{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9980430528375734,
  "eval_steps": 500,
  "global_step": 1149,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02609262883235486,
      "grad_norm": 2.5580028899176965,
      "learning_rate": 5e-06,
      "loss": 0.9679,
      "step": 10
    },
    {
      "epoch": 0.05218525766470972,
      "grad_norm": 1.2382684301314397,
      "learning_rate": 5e-06,
      "loss": 0.8485,
      "step": 20
    },
    {
      "epoch": 0.07827788649706457,
      "grad_norm": 1.7244888669737717,
      "learning_rate": 5e-06,
      "loss": 0.8086,
      "step": 30
    },
    {
      "epoch": 0.10437051532941943,
      "grad_norm": 0.6684245608414933,
      "learning_rate": 5e-06,
      "loss": 0.7921,
      "step": 40
    },
    {
      "epoch": 0.1304631441617743,
      "grad_norm": 1.2452393350160769,
      "learning_rate": 5e-06,
      "loss": 0.7827,
      "step": 50
    },
    {
      "epoch": 0.15655577299412915,
      "grad_norm": 0.596876900380745,
      "learning_rate": 5e-06,
      "loss": 0.7727,
      "step": 60
    },
    {
      "epoch": 0.182648401826484,
      "grad_norm": 0.6733648796408476,
      "learning_rate": 5e-06,
      "loss": 0.7657,
      "step": 70
    },
    {
      "epoch": 0.20874103065883887,
      "grad_norm": 0.6054360604030639,
      "learning_rate": 5e-06,
      "loss": 0.7592,
      "step": 80
    },
    {
      "epoch": 0.23483365949119372,
      "grad_norm": 0.6078628726535039,
      "learning_rate": 5e-06,
      "loss": 0.7575,
      "step": 90
    },
    {
      "epoch": 0.2609262883235486,
      "grad_norm": 0.667808601099263,
      "learning_rate": 5e-06,
      "loss": 0.7535,
      "step": 100
    },
    {
      "epoch": 0.28701891715590344,
      "grad_norm": 0.5998036644971939,
      "learning_rate": 5e-06,
      "loss": 0.7505,
      "step": 110
    },
    {
      "epoch": 0.3131115459882583,
      "grad_norm": 0.601234098332995,
      "learning_rate": 5e-06,
      "loss": 0.7462,
      "step": 120
    },
    {
      "epoch": 0.33920417482061316,
      "grad_norm": 0.5470502921448889,
      "learning_rate": 5e-06,
      "loss": 0.7464,
      "step": 130
    },
    {
      "epoch": 0.365296803652968,
      "grad_norm": 0.5354881588471849,
      "learning_rate": 5e-06,
      "loss": 0.7448,
      "step": 140
    },
    {
      "epoch": 0.3913894324853229,
      "grad_norm": 0.638179541327912,
      "learning_rate": 5e-06,
      "loss": 0.7403,
      "step": 150
    },
    {
      "epoch": 0.41748206131767773,
      "grad_norm": 0.7539597801432362,
      "learning_rate": 5e-06,
      "loss": 0.7415,
      "step": 160
    },
    {
      "epoch": 0.4435746901500326,
      "grad_norm": 0.6349380651529639,
      "learning_rate": 5e-06,
      "loss": 0.7397,
      "step": 170
    },
    {
      "epoch": 0.46966731898238745,
      "grad_norm": 0.6486113350477142,
      "learning_rate": 5e-06,
      "loss": 0.7346,
      "step": 180
    },
    {
      "epoch": 0.4957599478147423,
      "grad_norm": 0.524423192769518,
      "learning_rate": 5e-06,
      "loss": 0.7336,
      "step": 190
    },
    {
      "epoch": 0.5218525766470972,
      "grad_norm": 0.8675137881298488,
      "learning_rate": 5e-06,
      "loss": 0.7357,
      "step": 200
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 0.6181294717672231,
      "learning_rate": 5e-06,
      "loss": 0.7324,
      "step": 210
    },
    {
      "epoch": 0.5740378343118069,
      "grad_norm": 0.5310035054090366,
      "learning_rate": 5e-06,
      "loss": 0.7311,
      "step": 220
    },
    {
      "epoch": 0.6001304631441617,
      "grad_norm": 0.5616827688931022,
      "learning_rate": 5e-06,
