuracy": 0.9887343898415566,
      "num_tokens": 294684.0,
      "step": 100
    },
    {
      "entropy": 0.08415136393159628,
      "epoch": 0.4819277108433735,
      "grad_norm": 0.47038641571998596,
      "learning_rate": 2e-05,
      "loss": 0.0702,
      "mean_token_accuracy": 0.9906241714954376,
      "num_tokens": 324140.0,
      "step": 110
    },
    {
      "entropy": 0.07522545410320162,
      "epoch": 0.5257393209200438,
      "grad_norm": 0.44743451476097107,
      "learning_rate": 2e-05,
      "loss": 0.0682,
      "mean_token_accuracy": 0.9909884691238403,
      "num_tokens": 353646.0,
      "step": 120
    },
    {
      "entropy": 0.07190036196261644,
      "epoch": 0.5695509309967142,
      "grad_norm": 0.4326375722885132,
      "learning_rate": 2e-05,
      "loss": 0.0679,
      "mean_token_accuracy": 0.990379473567009,
      "num_tokens": 383164.0,
      "step": 130
    },
    {
      "entropy": 0.06936099929735065,
      "epoch": 0.6133625410733844,
      "grad_norm": 0.3178843557834625,
      "learning_rate": 2e-05,
      "loss": 0.0638,
      "mean_token_accuracy": 0.9907173082232476,
      "num_tokens": 412692.0,
      "step": 140
    },
    {
      "entropy": 0.06100328806787729,
      "epoch": 0.6571741511500547,
      "grad_norm": 0.3546409606933594,
      "learning_rate": 2e-05,
      "loss": 0.056,
      "mean_token_accuracy": 0.9925005912780762,
      "num_tokens": 442159.0,
      "step": 150
    },
    {
      "entropy": 0.060952140018343925,
      "epoch": 0.7009857612267251,
      "grad_norm": 0.34292343258857727,
      "learning_rate": 2e-05,
      "loss": 0.0551,
      "mean_token_accuracy": 0.991845327615738,
      "num_tokens": 471592.0,
      "step": 160
    },
    {
      "entropy": 0.06124872919172049,
      "epoch": 0.7447973713033954,
      "grad_norm": 0.3005734384059906,
      "learning_rate": 2e-05,
      "loss": 0.0552,
      "mean_token_accuracy": 0.9918943449854851,
      "num_tokens": 501090.0,
      "step": 170
    },
    {
      "entropy": 0.05876323413103819,
      "epoch": 0.7886089813800657,
      "grad_norm": 0.24807888269424438,
      "learning_rate": 2e-05,
      "loss": 0.0517,
      "mean_token_accuracy": 0.9916606426239014,
      "num_tokens": 530595.0,
      "step": 180
    },
    {
      "entropy": 0.058617806807160375,
      "epoch": 0.8324205914567361,
      "grad_norm": 0.4288617968559265,
      "learning_rate": 2e-05,
      "loss": 0.0572,
      "mean_token_accuracy": 0.9911757484078407,
      "num_tokens": 560059.0,
      "step": 190
    },
    {
      "entropy": 0.05736046200618148,
      "epoch": 0.8762322015334063,
      "grad_norm": 0.3320342004299164,
      "learning_rate": 2e-05,
      "loss": 0.05,
      "mean_token_accuracy": 0.9926259219646454,
      "num_tokens": 589375.0,
      "step": 200
    },
    {
      "entropy": 0.05618324866518378,
      "epoch": 0.9200438116100766,
      "grad_norm": 0.33638137578964233,
      "learning_rate": 2e-05,
      "loss": 0.0525,
      "mean_token_accuracy": 0.9919258087873459,
      "num_tokens": 618850.0,
      "step": 210
    },
    {
      "entropy": 0.05968485539779067,
      "epoch": 0.963855421686747,
      "grad_norm": 0.5651094317436218,
      "learning_rate": 2e-05,
      "loss": 0.052,
      "mean_token_accuracy": 0.9916668817400932,
      "num_tokens": 648391.0,
      "step": 220
    },
    {
      "entropy": 0.060051426794883365,
      "epoch": 1.004381161007667,
      "grad_norm": 0.31502196192741394,
      "learning_rate": 2e-05,
      "loss": 0.0486,
      "mean_token_accuracy": 0.9919279153282577,
      "num_tokens": 675667.0,
      "step": 230
    },
    {
      "entropy": 0.05292966021224856,
      "epoch": 1.0481927710843373,
      "grad_norm": 0.40586429834365845,
      "learning_rate": 2e-05,
      "loss": 0.0475,
      "mean_token_accuracy": 0.9920881032943726,
      "num_tokens": 705116.0,
      "step": 240
    },
    {
      "entropy": 0.052853992022573945,
      "epoch": 1.0920043811610076,
      "grad_norm": 0.2613174021244049,
      "learning_rate": 2e-05,
      "loss": 0.045,
      "mean_token_accuracy": 0.9926601052284241,
      "num_tokens": 734545.0,
      "step": 250
    },
    {
      "entropy": 0.052618366107344626,
      "epoch": 1.135815991237678,
      "grad_norm": 0.3835960030555725,
      "learning_rate": 2e-05,
      "loss": 0.044,
      "mean_token_accuracy": 0.9924288704991341,
      "num_tokens": 764013.0,
      "step": 260
    },
    {
      "entropy": 0.052085249312222005,
      "epoch": 1.1796276013143483,
      "grad_norm": 0.35782596468925476,
      "learning_rate": 2e-05,
      "loss": 0.046,
      "mean_token_accuracy": 0.9919404909014702,
      "num_tokens": 793540.0,
      "step": 270
    },
    {
      "entropy": 0.05476817348971963,
      "epoch": 1.2234392113910186,
      "grad_norm": 0.30708274245262146,
      "learning_rate": 2e-05,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9918309196829795,
      "num_tokens": 823042.0,
      "step": 280
    },
    {
      "entropy": 0.05322778979316354,
      "epoch": 1.267250821467689,
      "grad_norm": 0.2604309022426605,
      "learning_rate": 2e-05,
      "loss": 0.0424,
      "mean_token_accuracy": 0.9924165293574333,
      "num_tokens": 852457.0,
      "step": 290
    },
    {
      "entropy": 0.04979227380827069,
      "epoch": 1.3110624315443593,
      "grad_norm": 0.334300696849823,
      "learning_rate": 2e-05,
      "loss": 0.0404,
      "mean_token_accuracy": 0.9923597663640976,
      "num_tokens": 881910.0,
      "step": 300
    },
    {
      "entropy": 0.05080299507826567,
      "epoch": 1.3548740416210296,
      "grad_norm": 0.3369467854499817,
      "learning_rate": 2e-05,
      "loss": 0.0429,
      "mean_token_accuracy": 0.9918962031602859,
      "num_tokens": 911418.0,
      "step": 310
    },
    {
      "entropy": 0.052211628574877975,
      "epoch": 1.3986856516976998,
