{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 658,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1479004077613353,
      "epoch": 0.015197568389057751,
      "grad_norm": 0.51171875,
      "learning_rate": 1.8e-05,
      "loss": 1.5924851417541503,
      "mean_token_accuracy": 0.6634366158396006,
      "num_tokens": 62452.0,
      "step": 10
    },
    {
      "entropy": 1.2071532987058162,
      "epoch": 0.030395136778115502,
      "grad_norm": 0.361328125,
      "learning_rate": 3.8e-05,
      "loss": 1.5587355613708496,
      "mean_token_accuracy": 0.6596032619476319,
      "num_tokens": 125614.0,
      "step": 20
    },
    {
      "entropy": 1.1939050488173961,
      "epoch": 0.04559270516717325,
      "grad_norm": 0.40625,
      "learning_rate": 5.8e-05,
      "loss": 1.3669007301330567,
      "mean_token_accuracy": 0.6933483470231294,
      "num_tokens": 184892.0,
      "step": 30
    },
    {
      "entropy": 1.221787953004241,
      "epoch": 0.060790273556231005,
      "grad_norm": 0.353515625,
      "learning_rate": 7.800000000000001e-05,
      "loss": 1.23649320602417,
      "mean_token_accuracy": 0.7031858772039413,
      "num_tokens": 245953.0,
      "step": 40
    },
    {
      "entropy": 1.077309934794903,
      "epoch": 0.07598784194528875,
      "grad_norm": 0.28515625,
      "learning_rate": 9.8e-05,
      "loss": 1.0901852607727052,
      "mean_token_accuracy": 0.7312592580914498,
      "num_tokens": 306051.0,
      "step": 50
    },
    {
      "entropy": 1.1815281737595797,
      "epoch": 0.0911854103343465,
      "grad_norm": 0.2353515625,
      "learning_rate": 0.000118,
      "loss": 1.188125991821289,
      "mean_token_accuracy": 0.7160288453102112,
      "num_tokens": 366518.0,
      "step": 60
    },
    {
      "entropy": 1.0586148291826247,
      "epoch": 0.10638297872340426,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.000138,
      "loss": 1.0688477516174317,
      "mean_token_accuracy": 0.7354599207639694,
      "num_tokens": 430087.0,
      "step": 70
    },
    {
      "entropy": 1.050880615785718,
      "epoch": 0.12158054711246201,
      "grad_norm": 0.3515625,
      "learning_rate": 0.00015800000000000002,
      "loss": 1.051170825958252,
      "mean_token_accuracy": 0.73864571377635,
      "num_tokens": 491536.0,
      "step": 80
    },
    {
      "entropy": 1.0797226771712303,
      "epoch": 0.13677811550151975,
      "grad_norm": 0.298828125,
      "learning_rate": 0.00017800000000000002,
      "loss": 1.0682001113891602,
      "mean_token_accuracy": 0.7340194799005986,
      "num_tokens": 552928.0,
      "step": 90
    },
    {
      "entropy": 1.0816405173391104,
      "epoch": 0.1519756838905775,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.00019800000000000002,
      "loss": 1.0748048782348634,
      "mean_token_accuracy": 0.7275203809142112,
      "num_tokens": 617585.0,
      "step": 100
    },
    {
      "entropy": 1.1387810289859772,
      "epoch": 0.16717325227963525,
      "grad_norm": 0.2421875,
      "learning_rate": 0.00019987165071710527,
      "loss": 1.1564626693725586,
      "mean_token_accuracy": 0.7178827051073313,
      "num_tokens": 682493.0,
      "step": 110
    },
    {
      "entropy": 1.0470557224005461,
      "epoch": 0.182370820668693,
      "grad_norm": 0.263671875,
      "learning_rate": 0.00019942839715782445,
      "loss": 1.0679837226867677,
      "mean_token_accuracy": 0.7393594264984131,
      "num_tokens": 741585.0,
      "step": 120
    },
    {
      "entropy": 1.022692532464862,
      "epoch": 0.19756838905775076,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0001986700590805916,
      "loss": 1.022587490081787,
      "mean_token_accuracy": 0.7484504476189613,
      "num_tokens": 805276.0,
      "step": 130
    },
    {
      "entropy": 1.0931762170046568,
