{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9943181818181817,
  "eval_steps": 500,
  "global_step": 702,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.6482378005981445,
      "epoch": 0.028409090909090908,
      "grad_norm": 0.7048962712287903,
      "learning_rate": 1.1062285714285713e-05,
      "loss": 0.2026,
      "mean_token_accuracy": 0.9482126772403717,
      "num_tokens": 609498.0,
      "step": 10
    },
    {
      "entropy": 0.7488664656877517,
      "epoch": 0.056818181818181816,
      "grad_norm": 0.6890662908554077,
      "learning_rate": 2.3353714285714284e-05,
      "loss": 0.3103,
      "mean_token_accuracy": 0.9245607912540436,
      "num_tokens": 1223082.0,
      "step": 20
    },
    {
      "entropy": 0.8222838968038559,
      "epoch": 0.08522727272727272,
      "grad_norm": 0.48184412717819214,
      "learning_rate": 3.564514285714286e-05,
      "loss": 0.1827,
      "mean_token_accuracy": 0.9501622468233109,
      "num_tokens": 1845231.0,
      "step": 30
    },
    {
      "entropy": 0.733986285328865,
      "epoch": 0.11363636363636363,
      "grad_norm": 1.120591163635254,
      "learning_rate": 4.301715405566566e-05,
      "loss": 0.2719,
      "mean_token_accuracy": 0.9292658746242524,
      "num_tokens": 2470218.0,
      "step": 40
    },
    {
      "entropy": 0.7203770101070404,
      "epoch": 0.14204545454545456,
      "grad_norm": 0.7335969805717468,
      "learning_rate": 4.2985148712356925e-05,
      "loss": 0.1647,
      "mean_token_accuracy": 0.9520361691713333,
      "num_tokens": 3092849.0,
      "step": 50
    },
    {
      "entropy": 0.7556389153003693,
      "epoch": 0.17045454545454544,
      "grad_norm": 0.7150644063949585,
      "learning_rate": 4.291765139580166e-05,
      "loss": 0.192,
      "mean_token_accuracy": 0.9466459959745407,
      "num_tokens": 3702183.0,
      "step": 60
    },
    {
      "entropy": 0.758555355668068,
      "epoch": 0.19886363636363635,
      "grad_norm": 0.6904290318489075,
      "learning_rate": 4.281481092356922e-05,
      "loss": 0.207,
      "mean_token_accuracy": 0.9449202001094819,
      "num_tokens": 4312917.0,
      "step": 70
    },
    {
      "entropy": 0.7541693866252899,
      "epoch": 0.22727272727272727,
      "grad_norm": 0.892300546169281,
      "learning_rate": 4.2676854037549315e-05,
      "loss": 0.2208,
      "mean_token_accuracy": 0.9379744172096253,
      "num_tokens": 4935760.0,
      "step": 80
    },
    {
      "entropy": 0.7415777057409286,
      "epoch": 0.2556818181818182,
      "grad_norm": 0.3602456748485565,
      "learning_rate": 4.250408490403328e-05,
      "loss": 0.2102,
      "mean_token_accuracy": 0.9398324251174927,
      "num_tokens": 5541924.0,
      "step": 90
    },
    {
      "entropy": 0.7558239758014679,
      "epoch": 0.2840909090909091,
      "grad_norm": 0.7124091386795044,
      "learning_rate": 4.2296884443090485e-05,
      "loss": 0.2233,
      "mean_token_accuracy": 0.940904438495636,
      "num_tokens": 6151290.0,
      "step": 100
    },
    {
      "entropy": 0.7424949496984482,
      "epoch": 0.3125,
      "grad_norm": 0.37076252698898315,
      "learning_rate": 4.205570948871875e-05,
      "loss": 0.2685,
      "mean_token_accuracy": 0.9265453785657882,
      "num_tokens": 6769338.0,
      "step": 110
    },
    {
      "entropy": 0.7575804442167282,
      "epoch": 0.3409090909090909,
      "grad_norm": 0.9943450093269348,
      "learning_rate": 4.178109178162016e-05,
      "loss": 0.1884,
      "mean_token_accuracy": 0.943663215637207,
      "num_tokens": 7383420.0,
      "step": 120
    },
    {
      "entropy": 0.7821463882923126,
      "epoch": 0.3693181818181818,
      "grad_norm": 0.34976962208747864,