      "grad_norm": 0.49774906039237976,
      "learning_rate": 2e-05,
      "loss": 0.0442,
      "mean_token_accuracy": 0.9922201976180076,
      "num_tokens": 940867.0,
      "step": 320
    },
    {
      "entropy": 0.05023647788912058,
      "epoch": 1.44249726177437,
      "grad_norm": 0.3457210659980774,
      "learning_rate": 2e-05,
      "loss": 0.0406,
      "mean_token_accuracy": 0.9923922121524811,
      "num_tokens": 970331.0,
      "step": 330
    },
    {
      "entropy": 0.047794731613248584,
      "epoch": 1.4863088718510404,
      "grad_norm": 0.35947972536087036,
      "learning_rate": 2e-05,
      "loss": 0.0417,
      "mean_token_accuracy": 0.992406377196312,
      "num_tokens": 999841.0,
      "step": 340
    },
    {
      "entropy": 0.047655423637479544,
      "epoch": 1.5301204819277108,
      "grad_norm": 0.37163057923316956,
      "learning_rate": 2e-05,
      "loss": 0.0405,
      "mean_token_accuracy": 0.9928415760397911,
      "num_tokens": 1029329.0,
      "step": 350
    },
    {
      "entropy": 0.04932913850061595,
      "epoch": 1.5739320920043811,
      "grad_norm": 0.34155628085136414,
      "learning_rate": 2e-05,
      "loss": 0.0409,
      "mean_token_accuracy": 0.992286778986454,
      "num_tokens": 1058772.0,
      "step": 360
    },
    {
      "entropy": 0.05013784933835268,
      "epoch": 1.6177437020810514,
      "grad_norm": 0.3228084444999695,
      "learning_rate": 2e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.9925702095031739,
      "num_tokens": 1088250.0,
      "step": 370
    },
    {
      "entropy": 0.04742059959098697,
      "epoch": 1.6615553121577218,
      "grad_norm": 0.2427096664905548,
      "learning_rate": 2e-05,
      "loss": 0.0348,
      "mean_token_accuracy": 0.9927995279431343,
      "num_tokens": 1117693.0,
      "step": 380
    },
    {
      "entropy": 0.04431099114008248,
      "epoch": 1.7053669222343921,
      "grad_norm": 0.2118023931980133,
      "learning_rate": 2e-05,
      "loss": 0.0349,
      "mean_token_accuracy": 0.9929380178451538,
      "num_tokens": 1147140.0,
      "step": 390
    },
    {
      "entropy": 0.04604467884637416,
      "epoch": 1.7491785323110625,
      "grad_norm": 0.25765758752822876,
      "learning_rate": 2e-05,
      "loss": 0.037,
      "mean_token_accuracy": 0.9926380544900895,
      "num_tokens": 1176644.0,
      "step": 400
    },
    {
      "entropy": 0.04647672027349472,
      "epoch": 1.7929901423877328,
      "grad_norm": 0.2914351522922516,
      "learning_rate": 2e-05,
      "loss": 0.0362,
      "mean_token_accuracy": 0.99278933852911,
      "num_tokens": 1206205.0,
      "step": 410
    },
    {
      "entropy": 0.04514645580202341,
      "epoch": 1.8368017524644031,
      "grad_norm": 0.23769572377204895,
      "learning_rate": 2e-05,
      "loss": 0.0375,
      "mean_token_accuracy": 0.992934164404869,
      "num_tokens": 1235651.0,
      "step": 420
    },
    {
      "entropy": 0.044267228711396454,
      "epoch": 1.8806133625410735,
      "grad_norm": 0.22526511549949646,
      "learning_rate": 2e-05,
      "loss": 0.0331,
      "mean_token_accuracy": 0.9943611547350883,
      "num_tokens": 1265105.0,
      "step": 430
    },
    {
      "entropy": 0.04463120717555284,
      "epoch": 1.9244249726177438,
      "grad_norm": 0.27114230394363403,
      "learning_rate": 2e-05,
      "loss": 0.0348,
      "mean_token_accuracy": 0.99424988925457,
      "num_tokens": 1294507.0,
      "step": 440
    },
    {
      "entropy": 0.045189128536731,
      "epoch": 1.9682365826944141,
      "grad_norm": 0.3721632957458496,
      "learning_rate": 2e-05,
      "loss": 0.0348,
      "mean_token_accuracy": 0.9939030453562736,
      "num_tokens": 1324016.0,
      "step": 450
    },
    {
      "entropy": 0.04366534243564348,
      "epoch": 2.008762322015334,
      "grad_norm": 0.2200087010860443,
      "learning_rate": 2e-05,
      "loss": 0.0344,
      "mean_token_accuracy": 0.9943878634555919,
      "num_tokens": 1351286.0,
      "step": 460
    },
    {
      "entropy": 0.04503620527684689,
      "epoch": 2.0525739320920042,
      "grad_norm": 0.2603790760040283,
      "learning_rate": 2e-05,
      "loss": 0.0321,
      "mean_token_accuracy": 0.9944434046745301,
      "num_tokens": 1380791.0,
      "step": 470
    },
    {
      "entropy": 0.04118769532069564,
      "epoch": 2.0963855421686746,
      "grad_norm": 0.29507240653038025,
      "learning_rate": 2e-05,
      "loss": 0.0314,
      "mean_token_accuracy": 0.994631552696228,
      "num_tokens": 1410230.0,
      "step": 480
    },
    {
      "entropy": 0.04305135570466519,
      "epoch": 2.140197152245345,
      "grad_norm": 0.2874581515789032,
      "learning_rate": 2e-05,
      "loss": 0.0325,
      "mean_token_accuracy": 0.9944758623838424,
      "num_tokens": 1439719.0,
      "step": 490
    },
    {
      "entropy": 0.0408366883173585,
      "epoch": 2.1840087623220152,
      "grad_norm": 0.3915460407733917,
      "learning_rate": 2e-05,
      "loss": 0.0304,
      "mean_token_accuracy": 0.9943425431847572,
      "num_tokens": 1469243.0,
      "step": 500
    },
    {
      "entropy": 0.041507516894489525,
      "epoch": 2.2278203723986856,
      "grad_norm": 0.2981090545654297,
      "learning_rate": 2e-05,
      "loss": 0.0299,
      "mean_token_accuracy": 0.9945579648017884,
      "num_tokens": 1498662.0,
      "step": 510
    },
    {
      "entropy": 0.03967566112987697,
      "epoch": 2.271631982475356,
      "grad_norm": 0.38977038860321045,
      "learning_rate": 2e-05,
      "loss": 0.0322,
      "mean_token_accuracy": 0.9944061517715455,
      "num_tokens": 1528141.0,
      "step": 520
    },
    {
      "entropy": 0.039826968545094134,
      "epoch": 2.3154435925520263,
      "grad_norm": 0.3456617295742035,
      "learning_rate": 2e-05,
      "loss": 0.0285,
      "mean_token_accuracy": 0.9949691370129585,