      "loss": 0.7286,
      "step": 230
    },
    {
      "epoch": 0.6262230919765166,
      "grad_norm": 0.6811648368494987,
      "learning_rate": 5e-06,
      "loss": 0.7291,
      "step": 240
    },
    {
      "epoch": 0.6523157208088715,
      "grad_norm": 0.7733201970691932,
      "learning_rate": 5e-06,
      "loss": 0.7276,
      "step": 250
    },
    {
      "epoch": 0.6784083496412263,
      "grad_norm": 0.6192873191830528,
      "learning_rate": 5e-06,
      "loss": 0.7272,
      "step": 260
    },
    {
      "epoch": 0.7045009784735812,
      "grad_norm": 0.8068236017156095,
      "learning_rate": 5e-06,
      "loss": 0.7249,
      "step": 270
    },
    {
      "epoch": 0.730593607305936,
      "grad_norm": 0.8150623115239886,
      "learning_rate": 5e-06,
      "loss": 0.728,
      "step": 280
    },
    {
      "epoch": 0.7566862361382909,
      "grad_norm": 0.724911564188439,
      "learning_rate": 5e-06,
      "loss": 0.7264,
      "step": 290
    },
    {
      "epoch": 0.7827788649706457,
      "grad_norm": 0.6359822287899792,
      "learning_rate": 5e-06,
      "loss": 0.723,
      "step": 300
    },
    {
      "epoch": 0.8088714938030006,
      "grad_norm": 0.5527636979800452,
      "learning_rate": 5e-06,
      "loss": 0.7232,
      "step": 310
    },
    {
      "epoch": 0.8349641226353555,
      "grad_norm": 0.499616870483793,
      "learning_rate": 5e-06,
      "loss": 0.7222,
      "step": 320
    },
    {
      "epoch": 0.8610567514677103,
      "grad_norm": 0.5494308637029189,
      "learning_rate": 5e-06,
      "loss": 0.7174,
      "step": 330
    },
    {
      "epoch": 0.8871493803000652,
      "grad_norm": 0.5742968590351128,
      "learning_rate": 5e-06,
      "loss": 0.719,
      "step": 340
    },
    {
      "epoch": 0.91324200913242,
      "grad_norm": 0.48431807220252887,
      "learning_rate": 5e-06,
      "loss": 0.7206,
      "step": 350
    },
    {
      "epoch": 0.9393346379647749,
      "grad_norm": 0.4853602229330187,
      "learning_rate": 5e-06,
      "loss": 0.7178,
      "step": 360
    },
    {
      "epoch": 0.9654272667971298,
      "grad_norm": 0.501484126053991,
      "learning_rate": 5e-06,
      "loss": 0.7204,
      "step": 370
    },
    {
      "epoch": 0.9915198956294846,
      "grad_norm": 0.5012192063059191,
      "learning_rate": 5e-06,
      "loss": 0.7179,
      "step": 380
    },
    {
      "epoch": 1.0176125244618395,
      "grad_norm": 0.9524671132017011,
      "learning_rate": 5e-06,
      "loss": 0.7451,
      "step": 390
    },
    {
      "epoch": 1.0437051532941943,
      "grad_norm": 0.524028424392148,
      "learning_rate": 5e-06,
      "loss": 0.6803,
      "step": 400
    },
    {
      "epoch": 1.0697977821265492,
      "grad_norm": 0.5407200294759114,
      "learning_rate": 5e-06,
      "loss": 0.6793,
      "step": 410
    },
    {
      "epoch": 1.095890410958904,
      "grad_norm": 0.5857287578156233,
      "learning_rate": 5e-06,
      "loss": 0.6811,
      "step": 420
    },
    {
      "epoch": 1.121983039791259,
      "grad_norm": 0.5280070429598902,
      "learning_rate": 5e-06,
      "loss": 0.6807,
      "step": 430
    },
    {
      "epoch": 1.1480756686236138,
      "grad_norm": 0.4857639927609277,
      "learning_rate": 5e-06,
      "loss": 0.6776,
      "step": 440
    },
    {
      "epoch": 1.1741682974559686,
      "grad_norm": 0.5945511698424941,
      "learning_rate": 5e-06,
      "loss": 0.68,
      "step": 450
    },
    {
      "epoch": 1.2002609262883235,
      "grad_norm": 0.589811732149492,
      "learning_rate": 5e-06,
      "loss": 0.6831,
      "step": 460
    },
    {
      "epoch": 1.2263535551206783,