      "learning_rate": 4.1473636796823234e-05,
      "loss": 0.1945,
      "mean_token_accuracy": 0.9438994735479355,
      "num_tokens": 7998446.0,
      "step": 130
    },
    {
      "entropy": 0.7348826706409455,
      "epoch": 0.3977272727272727,
      "grad_norm": 0.8730257153511047,
      "learning_rate": 4.113402240873622e-05,
      "loss": 0.1829,
      "mean_token_accuracy": 0.9465313255786896,
      "num_tokens": 8613496.0,
      "step": 140
    },
    {
      "entropy": 0.7473214626312256,
      "epoch": 0.42613636363636365,
      "grad_norm": 0.5930320024490356,
      "learning_rate": 4.076299739657473e-05,
      "loss": 0.1726,
      "mean_token_accuracy": 0.9504941254854202,
      "num_tokens": 9233303.0,
      "step": 150
    },
    {
      "entropy": 0.772968178987503,
      "epoch": 0.45454545454545453,
      "grad_norm": 0.6510328054428101,
      "learning_rate": 4.0361379793459054e-05,
      "loss": 0.1959,
      "mean_token_accuracy": 0.9439892411231995,
      "num_tokens": 9830889.0,
      "step": 160
    },
    {
      "entropy": 0.7498441338539124,
      "epoch": 0.48295454545454547,
      "grad_norm": 0.8020580410957336,
      "learning_rate": 3.993005508282095e-05,
      "loss": 0.2414,
      "mean_token_accuracy": 0.9348640829324723,
      "num_tokens": 10432314.0,
      "step": 170
    },
    {
      "entropy": 0.7491740703582763,
      "epoch": 0.5113636363636364,
      "grad_norm": 0.770262598991394,
      "learning_rate": 3.946997424609653e-05,
      "loss": 0.1539,
      "mean_token_accuracy": 0.9534165799617768,
      "num_tokens": 11040409.0,
      "step": 180
    },
    {
      "entropy": 0.7374202877283096,
      "epoch": 0.5397727272727273,
      "grad_norm": 0.593521773815155,
      "learning_rate": 3.898215166600964e-05,
      "loss": 0.2063,
      "mean_token_accuracy": 0.9433410108089447,
      "num_tokens": 11649158.0,
      "step": 190
    },
    {
      "entropy": 0.7478526562452317,
      "epoch": 0.5681818181818182,
      "grad_norm": 0.22562482953071594,
      "learning_rate": 3.8467662890068544e-05,
      "loss": 0.1686,
      "mean_token_accuracy": 0.9514732480049133,
      "num_tokens": 12258214.0,
      "step": 200
    },
    {
      "entropy": 0.7442194372415543,
      "epoch": 0.5965909090909091,
      "grad_norm": 0.33614301681518555,
      "learning_rate": 3.792764225920697e-05,
      "loss": 0.2228,
      "mean_token_accuracy": 0.9355043977499008,
      "num_tokens": 12873958.0,
      "step": 210
    },
    {
      "entropy": 0.7340408802032471,
      "epoch": 0.625,
      "grad_norm": 0.7484961152076721,
      "learning_rate": 3.736328040679789e-05,
      "loss": 0.1434,
      "mean_token_accuracy": 0.9569074451923371,
      "num_tokens": 13486044.0,
      "step": 220
    },
    {
      "entropy": 0.7286541283130645,
      "epoch": 0.6534090909090909,
      "grad_norm": 0.58507239818573,
      "learning_rate": 3.677582163355408e-05,
      "loss": 0.2203,
      "mean_token_accuracy": 0.9381987273693084,
      "num_tokens": 14094975.0,
      "step": 230
    },
    {
      "entropy": 0.7239507138729095,
      "epoch": 0.6818181818181818,
      "grad_norm": 0.36877039074897766,
      "learning_rate": 3.6166561164103445e-05,
      "loss": 0.1839,
      "mean_token_accuracy": 0.9480096578598023,
      "num_tokens": 14724816.0,
      "step": 240
    },
    {
      "entropy": 0.7435393124818802,
      "epoch": 0.7102272727272727,
      "grad_norm": 0.6260282397270203,
      "learning_rate": 3.553684229128757e-05,
      "loss": 0.1743,
      "mean_token_accuracy": 0.9525205612182617,
      "num_tokens": 15343654.0,
      "step": 250
    },
    {
      "entropy": 0.73450787961483,
      "epoch": 0.7386363636363636,