      "num_tokens": 1557578.0,
      "step": 530
    },
    {
      "entropy": 0.0421893454156816,
      "epoch": 2.3592552026286966,
      "grad_norm": 0.3418237566947937,
      "learning_rate": 2e-05,
      "loss": 0.0313,
      "mean_token_accuracy": 0.9939802244305611,
      "num_tokens": 1587144.0,
      "step": 540
    },
    {
      "entropy": 0.0395780008751899,
      "epoch": 2.403066812705367,
      "grad_norm": 0.30387938022613525,
      "learning_rate": 2e-05,
      "loss": 0.0292,
      "mean_token_accuracy": 0.9946425825357437,
      "num_tokens": 1616658.0,
      "step": 550
    },
    {
      "entropy": 0.03764454615302384,
      "epoch": 2.4468784227820373,
      "grad_norm": 0.33084434270858765,
      "learning_rate": 2e-05,
      "loss": 0.0296,
      "mean_token_accuracy": 0.9947469428181648,
      "num_tokens": 1646176.0,
      "step": 560
    },
    {
      "entropy": 0.037794701755046844,
      "epoch": 2.4906900328587076,
      "grad_norm": 0.2900339365005493,
      "learning_rate": 2e-05,
      "loss": 0.0294,
      "mean_token_accuracy": 0.9943759799003601,
      "num_tokens": 1675707.0,
      "step": 570
    },
    {
      "entropy": 0.03779870173893869,
      "epoch": 2.534501642935378,
      "grad_norm": 0.2534691095352173,
      "learning_rate": 2e-05,
      "loss": 0.0296,
      "mean_token_accuracy": 0.9940598547458649,
      "num_tokens": 1705162.0,
      "step": 580
    },
    {
      "entropy": 0.036255474342033266,
      "epoch": 2.5783132530120483,
      "grad_norm": 0.35891082882881165,
      "learning_rate": 2e-05,
      "loss": 0.0276,
      "mean_token_accuracy": 0.9944613263010978,
      "num_tokens": 1734615.0,
      "step": 590
    },
    {
      "entropy": 0.03256695491727442,
      "epoch": 2.6221248630887186,
      "grad_norm": 0.2845175266265869,
      "learning_rate": 2e-05,
      "loss": 0.0278,
      "mean_token_accuracy": 0.994592797756195,
      "num_tokens": 1764014.0,
      "step": 600
    },
    {
      "entropy": 0.03345188507810235,
      "epoch": 2.665936473165389,
      "grad_norm": 0.3167726397514343,
      "learning_rate": 2e-05,
      "loss": 0.0266,
      "mean_token_accuracy": 0.9946425467729568,
      "num_tokens": 1793503.0,
      "step": 610
    },
    {
      "entropy": 0.03313743188045919,
      "epoch": 2.7097480832420593,
      "grad_norm": 0.21710677444934845,
      "learning_rate": 2e-05,
      "loss": 0.0272,
      "mean_token_accuracy": 0.9945120304822922,
      "num_tokens": 1823006.0,
      "step": 620
    },
    {
      "entropy": 0.030652716779150067,
      "epoch": 2.7535596933187296,
      "grad_norm": 0.3033951222896576,
      "learning_rate": 2e-05,
      "loss": 0.0253,
      "mean_token_accuracy": 0.9948923841118813,
      "num_tokens": 1852394.0,
      "step": 630
    },
    {
      "entropy": 0.03487149984575808,
      "epoch": 2.7973713033953995,
      "grad_norm": 0.2704299986362457,
      "learning_rate": 2e-05,
      "loss": 0.029,
      "mean_token_accuracy": 0.9941735804080963,
      "num_tokens": 1881893.0,
      "step": 640
    },
    {
      "entropy": 0.031937569426372645,
      "epoch": 2.8411829134720703,
      "grad_norm": 0.3431037962436676,
      "learning_rate": 2e-05,
      "loss": 0.028,
      "mean_token_accuracy": 0.9944406807422638,
      "num_tokens": 1911381.0,
      "step": 650
    },
    {
      "entropy": 0.031133335269987582,
      "epoch": 2.88499452354874,
      "grad_norm": 0.35892975330352783,
      "learning_rate": 2e-05,
      "loss": 0.0268,
      "mean_token_accuracy": 0.9947311311960221,
      "num_tokens": 1940830.0,
      "step": 660
    },
    {
      "entropy": 0.030122329480946064,
      "epoch": 2.928806133625411,
      "grad_norm": 0.25242748856544495,
      "learning_rate": 2e-05,
      "loss": 0.0276,
      "mean_token_accuracy": 0.9946604892611504,
      "num_tokens": 1970238.0,
      "step": 670
    },
    {
      "entropy": 0.03375864648260176,
      "epoch": 2.972617743702081,
      "grad_norm": 0.22651821374893188,
      "learning_rate": 2e-05,
      "loss": 0.0284,
      "mean_token_accuracy": 0.9947338193655014,
      "num_tokens": 1999673.0,
      "step": 680
    },
    {
      "entropy": 0.02985721411233818,
      "epoch": 3.013143483023001,
      "grad_norm": 0.2214677631855011,
      "learning_rate": 2e-05,
      "loss": 0.0253,
      "mean_token_accuracy": 0.9945424746822666,
      "num_tokens": 2026976.0,
      "step": 690
    },
    {
      "entropy": 0.02816909532994032,
      "epoch": 3.0569550930996714,
      "grad_norm": 0.19869959354400635,
      "learning_rate": 2e-05,
      "loss": 0.0234,
      "mean_token_accuracy": 0.9953982174396515,
      "num_tokens": 2056329.0,
      "step": 700
    },
    {
      "entropy": 0.027280686935409904,
      "epoch": 3.1007667031763417,
      "grad_norm": 0.22260268032550812,
      "learning_rate": 2e-05,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9950461372733116,
      "num_tokens": 2085798.0,
      "step": 710
    },
    {
      "entropy": 0.028381537599489092,
      "epoch": 3.144578313253012,
      "grad_norm": 0.3607195019721985,
      "learning_rate": 2e-05,
      "loss": 0.0267,
      "mean_token_accuracy": 0.9944693371653557,
      "num_tokens": 2115279.0,
      "step": 720
    },
    {
      "entropy": 0.029735187301412225,
      "epoch": 3.1883899233296824,
      "grad_norm": 0.25269201397895813,
      "learning_rate": 2e-05,
      "loss": 0.0273,
      "mean_token_accuracy": 0.9947093442082405,
      "num_tokens": 2144765.0,
      "step": 730
    },
    {
      "entropy": 0.027832537749782206,
      "epoch": 3.2322015334063527,
      "grad_norm": 0.25261572003364563,
      "learning_rate": 2e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9951098829507827,
      "num_tokens": 2174228.0,
      "step": 740
    },
    {
      "entropy": 0.02715804339386523,
      "epoch": 3.276013143483023,