      "grad_norm": 0.5534380098945336,
      "learning_rate": 5e-06,
      "loss": 0.6821,
      "step": 470
    },
    {
      "epoch": 1.2524461839530332,
      "grad_norm": 0.5126525326289294,
      "learning_rate": 5e-06,
      "loss": 0.6798,
      "step": 480
    },
    {
      "epoch": 1.278538812785388,
      "grad_norm": 0.6254986498705398,
      "learning_rate": 5e-06,
      "loss": 0.6797,
      "step": 490
    },
    {
      "epoch": 1.304631441617743,
      "grad_norm": 0.49035273013831415,
      "learning_rate": 5e-06,
      "loss": 0.683,
      "step": 500
    },
    {
      "epoch": 1.3307240704500978,
      "grad_norm": 0.5403945589242636,
      "learning_rate": 5e-06,
      "loss": 0.6769,
      "step": 510
    },
    {
      "epoch": 1.3568166992824526,
      "grad_norm": 0.5501681060254049,
      "learning_rate": 5e-06,
      "loss": 0.6807,
      "step": 520
    },
    {
      "epoch": 1.3829093281148075,
      "grad_norm": 0.6772693119372506,
      "learning_rate": 5e-06,
      "loss": 0.6806,
      "step": 530
    },
    {
      "epoch": 1.4090019569471623,
      "grad_norm": 0.5937727651699936,
      "learning_rate": 5e-06,
      "loss": 0.6797,
      "step": 540
    },
    {
      "epoch": 1.4350945857795172,
      "grad_norm": 0.5931432683082334,
      "learning_rate": 5e-06,
      "loss": 0.6769,
      "step": 550
    },
    {
      "epoch": 1.461187214611872,
      "grad_norm": 0.6044257552885106,
      "learning_rate": 5e-06,
      "loss": 0.6821,
      "step": 560
    },
    {
      "epoch": 1.487279843444227,
      "grad_norm": 0.5487167986355509,
      "learning_rate": 5e-06,
      "loss": 0.6779,
      "step": 570
    },
    {
      "epoch": 1.5133724722765818,
      "grad_norm": 0.5046084173932228,
      "learning_rate": 5e-06,
      "loss": 0.6797,
      "step": 580
    },
    {
      "epoch": 1.5394651011089366,
      "grad_norm": 0.588333435552722,
      "learning_rate": 5e-06,
      "loss": 0.681,
      "step": 590
    },
    {
      "epoch": 1.5655577299412915,
      "grad_norm": 0.5857191319203284,
      "learning_rate": 5e-06,
      "loss": 0.6821,
      "step": 600
    },
    {
      "epoch": 1.5916503587736464,
      "grad_norm": 0.5197393978903909,
      "learning_rate": 5e-06,
      "loss": 0.6815,
      "step": 610
    },
    {
      "epoch": 1.6177429876060012,
      "grad_norm": 0.5171612878613464,
      "learning_rate": 5e-06,
      "loss": 0.6786,
      "step": 620
    },
    {
      "epoch": 1.643835616438356,
      "grad_norm": 0.566201693304294,
      "learning_rate": 5e-06,
      "loss": 0.6793,
      "step": 630
    },
    {
      "epoch": 1.669928245270711,
      "grad_norm": 0.5044017255239589,
      "learning_rate": 5e-06,
      "loss": 0.6776,
      "step": 640
    },
    {
      "epoch": 1.6960208741030658,
      "grad_norm": 0.478490790469601,
      "learning_rate": 5e-06,
      "loss": 0.6754,
      "step": 650
    },
    {
      "epoch": 1.7221135029354206,
      "grad_norm": 0.4963811798560764,
      "learning_rate": 5e-06,
      "loss": 0.6798,
      "step": 660
    },
    {
      "epoch": 1.7482061317677755,
      "grad_norm": 0.5278784589567511,
      "learning_rate": 5e-06,
      "loss": 0.6768,
      "step": 670
    },
    {
      "epoch": 1.7742987606001304,
      "grad_norm": 0.5387531965847336,
      "learning_rate": 5e-06,
      "loss": 0.6795,
      "step": 680
    },
    {
      "epoch": 1.8003913894324852,
      "grad_norm": 0.5875937914583538,
      "learning_rate": 5e-06,
      "loss": 0.6781,
      "step": 690
    },
    {
      "epoch": 1.82648401826484,
      "grad_norm": 0.5355905144164954,
      "learning_rate": 5e-06,