      "grad_norm": 0.5552809238433838,
      "learning_rate": 3.488805341447997e-05,
      "loss": 0.1195,
      "mean_token_accuracy": 0.9629803538322449,
      "num_tokens": 15964994.0,
      "step": 260
    },
    {
      "entropy": 0.7273074865341187,
      "epoch": 0.7670454545454546,
      "grad_norm": 0.13038885593414307,
      "learning_rate": 3.422162497845362e-05,
      "loss": 0.1769,
      "mean_token_accuracy": 0.9498629450798035,
      "num_tokens": 16581172.0,
      "step": 270
    },
    {
      "entropy": 0.7412796974182129,
      "epoch": 0.7954545454545454,
      "grad_norm": 0.4582144320011139,
      "learning_rate": 3.353902631954733e-05,
      "loss": 0.1346,
      "mean_token_accuracy": 0.9582979947328567,
      "num_tokens": 17187872.0,
      "step": 280
    },
    {
      "entropy": 0.7395242869853973,
      "epoch": 0.8238636363636364,
      "grad_norm": 0.6720832586288452,
      "learning_rate": 3.28417624260841e-05,
      "loss": 0.1547,
      "mean_token_accuracy": 0.9536517858505249,
      "num_tokens": 17803438.0,
      "step": 290
    },
    {
      "entropy": 0.7136934667825698,
      "epoch": 0.8522727272727273,
      "grad_norm": 0.2975660264492035,
      "learning_rate": 3.213137062018437e-05,
      "loss": 0.1632,
      "mean_token_accuracy": 0.9506705641746521,
      "num_tokens": 18413557.0,
      "step": 300
    },
    {
      "entropy": 0.7071197032928467,
      "epoch": 0.8806818181818182,
      "grad_norm": 0.2304810881614685,
      "learning_rate": 3.140941716828991e-05,
      "loss": 0.1622,
      "mean_token_accuracy": 0.9519901126623154,
      "num_tokens": 19043933.0,
      "step": 310
    },
    {
      "entropy": 0.7309081524610519,
      "epoch": 0.9090909090909091,
      "grad_norm": 0.674328088760376,
      "learning_rate": 3.06774938278715e-05,
      "loss": 0.1646,
      "mean_token_accuracy": 0.9532064646482468,
      "num_tokens": 19662432.0,
      "step": 320
    },
    {
      "entropy": 0.7218959659337998,
      "epoch": 0.9375,
      "grad_norm": 0.4572397768497467,
      "learning_rate": 2.993721433793419e-05,
      "loss": 0.1446,
      "mean_token_accuracy": 0.9599547356367111,
      "num_tokens": 20272284.0,
      "step": 330
    },
    {
      "entropy": 0.728670009970665,
      "epoch": 0.9659090909090909,
      "grad_norm": 0.7572547197341919,
      "learning_rate": 2.9190210861057815e-05,
      "loss": 0.1595,
      "mean_token_accuracy": 0.9513385713100433,
      "num_tokens": 20878084.0,
      "step": 340
    },
    {
      "entropy": 0.7261628389358521,
      "epoch": 0.9943181818181818,
      "grad_norm": 0.6451911330223083,
      "learning_rate": 2.84381303848174e-05,
      "loss": 0.1747,
      "mean_token_accuracy": 0.9511417508125305,
      "num_tokens": 21488473.0,
      "step": 350
    },
    {
      "entropy": 0.692512372136116,
      "epoch": 1.0227272727272727,
      "grad_norm": 0.5031061172485352,
      "learning_rate": 2.7682631090517495e-05,
      "loss": 0.1095,
      "mean_token_accuracy": 0.9642770558595657,
      "num_tokens": 22101186.0,
      "step": 360
    },
    {
      "entropy": 0.6420660585165023,
      "epoch": 1.0511363636363635,
      "grad_norm": 0.5090164542198181,
      "learning_rate": 2.6925378697246658e-05,
      "loss": 0.106,
      "mean_token_accuracy": 0.9648567944765091,
      "num_tokens": 22714309.0,
      "step": 370
    },
    {
      "entropy": 0.6293762028217316,
      "epoch": 1.0795454545454546,
      "grad_norm": 0.31596750020980835,
      "learning_rate": 2.6168042789312696e-05,
      "loss": 0.0946,
      "mean_token_accuracy": 0.9694789320230484,
      "num_tokens": 23332381.0,
      "step": 380
    },
    {
      "entropy": 0.6264711201190949,