      "grad_norm": 0.27859893441200256,
      "learning_rate": 2e-05,
      "loss": 0.0266,
      "mean_token_accuracy": 0.9944431126117707,
      "num_tokens": 2203757.0,
      "step": 750
    },
    {
      "entropy": 0.028143454669043423,
      "epoch": 3.3198247535596934,
      "grad_norm": 0.2973248064517975,
      "learning_rate": 2e-05,
      "loss": 0.0262,
      "mean_token_accuracy": 0.9948585823178291,
      "num_tokens": 2233150.0,
      "step": 760
    },
    {
      "entropy": 0.030359381577000022,
      "epoch": 3.3636363636363638,
      "grad_norm": 0.2367042601108551,
      "learning_rate": 2e-05,
      "loss": 0.0268,
      "mean_token_accuracy": 0.9942718967795372,
      "num_tokens": 2262643.0,
      "step": 770
    },
    {
      "entropy": 0.026462095649912955,
      "epoch": 3.407447973713034,
      "grad_norm": 0.34042081236839294,
      "learning_rate": 2e-05,
      "loss": 0.0243,
      "mean_token_accuracy": 0.9947688922286033,
      "num_tokens": 2292090.0,
      "step": 780
    },
    {
      "entropy": 0.026516364049166442,
      "epoch": 3.4512595837897044,
      "grad_norm": 0.29167208075523376,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.995012117922306,
      "num_tokens": 2321564.0,
      "step": 790
    },
    {
      "entropy": 0.026681744982488452,
      "epoch": 3.4950711938663748,
      "grad_norm": 0.2620503604412079,
      "learning_rate": 2e-05,
      "loss": 0.0265,
      "mean_token_accuracy": 0.9948214083909989,
      "num_tokens": 2351115.0,
      "step": 800
    },
    {
      "entropy": 0.027950569428503512,
      "epoch": 3.5388828039430447,
      "grad_norm": 0.3914905786514282,
      "learning_rate": 2e-05,
      "loss": 0.0264,
      "mean_token_accuracy": 0.994230942428112,
      "num_tokens": 2380589.0,
      "step": 810
    },
    {
      "entropy": 0.029146080673672258,
      "epoch": 3.5826944140197154,
      "grad_norm": 0.2935093343257904,
      "learning_rate": 2e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9945803046226501,
      "num_tokens": 2410081.0,
      "step": 820
    },
    {
      "entropy": 0.029389529721811414,
      "epoch": 3.6265060240963853,
      "grad_norm": 0.20426620543003082,
      "learning_rate": 2e-05,
      "loss": 0.0251,
      "mean_token_accuracy": 0.9949777990579605,
      "num_tokens": 2439552.0,
      "step": 830
    },
    {
      "entropy": 0.02622994459234178,
      "epoch": 3.670317634173056,
      "grad_norm": 0.37980544567108154,
      "learning_rate": 2e-05,
      "loss": 0.0254,
      "mean_token_accuracy": 0.9946694761514664,
      "num_tokens": 2469015.0,
      "step": 840
    },
    {
      "entropy": 0.027297466271556915,
      "epoch": 3.714129244249726,
      "grad_norm": 0.2875664234161377,
      "learning_rate": 2e-05,
      "loss": 0.0245,
      "mean_token_accuracy": 0.9948430895805359,
      "num_tokens": 2498494.0,
      "step": 850
    },
    {
      "entropy": 0.026779479440301658,
      "epoch": 3.7579408543263964,
      "grad_norm": 0.24819296598434448,
      "learning_rate": 2e-05,
      "loss": 0.0269,
      "mean_token_accuracy": 0.994810126721859,
      "num_tokens": 2527986.0,
      "step": 860
    },
    {
      "entropy": 0.02729630549438298,
      "epoch": 3.8017524644030667,
      "grad_norm": 0.24704840779304504,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9947375372052193,
      "num_tokens": 2557430.0,
      "step": 870
    },
    {
      "entropy": 0.02764119957573712,
      "epoch": 3.845564074479737,
      "grad_norm": 0.3414749503135681,
      "learning_rate": 2e-05,
      "loss": 0.027,
      "mean_token_accuracy": 0.9947045534849167,
      "num_tokens": 2586894.0,
      "step": 880
    },
    {
      "entropy": 0.02578033790923655,
      "epoch": 3.8893756845564074,
      "grad_norm": 0.37503501772880554,
      "learning_rate": 2e-05,
      "loss": 0.0248,
      "mean_token_accuracy": 0.9950520157814026,
      "num_tokens": 2616380.0,
      "step": 890
    },
    {
      "entropy": 0.027033341675996782,
      "epoch": 3.9331872946330777,
      "grad_norm": 0.2305474430322647,
      "learning_rate": 2e-05,
      "loss": 0.0256,
      "mean_token_accuracy": 0.9948741808533669,
      "num_tokens": 2645835.0,
      "step": 900
    },
    {
      "entropy": 0.027415974205359815,
      "epoch": 3.976998904709748,
      "grad_norm": 0.2407739907503128,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.9946468979120254,
      "num_tokens": 2675356.0,
      "step": 910
    },
    {
      "entropy": 0.02766735479235649,
      "epoch": 4.017524644030668,
      "grad_norm": 0.3133447766304016,
      "learning_rate": 2e-05,
      "loss": 0.0266,
      "mean_token_accuracy": 0.9944937438578219,
      "num_tokens": 2702602.0,
      "step": 920
    },
    {
      "entropy": 0.028550118627026676,
      "epoch": 4.061336254107339,
      "grad_norm": 0.25123274326324463,
      "learning_rate": 2e-05,
      "loss": 0.0262,
      "mean_token_accuracy": 0.9943874076008796,
      "num_tokens": 2732020.0,
      "step": 930
    },
    {
      "entropy": 0.024479676922783255,
      "epoch": 4.1051478641840085,
      "grad_norm": 0.45999675989151,
      "learning_rate": 2e-05,
      "loss": 0.0244,
      "mean_token_accuracy": 0.9946165904402733,
      "num_tokens": 2761530.0,
      "step": 940
    },
    {
      "entropy": 0.025697663542814553,
      "epoch": 4.148959474260679,
      "grad_norm": 0.23499128222465515,
      "learning_rate": 2e-05,
      "loss": 0.0235,
      "mean_token_accuracy": 0.9952784523367881,
      "num_tokens": 2790969.0,
      "step": 950
    },
    {
      "entropy": 0.024972660979256035,
      "epoch": 4.192771084337349,
      "grad_norm": 0.27967652678489685,
      "learning_rate": 2e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9953790575265884,
      "num_tokens": 2820393.0,
      "step": 960
    },
    {
      "entropy": 0.02475374243222177,