      "epoch": 0.2127659574468085,
      "grad_norm": 0.2890625,
      "learning_rate": 0.00019759903962771156,
      "loss": 1.0889897346496582,
      "mean_token_accuracy": 0.7326500549912452,
      "num_tokens": 870605.0,
      "step": 140
    },
    {
      "entropy": 1.046710380911827,
      "epoch": 0.22796352583586627,
      "grad_norm": 0.283203125,
      "learning_rate": 0.00019621873281596092,
      "loss": 1.088566493988037,
      "mean_token_accuracy": 0.7342231959104538,
      "num_tokens": 932551.0,
      "step": 150
    },
    {
      "entropy": 1.0759602926671505,
      "epoch": 0.24316109422492402,
      "grad_norm": 0.2265625,
      "learning_rate": 0.00019453351278108806,
      "loss": 1.0699708938598633,
      "mean_token_accuracy": 0.735610119625926,
      "num_tokens": 991401.0,
      "step": 160
    },
    {
      "entropy": 1.0602354250848294,
      "epoch": 0.25835866261398177,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.00019254871991635598,
      "loss": 1.0612274169921876,
      "mean_token_accuracy": 0.7369228135794401,
      "num_tokens": 1053437.0,
      "step": 170
    },
    {
      "entropy": 1.0833635926246643,
      "epoch": 0.2735562310030395,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.00019027064394905473,
      "loss": 1.0936810493469238,
      "mean_token_accuracy": 0.7335428461432457,
      "num_tokens": 1116692.0,
      "step": 180
    },
    {
      "entropy": 1.0710609000176192,
      "epoch": 0.2887537993920973,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.00018770650400861357,
      "loss": 1.0741844177246094,
      "mean_token_accuracy": 0.7365527264773846,
      "num_tokens": 1176488.0,
      "step": 190
    },
    {
      "entropy": 1.0221609488129615,
      "epoch": 0.303951367781155,
      "grad_norm": 0.302734375,
      "learning_rate": 0.00018486442574947511,
      "loss": 1.0395893096923827,
      "mean_token_accuracy": 0.7463255256414414,
      "num_tokens": 1234621.0,
      "step": 200
    },
    {
      "entropy": 1.028166577219963,
      "epoch": 0.3191489361702128,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0001817534156012295,
      "loss": 1.0307375907897949,
      "mean_token_accuracy": 0.7422413423657417,
      "num_tokens": 1293632.0,
      "step": 210
    },
    {
      "entropy": 1.0808881603181362,
      "epoch": 0.3343465045592705,
      "grad_norm": 0.220703125,
      "learning_rate": 0.00017838333222760792,
      "loss": 1.0678336143493652,
      "mean_token_accuracy": 0.7352703791111708,
      "num_tokens": 1355301.0,
      "step": 220
    },
    {
      "entropy": 1.0897805251181125,
      "epoch": 0.3495440729483283,
      "grad_norm": 0.232421875,
      "learning_rate": 0.00017476485528478093,
      "loss": 1.0985455513000488,
      "mean_token_accuracy": 0.7289470963180065,
      "num_tokens": 1416981.0,
      "step": 230
    },
    {
      "entropy": 1.0927846007049085,
      "epoch": 0.364741641337386,
      "grad_norm": 0.23828125,
      "learning_rate": 0.0001709094515779655,
      "loss": 1.0999566078186036,
      "mean_token_accuracy": 0.7316457532346249,
      "num_tokens": 1477934.0,
      "step": 240
    },
    {
      "entropy": 1.0481801606714725,
      "epoch": 0.3799392097264438,
      "grad_norm": 0.2421875,
      "learning_rate": 0.00016682933872358912,
      "loss": 1.0316462516784668,
      "mean_token_accuracy": 0.7426637083292007,
      "num_tokens": 1539241.0,
      "step": 250
    },
    {
      "entropy": 1.0000269718468189,
      "epoch": 0.3951367781155015,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00016253744643216368,
      "loss": 1.0046791076660155,
      "mean_token_accuracy": 0.747460788488388,
      "num_tokens": 1604048.0,
      "step": 260
    },
    {
      "entropy": 1.0748730458319187,
      "epoch": 0.41033434650455924,