      "epoch": 1.1079545454545454,
      "grad_norm": 0.7840130925178528,
      "learning_rate": 2.5412293135155906e-05,
      "loss": 0.1266,
      "mean_token_accuracy": 0.9626366287469864,
      "num_tokens": 23939715.0,
      "step": 390
    },
    {
      "entropy": 0.6454939782619477,
      "epoch": 1.1363636363636362,
      "grad_norm": 0.7014925479888916,
      "learning_rate": 2.465979600585637e-05,
      "loss": 0.1407,
      "mean_token_accuracy": 0.9541082143783569,
      "num_tokens": 24548690.0,
      "step": 400
    },
    {
      "entropy": 0.6302516043186188,
      "epoch": 1.1647727272727273,
      "grad_norm": 0.2765543460845947,
      "learning_rate": 2.3912210501352112e-05,
      "loss": 0.0847,
      "mean_token_accuracy": 0.971521845459938,
      "num_tokens": 25162955.0,
      "step": 410
    },
    {
      "entropy": 0.6302102357149124,
      "epoch": 1.1931818181818181,
      "grad_norm": 0.22406093776226044,
      "learning_rate": 2.3171184892468237e-05,
      "loss": 0.1074,
      "mean_token_accuracy": 0.9653314292430878,
      "num_tokens": 25777238.0,
      "step": 420
    },
    {
      "entropy": 0.635069951415062,
      "epoch": 1.2215909090909092,
      "grad_norm": 0.6240288615226746,
      "learning_rate": 2.243835298682196e-05,
      "loss": 0.1153,
      "mean_token_accuracy": 0.9618167370557785,
      "num_tokens": 26384922.0,
      "step": 430
    },
    {
      "entropy": 0.6203290760517121,
      "epoch": 1.25,
      "grad_norm": 0.9027181267738342,
      "learning_rate": 2.1715330526616013e-05,
      "loss": 0.0932,
      "mean_token_accuracy": 0.9683524072170258,
      "num_tokens": 26998094.0,
      "step": 440
    },
    {
      "entropy": 0.6006172508001327,
      "epoch": 1.2784090909090908,
      "grad_norm": 1.025183916091919,
      "learning_rate": 2.1003711626262565e-05,
      "loss": 0.1201,
      "mean_token_accuracy": 0.9617148607969284,
      "num_tokens": 27610230.0,
      "step": 450
    },
    {
      "entropy": 0.5955485552549362,
      "epoch": 1.3068181818181819,
      "grad_norm": 0.7510467767715454,
      "learning_rate": 2.0305065257691882e-05,
      "loss": 0.1126,
      "mean_token_accuracy": 0.9611585140228271,
      "num_tokens": 28234010.0,
      "step": 460
    },
    {
      "entropy": 0.6035895735025406,
      "epoch": 1.3352272727272727,
      "grad_norm": 0.5905774831771851,
      "learning_rate": 1.962093179109499e-05,
      "loss": 0.0966,
      "mean_token_accuracy": 0.9683466464281082,
      "num_tokens": 28852800.0,
      "step": 470
    },
    {
      "entropy": 0.6095875680446625,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.9473235607147217,
      "learning_rate": 1.8952819598727202e-05,
      "loss": 0.1591,
      "mean_token_accuracy": 0.9514560580253602,
      "num_tokens": 29476443.0,
      "step": 480
    },
    {
      "entropy": 0.6039719879627228,
      "epoch": 1.3920454545454546,
      "grad_norm": 0.6899463534355164,
      "learning_rate": 1.8302201729260465e-05,
      "loss": 0.1228,
      "mean_token_accuracy": 0.9596842557191849,
      "num_tokens": 30089243.0,
      "step": 490
    },
    {
      "entropy": 0.5989584535360336,
      "epoch": 1.4204545454545454,
      "grad_norm": 0.7862110733985901,
      "learning_rate": 1.7670512660016864e-05,
      "loss": 0.0991,
      "mean_token_accuracy": 0.9657784640789032,
      "num_tokens": 30711050.0,
      "step": 500
    },
    {
      "entropy": 0.5999628603458405,
      "epoch": 1.4488636363636362,
      "grad_norm": 0.8022314310073853,
      "learning_rate": 1.7059145134243955e-05,
      "loss": 0.1361,
      "mean_token_accuracy": 0.9545314937829972,
      "num_tokens": 31339225.0,
      "step": 510