      "epoch": 4.23658269441402,
      "grad_norm": 0.25424709916114807,
      "learning_rate": 2e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9946705937385559,
      "num_tokens": 2849863.0,
      "step": 970
    },
    {
      "entropy": 0.02558194790035486,
      "epoch": 4.28039430449069,
      "grad_norm": 0.28516727685928345,
      "learning_rate": 2e-05,
      "loss": 0.0229,
      "mean_token_accuracy": 0.9950990095734596,
      "num_tokens": 2879274.0,
      "step": 980
    },
    {
      "entropy": 0.024959080247208477,
      "epoch": 4.324205914567361,
      "grad_norm": 0.20538896322250366,
      "learning_rate": 2e-05,
      "loss": 0.024,
      "mean_token_accuracy": 0.9950569152832032,
      "num_tokens": 2908801.0,
      "step": 990
    },
    {
      "entropy": 0.025352832465432584,
      "epoch": 4.3680175246440305,
      "grad_norm": 0.2579675018787384,
      "learning_rate": 2e-05,
      "loss": 0.0245,
      "mean_token_accuracy": 0.9950070321559906,
      "num_tokens": 2938257.0,
      "step": 1000
    },
    {
      "entropy": 0.02450204298365861,
      "epoch": 4.411829134720701,
      "grad_norm": 0.2892380952835083,
      "learning_rate": 2e-05,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9949192464351654,
      "num_tokens": 2967762.0,
      "step": 1010
    },
    {
      "entropy": 0.02369625587016344,
      "epoch": 4.455640744797371,
      "grad_norm": 0.26433488726615906,
      "learning_rate": 2e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9950045317411422,
      "num_tokens": 2997222.0,
      "step": 1020
    },
    {
      "entropy": 0.027335537271574138,
      "epoch": 4.499452354874042,
      "grad_norm": 0.37580329179763794,
      "learning_rate": 2e-05,
      "loss": 0.0252,
      "mean_token_accuracy": 0.9948343947529793,
      "num_tokens": 3026671.0,
      "step": 1030
    },
    {
      "entropy": 0.025261221639811992,
      "epoch": 4.543263964950712,
      "grad_norm": 0.2803178131580353,
      "learning_rate": 2e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9950425833463669,
      "num_tokens": 3056097.0,
      "step": 1040
    },
    {
      "entropy": 0.026201538788154722,
      "epoch": 4.587075575027383,
      "grad_norm": 0.3461656868457794,
      "learning_rate": 2e-05,
      "loss": 0.0235,
      "mean_token_accuracy": 0.9950102731585503,
      "num_tokens": 3085560.0,
      "step": 1050
    },
    {
      "entropy": 0.024683090089820326,
      "epoch": 4.6308871851040525,
      "grad_norm": 0.22958162426948547,
      "learning_rate": 2e-05,
      "loss": 0.0245,
      "mean_token_accuracy": 0.9952207550406456,
      "num_tokens": 3115048.0,
      "step": 1060
    },
    {
      "entropy": 0.024839983228594063,
      "epoch": 4.674698795180722,
      "grad_norm": 0.3140292763710022,
      "learning_rate": 2e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9952376142144204,
      "num_tokens": 3144452.0,
      "step": 1070
    },
    {
      "entropy": 0.026491868868470193,
      "epoch": 4.718510405257393,
      "grad_norm": 0.2557467818260193,
      "learning_rate": 2e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9949898362159729,
      "num_tokens": 3173986.0,
      "step": 1080
    },
    {
      "entropy": 0.026718211872503162,
      "epoch": 4.762322015334064,
      "grad_norm": 0.3381554186344147,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9950517997145653,
      "num_tokens": 3203519.0,
      "step": 1090
    },
    {
      "entropy": 0.02580249684397131,
      "epoch": 4.806133625410734,
      "grad_norm": 0.2958730161190033,
      "learning_rate": 2e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9952193647623062,
      "num_tokens": 3233021.0,
      "step": 1100
    },
    {
      "entropy": 0.02244551058392972,
      "epoch": 4.849945235487404,
      "grad_norm": 0.2638761103153229,
      "learning_rate": 2e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9949140697717667,
      "num_tokens": 3262501.0,
      "step": 1110
    },
    {
      "entropy": 0.023888250160962344,
      "epoch": 4.8937568455640745,
      "grad_norm": 0.3221336007118225,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9950117602944374,
      "num_tokens": 3291991.0,
      "step": 1120
    },
    {
      "entropy": 0.023271886515431106,
      "epoch": 4.937568455640744,
      "grad_norm": 0.17379634082317352,
      "learning_rate": 2e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9950850084424019,
      "num_tokens": 3321471.0,
      "step": 1130
    },
    {
      "entropy": 0.025164074124768378,
      "epoch": 4.981380065717415,
      "grad_norm": 0.3978523313999176,
      "learning_rate": 2e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.9946789160370827,
      "num_tokens": 3350989.0,
      "step": 1140
    },
    {
      "entropy": 0.02359090921645229,
      "epoch": 5.021905805038335,
      "grad_norm": 0.2893853187561035,
      "learning_rate": 2e-05,
      "loss": 0.0209,
      "mean_token_accuracy": 0.9951927516911481,
      "num_tokens": 3378231.0,
      "step": 1150
    },
    {
      "entropy": 0.025032577151432634,
      "epoch": 5.065717415115006,
      "grad_norm": 0.3284905254840851,
      "learning_rate": 2e-05,
      "loss": 0.0248,
      "mean_token_accuracy": 0.9947940185666084,
      "num_tokens": 3407821.0,
      "step": 1160
    },
    {
      "entropy": 0.023946191160939635,
      "epoch": 5.109529025191676,
      "grad_norm": 0.2103855162858963,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.995547603070736,
      "num_tokens": 3437275.0,
      "step": 1170
    },
    {
      "entropy": 0.025330464565195145,
      "epoch": 5.153340635268346,
      "grad_norm": 0.24447986483573914,
      "learning_rate": 2e-05,
      "loss": 0.0218,
      "mean_token_accuracy": 0.9955911099910736,
      "num_tokens": 3466744.0,
      "step": 1180
    },
    {