      "grad_norm": 0.236328125,
      "learning_rate": 0.0001580473755345625,
      "loss": 1.0607515335083009,
      "mean_token_accuracy": 0.7351726233959198,
      "num_tokens": 1666904.0,
      "step": 270
    },
    {
      "entropy": 1.059519186615944,
      "epoch": 0.425531914893617,
      "grad_norm": 0.224609375,
      "learning_rate": 0.00015337335488154431,
      "loss": 1.0550406455993653,
      "mean_token_accuracy": 0.7438629917800427,
      "num_tokens": 1725233.0,
      "step": 280
    },
    {
      "entropy": 1.049350445717573,
      "epoch": 0.44072948328267475,
      "grad_norm": 0.203125,
      "learning_rate": 0.00014853019625310813,
      "loss": 1.0740507125854493,
      "mean_token_accuracy": 0.7419425651431084,
      "num_tokens": 1788516.0,
      "step": 290
    },
    {
      "entropy": 1.0794111423194408,
      "epoch": 0.45592705167173253,
      "grad_norm": 0.2734375,
      "learning_rate": 0.000143533247420569,
      "loss": 1.0830384254455567,
      "mean_token_accuracy": 0.7354968316853047,
      "num_tokens": 1849950.0,
      "step": 300
    },
    {
      "entropy": 1.0809252394363285,
      "epoch": 0.47112462006079026,
      "grad_norm": 0.271484375,
      "learning_rate": 0.00013839834351009954,
      "loss": 1.106314754486084,
      "mean_token_accuracy": 0.7333131659775972,
      "num_tokens": 1911826.0,
      "step": 310
    },
    {
      "entropy": 1.0794141918420792,
      "epoch": 0.48632218844984804,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0001331417568218636,
      "loss": 1.080414390563965,
      "mean_token_accuracy": 0.7330277658998966,
      "num_tokens": 1971255.0,
      "step": 320
    },
    {
      "entropy": 1.0166626147925855,
      "epoch": 0.5015197568389058,
      "grad_norm": 0.2412109375,
      "learning_rate": 0.00012778014526376353,
      "loss": 1.0223579406738281,
      "mean_token_accuracy": 0.7451204493641853,
      "num_tokens": 2035666.0,
      "step": 330
    },
    {
      "entropy": 1.0680379424244166,
      "epoch": 0.5167173252279635,
      "grad_norm": 0.21875,
      "learning_rate": 0.0001223304995632124,
      "loss": 1.063144588470459,
      "mean_token_accuracy": 0.7386665888130665,
      "num_tokens": 2097745.0,
      "step": 340
    },
    {
      "entropy": 1.1010748460888862,
      "epoch": 0.5319148936170213,
      "grad_norm": 0.265625,
      "learning_rate": 0.00011681008942421483,
      "loss": 1.1296490669250487,
      "mean_token_accuracy": 0.7327280201017856,
      "num_tokens": 2154975.0,
      "step": 350
    },
    {
      "entropy": 1.0673069586977362,
      "epoch": 0.547112462006079,
      "grad_norm": 0.2392578125,
      "learning_rate": 0.00011123640880038233,
      "loss": 1.0613948822021484,
      "mean_token_accuracy": 0.7358887560665608,
      "num_tokens": 2218174.0,
      "step": 360
    },
    {
      "entropy": 1.120655293017626,
      "epoch": 0.5623100303951368,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.00010562712045731084,
      "loss": 1.1228485107421875,
      "mean_token_accuracy": 0.7270086117088794,
      "num_tokens": 2282763.0,
      "step": 370
    },
    {
      "entropy": 1.023940760269761,
      "epoch": 0.5775075987841946,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.0001,
      "loss": 1.01791353225708,
      "mean_token_accuracy": 0.7478193882852793,
      "num_tokens": 2342406.0,
      "step": 380
    },
    {
      "entropy": 1.0468089915812016,
      "epoch": 0.5927051671732523,
      "grad_norm": 0.296875,
      "learning_rate": 9.43728795426892e-05,
      "loss": 1.049548053741455,
      "mean_token_accuracy": 0.742385634034872,
      "num_tokens": 2403644.0,
      "step": 390
    },
    {
      "entropy": 1.0749794770032168,
      "epoch": 0.60790273556231,
      "grad_norm": 0.224609375,
      "learning_rate": 8.87635911996177e-05,