      "loss": 0.6792,
      "step": 700
    },
    {
      "epoch": 1.852576647097195,
      "grad_norm": 0.6338025629920409,
      "learning_rate": 5e-06,
      "loss": 0.6821,
      "step": 710
    },
    {
      "epoch": 1.8786692759295498,
      "grad_norm": 0.5291801310990664,
      "learning_rate": 5e-06,
      "loss": 0.677,
      "step": 720
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.590990170736626,
      "learning_rate": 5e-06,
      "loss": 0.6766,
      "step": 730
    },
    {
      "epoch": 1.9308545335942595,
      "grad_norm": 0.5482515525172131,
      "learning_rate": 5e-06,
      "loss": 0.6751,
      "step": 740
    },
    {
      "epoch": 1.9569471624266144,
      "grad_norm": 0.620356822725272,
      "learning_rate": 5e-06,
      "loss": 0.6765,
      "step": 750
    },
    {
      "epoch": 1.9830397912589692,
      "grad_norm": 0.5642132027456009,
      "learning_rate": 5e-06,
      "loss": 0.6776,
      "step": 760
    },
    {
      "epoch": 2.009132420091324,
      "grad_norm": 0.7543688135393036,
      "learning_rate": 5e-06,
      "loss": 0.7153,
      "step": 770
    },
    {
      "epoch": 2.035225048923679,
      "grad_norm": 0.5581679413180287,
      "learning_rate": 5e-06,
      "loss": 0.6374,
      "step": 780
    },
    {
      "epoch": 2.061317677756034,
      "grad_norm": 0.6393564816609882,
      "learning_rate": 5e-06,
      "loss": 0.6364,
      "step": 790
    },
    {
      "epoch": 2.0874103065883887,
      "grad_norm": 0.614009057667238,
      "learning_rate": 5e-06,
      "loss": 0.6402,
      "step": 800
    },
    {
      "epoch": 2.1135029354207435,
      "grad_norm": 0.5346940357958085,
      "learning_rate": 5e-06,
      "loss": 0.6407,
      "step": 810
    },
    {
      "epoch": 2.1395955642530984,
      "grad_norm": 0.559106756693009,
      "learning_rate": 5e-06,
      "loss": 0.638,
      "step": 820
    },
    {
      "epoch": 2.1656881930854532,
      "grad_norm": 0.603074822031344,
      "learning_rate": 5e-06,
      "loss": 0.6363,
      "step": 830
    },
    {
      "epoch": 2.191780821917808,
      "grad_norm": 0.6669696137807871,
      "learning_rate": 5e-06,
      "loss": 0.6394,
      "step": 840
    },
    {
      "epoch": 2.217873450750163,
      "grad_norm": 0.6196755868842616,
      "learning_rate": 5e-06,
      "loss": 0.6407,
      "step": 850
    },
    {
      "epoch": 2.243966079582518,
      "grad_norm": 0.5462481539172109,
      "learning_rate": 5e-06,
      "loss": 0.6407,
      "step": 860
    },
    {
      "epoch": 2.2700587084148727,
      "grad_norm": 0.5378190251307421,
      "learning_rate": 5e-06,
      "loss": 0.6401,
      "step": 870
    },
    {
      "epoch": 2.2961513372472275,
      "grad_norm": 0.5143159135941981,
      "learning_rate": 5e-06,
      "loss": 0.6403,
      "step": 880
    },
    {
      "epoch": 2.3222439660795824,
      "grad_norm": 0.514592343488391,
      "learning_rate": 5e-06,
      "loss": 0.6412,
      "step": 890
    },
    {
      "epoch": 2.3483365949119372,
      "grad_norm": 0.5516158921785709,
      "learning_rate": 5e-06,
      "loss": 0.6434,
      "step": 900
    },
    {
      "epoch": 2.374429223744292,
      "grad_norm": 0.669009619898686,
      "learning_rate": 5e-06,
      "loss": 0.6386,
      "step": 910
    },
    {
      "epoch": 2.400521852576647,
      "grad_norm": 0.6127707444182544,
      "learning_rate": 5e-06,
      "loss": 0.6378,
      "step": 920
    },
    {
      "epoch": 2.426614481409002,
      "grad_norm": 0.6147599175693709,
      "learning_rate": 5e-06,
      "loss": 0.6411,
      "step": 930
    },
    {
      "epoch": 2.4527071102413567,