      "entropy": 0.02319594454020262,
      "epoch": 5.197152245345016,
      "grad_norm": 0.2545948922634125,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9949484288692474,
      "num_tokens": 3496227.0,
      "step": 1190
    },
    {
      "entropy": 0.024116577091626824,
      "epoch": 5.240963855421687,
      "grad_norm": 0.3212072551250458,
      "learning_rate": 2e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.9953089833259583,
      "num_tokens": 3525658.0,
      "step": 1200
    },
    {
      "entropy": 0.024795983172953128,
      "epoch": 5.284775465498357,
      "grad_norm": 0.3668806254863739,
      "learning_rate": 2e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.994914136826992,
      "num_tokens": 3555128.0,
      "step": 1210
    },
    {
      "entropy": 0.02205991530790925,
      "epoch": 5.328587075575028,
      "grad_norm": 0.2271159589290619,
      "learning_rate": 2e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9950777113437652,
      "num_tokens": 3584595.0,
      "step": 1220
    },
    {
      "entropy": 0.02077565067447722,
      "epoch": 5.372398685651698,
      "grad_norm": 0.2420947551727295,
      "learning_rate": 2e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9958102077245712,
      "num_tokens": 3613973.0,
      "step": 1230
    },
    {
      "entropy": 0.02363784979097545,
      "epoch": 5.416210295728368,
      "grad_norm": 0.1963256448507309,
      "learning_rate": 2e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9953210085630417,
      "num_tokens": 3643480.0,
      "step": 1240
    },
    {
      "entropy": 0.021049648011103272,
      "epoch": 5.460021905805038,
      "grad_norm": 0.31181660294532776,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9951788693666458,
      "num_tokens": 3672937.0,
      "step": 1250
    },
    {
      "entropy": 0.023798084165900946,
      "epoch": 5.503833515881709,
      "grad_norm": 0.21456217765808105,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.994881397485733,
      "num_tokens": 3702446.0,
      "step": 1260
    },
    {
      "entropy": 0.02499352244194597,
      "epoch": 5.547645125958379,
      "grad_norm": 0.28611499071121216,
      "learning_rate": 2e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.995249642431736,
      "num_tokens": 3731934.0,
      "step": 1270
    },
    {
      "entropy": 0.022820547095034272,
      "epoch": 5.591456736035049,
      "grad_norm": 0.25996023416519165,
      "learning_rate": 2e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9951062291860581,
      "num_tokens": 3761366.0,
      "step": 1280
    },
    {
      "entropy": 0.02497695847414434,
      "epoch": 5.63526834611172,
      "grad_norm": 0.309165894985199,
      "learning_rate": 2e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9950066149234772,
      "num_tokens": 3790826.0,
      "step": 1290
    },
    {
      "entropy": 0.023493261518888177,
      "epoch": 5.67907995618839,
      "grad_norm": 0.24247996509075165,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9947842925786972,
      "num_tokens": 3820342.0,
      "step": 1300
    },
    {
      "entropy": 0.023278132849372923,
      "epoch": 5.72289156626506,
      "grad_norm": 0.358727365732193,
      "learning_rate": 2e-05,
      "loss": 0.0212,
      "mean_token_accuracy": 0.9951738849282264,
      "num_tokens": 3849780.0,
      "step": 1310
    },
    {
      "entropy": 0.020839552069082855,
      "epoch": 5.76670317634173,
      "grad_norm": 0.4976311922073364,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9948423251509666,
      "num_tokens": 3879265.0,
      "step": 1320
    },
    {
      "entropy": 0.022906254278495908,
      "epoch": 5.810514786418401,
      "grad_norm": 0.3438393175601959,
      "learning_rate": 2e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9953504383563996,
      "num_tokens": 3908719.0,
      "step": 1330
    },
    {
      "entropy": 0.024066617572680117,
      "epoch": 5.854326396495071,
      "grad_norm": 0.22441036999225616,
      "learning_rate": 2e-05,
      "loss": 0.0237,
      "mean_token_accuracy": 0.9947593525052071,
      "num_tokens": 3938284.0,
      "step": 1340
    },
    {
      "entropy": 0.023073240509256722,
      "epoch": 5.898138006571742,
      "grad_norm": 0.3324027359485626,
      "learning_rate": 2e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.995069320499897,
      "num_tokens": 3967715.0,
      "step": 1350
    },
    {
      "entropy": 0.02433232117909938,
      "epoch": 5.941949616648412,
      "grad_norm": 0.3299965262413025,
      "learning_rate": 2e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9954777494072914,
      "num_tokens": 3997146.0,
      "step": 1360
    },
    {
      "entropy": 0.02257502309512347,
      "epoch": 5.985761226725082,
      "grad_norm": 0.2526472806930542,
      "learning_rate": 2e-05,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9951533004641533,
      "num_tokens": 4026648.0,
      "step": 1370
    },
    {
      "entropy": 0.02323322096285788,
      "epoch": 6.026286966046002,
      "grad_norm": 0.23763661086559296,
      "learning_rate": 2e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9952652873219671,
      "num_tokens": 4053900.0,
      "step": 1380
    },
    {
      "entropy": 0.021967137791216375,
      "epoch": 6.070098576122673,
      "grad_norm": 0.2572341859340668,
      "learning_rate": 2e-05,
      "loss": 0.0202,
      "mean_token_accuracy": 0.9953828155994415,
      "num_tokens": 4083360.0,
      "step": 1390
    },
    {
      "entropy": 0.02138982171891257,
      "epoch": 6.113910186199343,
      "grad_norm": 0.3185424506664276,
      "learning_rate": 2e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.995489752292633,
      "num_tokens": 4112847.0,
      "step": 1400
    },
    {
      "entropy": 0.020864816126413645,
      "epoch": 6.157721796276014,