      "loss": 1.0873553276062011,
      "mean_token_accuracy": 0.7330629404634237,
      "num_tokens": 2466781.0,
      "step": 400
    },
    {
      "entropy": 1.0800842259079217,
      "epoch": 0.6231003039513677,
      "grad_norm": 0.2734375,
      "learning_rate": 8.31899105757852e-05,
      "loss": 1.0950210571289063,
      "mean_token_accuracy": 0.7357834167778492,
      "num_tokens": 2529775.0,
      "step": 410
    },
    {
      "entropy": 1.0835391595959663,
      "epoch": 0.6382978723404256,
      "grad_norm": 0.2734375,
      "learning_rate": 7.766950043678764e-05,
      "loss": 1.0691117286682128,
      "mean_token_accuracy": 0.7371876426041126,
      "num_tokens": 2593744.0,
      "step": 420
    },
    {
      "entropy": 1.0298739653080702,
      "epoch": 0.6534954407294833,
      "grad_norm": 0.240234375,
      "learning_rate": 7.221985473623654e-05,
      "loss": 1.0413244247436524,
      "mean_token_accuracy": 0.7439139492809772,
      "num_tokens": 2656829.0,
      "step": 430
    },
    {
      "entropy": 1.1207933265715837,
      "epoch": 0.668693009118541,
      "grad_norm": 0.244140625,
      "learning_rate": 6.685824317813643e-05,
      "loss": 1.136990737915039,
      "mean_token_accuracy": 0.7278301935642958,
      "num_tokens": 2719001.0,
      "step": 440
    },
    {
      "entropy": 1.0710795857012272,
      "epoch": 0.6838905775075987,
      "grad_norm": 0.263671875,
      "learning_rate": 6.160165648990048e-05,
      "loss": 1.0599514961242675,
      "mean_token_accuracy": 0.7398943588137626,
      "num_tokens": 2783541.0,
      "step": 450
    },
    {
      "entropy": 1.0919535238295794,
      "epoch": 0.6990881458966566,
      "grad_norm": 0.609375,
      "learning_rate": 5.6466752579431016e-05,
      "loss": 1.0944193840026855,
      "mean_token_accuracy": 0.7346091568470001,
      "num_tokens": 2845061.0,
      "step": 460
    },
    {
      "entropy": 1.0125693645328284,
      "epoch": 0.7142857142857143,
      "grad_norm": 0.2470703125,
      "learning_rate": 5.146980374689192e-05,
      "loss": 1.0110005378723144,
      "mean_token_accuracy": 0.743566332012415,
      "num_tokens": 2905588.0,
      "step": 470
    },
    {
      "entropy": 1.0804476391524076,
      "epoch": 0.729483282674772,
      "grad_norm": 0.224609375,
      "learning_rate": 4.662664511845568e-05,
      "loss": 1.065206813812256,
      "mean_token_accuracy": 0.7375747956335544,
      "num_tokens": 2969406.0,
      "step": 480
    },
    {
      "entropy": 1.076500639691949,
      "epoch": 0.7446808510638298,
      "grad_norm": 0.2109375,
      "learning_rate": 4.195262446543753e-05,
      "loss": 1.0726984977722167,
      "mean_token_accuracy": 0.7347712397575379,
      "num_tokens": 3033325.0,
      "step": 490
    },
    {
      "entropy": 1.0519217938184737,
      "epoch": 0.7598784194528876,
      "grad_norm": 0.2373046875,
      "learning_rate": 3.746255356783632e-05,
      "loss": 1.0575304985046388,
      "mean_token_accuracy": 0.7421793609857559,
      "num_tokens": 3096358.0,
      "step": 500
    },
    {
      "entropy": 1.1361971575766803,
      "epoch": 0.7750759878419453,
      "grad_norm": 0.26953125,
      "learning_rate": 3.317066127641091e-05,
      "loss": 1.14003267288208,
      "mean_token_accuracy": 0.7234445497393608,
      "num_tokens": 3157442.0,
      "step": 510
    },
    {
      "entropy": 1.0398458503186703,
      "epoch": 0.790273556231003,
      "grad_norm": 0.263671875,
      "learning_rate": 2.9090548422034525e-05,
      "loss": 1.0192261695861817,
      "mean_token_accuracy": 0.7409338817000389,
      "num_tokens": 3217105.0,
      "step": 520
    },
    {
      "entropy": 1.0723623022437097,
      "epoch": 0.8054711246200608,
      "grad_norm": 0.208984375,
      "learning_rate": 2.523514471521913e-05,