      "grad_norm": 0.5183387934748579,
      "learning_rate": 5e-06,
      "loss": 0.6431,
      "step": 940
    },
    {
      "epoch": 2.4787997390737115,
      "grad_norm": 0.5481797536089282,
      "learning_rate": 5e-06,
      "loss": 0.6415,
      "step": 950
    },
    {
      "epoch": 2.5048923679060664,
      "grad_norm": 0.49281844671077113,
      "learning_rate": 5e-06,
      "loss": 0.6416,
      "step": 960
    },
    {
      "epoch": 2.5309849967384213,
      "grad_norm": 0.5590165679106437,
      "learning_rate": 5e-06,
      "loss": 0.6413,
      "step": 970
    },
    {
      "epoch": 2.557077625570776,
      "grad_norm": 0.4928056280476181,
      "learning_rate": 5e-06,
      "loss": 0.6398,
      "step": 980
    },
    {
      "epoch": 2.583170254403131,
      "grad_norm": 0.6836197415170101,
      "learning_rate": 5e-06,
      "loss": 0.6458,
      "step": 990
    },
    {
      "epoch": 2.609262883235486,
      "grad_norm": 0.5754142213185701,
      "learning_rate": 5e-06,
      "loss": 0.6391,
      "step": 1000
    },
    {
      "epoch": 2.6353555120678407,
      "grad_norm": 0.5860168635077211,
      "learning_rate": 5e-06,
      "loss": 0.6412,
      "step": 1010
    },
    {
      "epoch": 2.6614481409001955,
      "grad_norm": 0.7689811537018695,
      "learning_rate": 5e-06,
      "loss": 0.6419,
      "step": 1020
    },
    {
      "epoch": 2.6875407697325504,
      "grad_norm": 0.7522119854978059,
      "learning_rate": 5e-06,
      "loss": 0.643,
      "step": 1030
    },
    {
      "epoch": 2.7136333985649053,
      "grad_norm": 0.5487355635946237,
      "learning_rate": 5e-06,
      "loss": 0.6395,
      "step": 1040
    },
    {
      "epoch": 2.73972602739726,
      "grad_norm": 0.6331210725666454,
      "learning_rate": 5e-06,
      "loss": 0.6409,
      "step": 1050
    },
    {
      "epoch": 2.765818656229615,
      "grad_norm": 0.5745683262495273,
      "learning_rate": 5e-06,
      "loss": 0.6447,
      "step": 1060
    },
    {
      "epoch": 2.79191128506197,
      "grad_norm": 0.5686919776335425,
      "learning_rate": 5e-06,
      "loss": 0.6396,
      "step": 1070
    },
    {
      "epoch": 2.8180039138943247,
      "grad_norm": 0.6015098918868974,
      "learning_rate": 5e-06,
      "loss": 0.6414,
      "step": 1080
    },
    {
      "epoch": 2.8440965427266796,
      "grad_norm": 0.4915249083089861,
      "learning_rate": 5e-06,
      "loss": 0.641,
      "step": 1090
    },
    {
      "epoch": 2.8701891715590344,
      "grad_norm": 0.5923242648103054,
      "learning_rate": 5e-06,
      "loss": 0.6418,
      "step": 1100
    },
    {
      "epoch": 2.8962818003913893,
      "grad_norm": 0.5453451599354968,
      "learning_rate": 5e-06,
      "loss": 0.6435,
      "step": 1110
    },
    {
      "epoch": 2.922374429223744,
      "grad_norm": 0.6120521256139003,
      "learning_rate": 5e-06,
      "loss": 0.6448,
      "step": 1120
    },
    {
      "epoch": 2.948467058056099,
      "grad_norm": 0.5414568129256186,
      "learning_rate": 5e-06,
      "loss": 0.6423,
      "step": 1130
    },
    {
      "epoch": 2.974559686888454,
      "grad_norm": 0.5372353896161117,
      "learning_rate": 5e-06,
      "loss": 0.6412,
      "step": 1140
    },
    {
      "epoch": 2.9980430528375734,
      "step": 1149,
      "total_flos": 1924199035699200.0,
      "train_loss": 0.6906216756689538,
      "train_runtime": 17455.8647,
      "train_samples_per_second": 33.704,
      "train_steps_per_second": 0.066
    }
  ],
  "logging_steps": 10,
  "max_steps": 1149,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1924199035699200.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}