      "grad_norm": 0.3312909007072449,
      "learning_rate": 2e-05,
      "loss": 0.0207,
      "mean_token_accuracy": 0.9954506799578666,
      "num_tokens": 4142296.0,
      "step": 1410
    },
    {
      "entropy": 0.02233279454521835,
      "epoch": 6.2015334063526835,
      "grad_norm": 0.17053531110286713,
      "learning_rate": 2e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9955132082104683,
      "num_tokens": 4171692.0,
      "step": 1420
    },
    {
      "entropy": 0.024310170486569405,
      "epoch": 6.245345016429353,
      "grad_norm": 0.2670746147632599,
      "learning_rate": 2e-05,
      "loss": 0.0205,
      "mean_token_accuracy": 0.9951695516705513,
      "num_tokens": 4201089.0,
      "step": 1430
    },
    {
      "entropy": 0.020984722138382494,
      "epoch": 6.289156626506024,
      "grad_norm": 0.25105613470077515,
      "learning_rate": 2e-05,
      "loss": 0.0206,
      "mean_token_accuracy": 0.9953115940093994,
      "num_tokens": 4230559.0,
      "step": 1440
    },
    {
      "entropy": 0.02055248327087611,
      "epoch": 6.332968236582694,
      "grad_norm": 0.24033121764659882,
      "learning_rate": 2e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9958187058568001,
      "num_tokens": 4259996.0,
      "step": 1450
    },
    {
      "entropy": 0.022047754074446858,
      "epoch": 6.376779846659365,
      "grad_norm": 0.22444288432598114,
      "learning_rate": 2e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9950413301587104,
      "num_tokens": 4289454.0,
      "step": 1460
    },
    {
      "entropy": 0.022383451019413768,
      "epoch": 6.420591456736035,
      "grad_norm": 0.23084090650081635,
      "learning_rate": 2e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.995081527531147,
      "num_tokens": 4318939.0,
      "step": 1470
    },
    {
      "entropy": 0.022706201835535466,
      "epoch": 6.4644030668127055,
      "grad_norm": 0.3231358528137207,
      "learning_rate": 2e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9955555170774459,
      "num_tokens": 4348417.0,
      "step": 1480
    },
    {
      "entropy": 0.023630459420382978,
      "epoch": 6.508214676889375,
      "grad_norm": 0.481501966714859,
      "learning_rate": 2e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9954674571752549,
      "num_tokens": 4377986.0,
      "step": 1490
    },
    {
      "entropy": 0.020876687136478722,
      "epoch": 6.552026286966046,
      "grad_norm": 0.35793283581733704,
      "learning_rate": 2e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9948447868227959,
      "num_tokens": 4407470.0,
      "step": 1500
    },
    {
      "entropy": 0.02328007298056036,
      "epoch": 6.595837897042716,
      "grad_norm": 0.20913559198379517,
      "learning_rate": 2e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9952251821756363,
      "num_tokens": 4437011.0,
      "step": 1510
    },
    {
      "entropy": 0.02375670406036079,
      "epoch": 6.639649507119387,
      "grad_norm": 0.2503627836704254,
      "learning_rate": 2e-05,
      "loss": 0.0209,
      "mean_token_accuracy": 0.9950839385390282,
      "num_tokens": 4466530.0,
      "step": 1520
    },
    {
      "entropy": 0.021242060861550272,
      "epoch": 6.683461117196057,
      "grad_norm": 0.27485519647598267,
      "learning_rate": 2e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9954615503549575,
      "num_tokens": 4496057.0,
      "step": 1530
    },
    {
      "entropy": 0.021616383735090495,
      "epoch": 6.7272727272727275,
      "grad_norm": 0.19043618440628052,
      "learning_rate": 2e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9952506437897682,
      "num_tokens": 4525543.0,
      "step": 1540
    },
    {
      "entropy": 0.023494412051513792,
      "epoch": 6.771084337349397,
      "grad_norm": 0.19691585004329681,
      "learning_rate": 2e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.99548791795969,
      "num_tokens": 4555013.0,
      "step": 1550
    },
    {
      "entropy": 0.02191384304314852,
      "epoch": 6.814895947426068,
      "grad_norm": 0.29944664239883423,
      "learning_rate": 2e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9951483547687531,
      "num_tokens": 4584494.0,
      "step": 1560
    },
    {
      "entropy": 0.02303541658911854,
      "epoch": 6.858707557502738,
      "grad_norm": 0.2113044410943985,
      "learning_rate": 2e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9948215037584305,
      "num_tokens": 4614045.0,
      "step": 1570
    },
    {
      "entropy": 0.021106354193761945,
      "epoch": 6.902519167579409,
      "grad_norm": 0.21171341836452484,
      "learning_rate": 2e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9955798342823983,
      "num_tokens": 4643457.0,
      "step": 1580
    },
    {
      "entropy": 0.02150448861066252,
      "epoch": 6.946330777656079,
      "grad_norm": 0.24946355819702148,
      "learning_rate": 2e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9956050664186478,
      "num_tokens": 4672821.0,
      "step": 1590
    },
    {
      "entropy": 0.021063918247818947,
      "epoch": 6.9901423877327495,
      "grad_norm": 0.3385717570781708,
      "learning_rate": 2e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9951394200325012,
      "num_tokens": 4702271.0,
      "step": 1600
    },
    {
      "entropy": 0.02021917436473273,
      "epoch": 7.030668127053669,
      "grad_norm": 0.2704486548900604,
      "learning_rate": 2e-05,
      "loss": 0.0188,
      "mean_token_accuracy": 0.9957452045904623,
      "num_tokens": 4729525.0,
      "step": 1610
    },
    {
      "entropy": 0.021311909216456114,
      "epoch": 7.074479737130339,
      "grad_norm": 0.320721298456192,
      "learning_rate": 2e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9953247964382171,
      "num_tokens": 4759055.0,
      "step": 1620
    },
