{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 458,
  "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
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 840521320303104.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}