      "loss": 1.0806448936462403,
      "mean_token_accuracy": 0.739747503399849,
      "num_tokens": 3279648.0,
      "step": 530
    },
    {
      "entropy": 1.0766091130673885,
      "epoch": 0.8206686930091185,
      "grad_norm": 0.2578125,
      "learning_rate": 2.1616667772392074e-05,
      "loss": 1.0830445289611816,
      "mean_token_accuracy": 0.7354145631194114,
      "num_tokens": 3340664.0,
      "step": 540
    },
    {
      "entropy": 1.0176336735486984,
      "epoch": 0.8358662613981763,
      "grad_norm": 0.2294921875,
      "learning_rate": 1.8246584398770493e-05,
      "loss": 1.0089756965637207,
      "mean_token_accuracy": 0.7476604901254177,
      "num_tokens": 3403315.0,
      "step": 550
    },
    {
      "entropy": 1.1855169147253037,
      "epoch": 0.851063829787234,
      "grad_norm": 0.251953125,
      "learning_rate": 1.5135574250524897e-05,
      "loss": 1.2196799278259278,
      "mean_token_accuracy": 0.711599162966013,
      "num_tokens": 3465413.0,
      "step": 560
    },
    {
      "entropy": 1.0133992433547974,
      "epoch": 0.8662613981762918,
      "grad_norm": 0.22265625,
      "learning_rate": 1.229349599138645e-05,
      "loss": 1.0138180732727051,
      "mean_token_accuracy": 0.7445756837725639,
      "num_tokens": 3527634.0,
      "step": 570
    },
    {
      "entropy": 1.0065228387713432,
      "epoch": 0.8814589665653495,
      "grad_norm": 0.2470703125,
      "learning_rate": 9.729356050945271e-06,
      "loss": 1.0059442520141602,
      "mean_token_accuracy": 0.7528380408883095,
      "num_tokens": 3589816.0,
      "step": 580
    },
    {
      "entropy": 1.026823963969946,
      "epoch": 0.8966565349544073,
      "grad_norm": 0.2265625,
      "learning_rate": 7.4512800836440525e-06,
      "loss": 1.0102774620056152,
      "mean_token_accuracy": 0.7461944825947284,
      "num_tokens": 3656572.0,
      "step": 590
    },
    {
      "entropy": 1.041152635589242,
      "epoch": 0.9118541033434651,
      "grad_norm": 0.2333984375,
      "learning_rate": 5.466487218911942e-06,
      "loss": 1.0332704544067384,
      "mean_token_accuracy": 0.7453692108392715,
      "num_tokens": 3720528.0,
      "step": 600
    },
    {
      "entropy": 1.0129390254616737,
      "epoch": 0.9270516717325228,
      "grad_norm": 0.259765625,
      "learning_rate": 3.7812671840390835e-06,
      "loss": 1.0002843856811523,
      "mean_token_accuracy": 0.7472102656960488,
      "num_tokens": 3784254.0,
      "step": 610
    },
    {
      "entropy": 1.0242063857614994,
      "epoch": 0.9422492401215805,
      "grad_norm": 0.265625,
      "learning_rate": 2.4009603722884742e-06,
      "loss": 0.9934074401855468,
      "mean_token_accuracy": 0.7479867108166218,
      "num_tokens": 3845304.0,
      "step": 620
    },
    {
      "entropy": 1.023862723633647,
      "epoch": 0.9574468085106383,
      "grad_norm": 0.20703125,
      "learning_rate": 1.3299409194084122e-06,
      "loss": 1.0128738403320312,
      "mean_token_accuracy": 0.7516457572579384,
      "num_tokens": 3910188.0,
      "step": 630
    },
    {
      "entropy": 1.0282081000506877,
      "epoch": 0.9726443768996961,
      "grad_norm": 0.2001953125,
      "learning_rate": 5.716028421755671e-07,
      "loss": 1.0188997268676758,
      "mean_token_accuracy": 0.7448701910674572,
      "num_tokens": 3973174.0,
      "step": 640
    },
    {
      "entropy": 1.0912907514721155,
      "epoch": 0.9878419452887538,
      "grad_norm": 0.275390625,
      "learning_rate": 1.2834928289472416e-07,
      "loss": 1.1018651962280273,
      "mean_token_accuracy": 0.7346004512161016,
      "num_tokens": 4038046.0,
      "step": 650
    }
  ],
  "logging_steps": 10,
  "max_steps": 658,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 6.882623888257843e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}