    {
      "entropy": 0.020557987270876765,
      "epoch": 7.11829134720701,
      "grad_norm": 0.32253098487854004,
      "learning_rate": 2e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9955166086554528,
      "num_tokens": 4788511.0,
      "step": 1630
    },
    {
      "entropy": 0.019870327040553092,
      "epoch": 7.16210295728368,
      "grad_norm": 0.2977895140647888,
      "learning_rate": 2e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9952198579907418,
      "num_tokens": 4818030.0,
      "step": 1640
    },
    {
      "entropy": 0.022079354885499926,
      "epoch": 7.205914567360351,
      "grad_norm": 0.19379644095897675,
      "learning_rate": 2e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9956528976559639,
      "num_tokens": 4847471.0,
      "step": 1650
    },
    {
      "entropy": 0.022374613489955664,
      "epoch": 7.2497261774370205,
      "grad_norm": 0.2583862245082855,
      "learning_rate": 2e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9954133436083794,
      "num_tokens": 4876921.0,
      "step": 1660
    },
    {
      "entropy": 0.019138680095784365,
      "epoch": 7.293537787513691,
      "grad_norm": 0.35674577951431274,
      "learning_rate": 2e-05,
      "loss": 0.0196,
      "mean_token_accuracy": 0.9955156922340394,
      "num_tokens": 4906393.0,
      "step": 1670
    },
    {
      "entropy": 0.020489802700467408,
      "epoch": 7.337349397590361,
      "grad_norm": 0.2793170213699341,
      "learning_rate": 2e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9954265221953392,
      "num_tokens": 4935899.0,
      "step": 1680
    },
    {
      "entropy": 0.019693873031064867,
      "epoch": 7.381161007667032,
      "grad_norm": 0.27242857217788696,
      "learning_rate": 2e-05,
      "loss": 0.0198,
      "mean_token_accuracy": 0.9957166820764541,
      "num_tokens": 4965311.0,
      "step": 1690
    },
    {
      "entropy": 0.021065186546184122,
      "epoch": 7.424972617743702,
      "grad_norm": 0.33647313714027405,
      "learning_rate": 2e-05,
      "loss": 0.0189,
      "mean_token_accuracy": 0.9956855684518814,
      "num_tokens": 4994761.0,
      "step": 1700
    },
    {
      "entropy": 0.020657581137493254,
      "epoch": 7.468784227820373,
      "grad_norm": 0.2646650969982147,
      "learning_rate": 2e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9955179765820503,
      "num_tokens": 5024232.0,
      "step": 1710
    },
    {
      "entropy": 0.021019189269281924,
      "epoch": 7.5125958378970425,
      "grad_norm": 0.35804569721221924,
      "learning_rate": 2e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9950902581214904,
      "num_tokens": 5053767.0,
      "step": 1720
    },
    {
      "entropy": 0.023458714364096522,
      "epoch": 7.556407447973713,
      "grad_norm": 0.2962614595890045,
      "learning_rate": 2e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9953871309757233,
      "num_tokens": 5083257.0,
      "step": 1730
    },
    {
      "entropy": 0.020619858335703612,
      "epoch": 7.600219058050383,
      "grad_norm": 0.4548715353012085,
      "learning_rate": 2e-05,
      "loss": 0.0205,
      "mean_token_accuracy": 0.9953182607889175,
      "num_tokens": 5112709.0,
      "step": 1740
    },
    {
      "entropy": 0.018665279296692462,
      "epoch": 7.644030668127054,
      "grad_norm": 0.2674151062965393,
      "learning_rate": 2e-05,
      "loss": 0.0187,
      "mean_token_accuracy": 0.9960940897464752,
      "num_tokens": 5142149.0,
      "step": 1750
    },
    {
      "entropy": 0.020423237257637082,
      "epoch": 7.687842278203724,
      "grad_norm": 0.26541629433631897,
      "learning_rate": 2e-05,
      "loss": 0.021,
      "mean_token_accuracy": 0.9951240062713623,
      "num_tokens": 5171658.0,
      "step": 1760
    },
    {
      "entropy": 0.02253671985818073,
      "epoch": 7.731653888280395,
      "grad_norm": 0.3901165723800659,
      "learning_rate": 2e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9952122256159782,
      "num_tokens": 5201113.0,
      "step": 1770
    },
    {
      "entropy": 0.021983299939893185,
      "epoch": 7.775465498357065,
      "grad_norm": 0.2226109355688095,
      "learning_rate": 2e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9953880906105042,
      "num_tokens": 5230603.0,
      "step": 1780
    },
    {
      "entropy": 0.019754715240560473,
      "epoch": 7.8192771084337345,
      "grad_norm": 0.3183552026748657,
      "learning_rate": 2e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9956612795591354,
      "num_tokens": 5260108.0,
      "step": 1790
    },
    {
      "entropy": 0.021397983329370616,
      "epoch": 7.863088718510405,
      "grad_norm": 0.18899278342723846,
      "learning_rate": 2e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9951773226261139,
      "num_tokens": 5289555.0,
      "step": 1800
    },
    {
      "entropy": 0.022004493232816457,
      "epoch": 7.906900328587076,
      "grad_norm": 0.19192738831043243,
      "learning_rate": 2e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9956512361764908,
      "num_tokens": 5318990.0,
      "step": 1810
    },
    {
      "entropy": 0.020234017237089574,
      "epoch": 7.950711938663746,
      "grad_norm": 0.19193395972251892,
      "learning_rate": 2e-05,
      "loss": 0.0182,
      "mean_token_accuracy": 0.9957286223769188,
      "num_tokens": 5348497.0,
      "step": 1820
    },
    {
      "entropy": 0.020024944259785116,
      "epoch": 7.994523548740416,
      "grad_norm": 0.19811798632144928,
      "learning_rate": 2e-05,
      "loss": 0.0202,
      "mean_token_accuracy": 0.9952393263578415,
      "num_tokens": 5377914.0,
      "step": 1830
    }
  ],
  "logging_steps": 10,
  "max_steps": 1832,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3362085281212416.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}