    },
    {
      "entropy": 0.5943899869918823,
      "epoch": 1.4772727272727273,
      "grad_norm": 0.7629591822624207,
      "learning_rate": 1.646944709040508e-05,
      "loss": 0.1062,
      "mean_token_accuracy": 0.9659716308116912,
      "num_tokens": 31955130.0,
      "step": 520
    },
    {
      "entropy": 0.6067788809537887,
      "epoch": 1.5056818181818183,
      "grad_norm": 0.3988562524318695,
      "learning_rate": 1.5902718690254872e-05,
      "loss": 0.0886,
      "mean_token_accuracy": 0.9703865736722946,
      "num_tokens": 32562124.0,
      "step": 530
    },
    {
      "entropy": 0.6277791023254394,
      "epoch": 1.5340909090909092,
      "grad_norm": 0.6378304362297058,
      "learning_rate": 1.5360209452252548e-05,
      "loss": 0.1236,
      "mean_token_accuracy": 0.959175643324852,
      "num_tokens": 33166069.0,
      "step": 540
    },
    {
      "entropy": 0.5966621428728104,
      "epoch": 1.5625,
      "grad_norm": 0.283119797706604,
      "learning_rate": 1.4843115496633136e-05,
      "loss": 0.0943,
      "mean_token_accuracy": 0.9684677690267562,
      "num_tokens": 33772029.0,
      "step": 550
    },
    {
      "entropy": 0.5906620651483536,
      "epoch": 1.5909090909090908,
      "grad_norm": 0.5465644598007202,
      "learning_rate": 1.4352576908210681e-05,
      "loss": 0.1051,
      "mean_token_accuracy": 0.966205295920372,
      "num_tokens": 34381610.0,
      "step": 560
    },
    {
      "entropy": 0.5959595113992691,
      "epoch": 1.6193181818181817,
      "grad_norm": 0.21132446825504303,
      "learning_rate": 1.3889675222727937e-05,
      "loss": 0.1054,
      "mean_token_accuracy": 0.9661400139331817,
      "num_tokens": 34993163.0,
      "step": 570
    },
    {
      "entropy": 0.5904346704483032,
      "epoch": 1.6477272727272727,
      "grad_norm": 0.8054311275482178,
      "learning_rate": 1.3455431042294594e-05,
      "loss": 0.1046,
      "mean_token_accuracy": 0.9653222233057022,
      "num_tokens": 35619719.0,
      "step": 580
    },
    {
      "entropy": 0.5979159325361252,
      "epoch": 1.6761363636363638,
      "grad_norm": 0.2166314423084259,
      "learning_rate": 1.305080178517153e-05,
      "loss": 0.1043,
      "mean_token_accuracy": 0.9669523984193802,
      "num_tokens": 36229069.0,
      "step": 590
    },
    {
      "entropy": 0.5969894379377365,
      "epoch": 1.7045454545454546,
      "grad_norm": 0.8285605311393738,
      "learning_rate": 1.2676679574862318e-05,
      "loss": 0.1398,
      "mean_token_accuracy": 0.9544253587722779,
      "num_tokens": 36847744.0,
      "step": 600
    },
    {
      "entropy": 0.5924331277608872,
      "epoch": 1.7329545454545454,
      "grad_norm": 0.5530591011047363,
      "learning_rate": 1.2333889273166127e-05,
      "loss": 0.1127,
      "mean_token_accuracy": 0.9630173146724701,
      "num_tokens": 37468633.0,
      "step": 610
    },
    {
      "entropy": 0.5973960250616074,
      "epoch": 1.7613636363636362,
      "grad_norm": 0.832360565662384,
      "learning_rate": 1.20231866615287e-05,
      "loss": 0.1258,
      "mean_token_accuracy": 0.9604001015424728,
      "num_tokens": 38078521.0,
      "step": 620
    },
    {
      "entropy": 0.5849107921123504,
      "epoch": 1.7897727272727273,
      "grad_norm": 0.6660439968109131,
      "learning_rate": 1.1745256774701201e-05,
      "loss": 0.1223,
      "mean_token_accuracy": 0.9596758246421814,
      "num_tokens": 38700025.0,
      "step": 630
    },
    {
      "entropy": 0.5932971000671386,
      "epoch": 1.8181818181818183,
      "grad_norm": 0.7642334699630737,
      "learning_rate": 1.150071239038078e-05,
      "loss": 0.1081,
      "mean_token_accuracy": 0.9666550129652023,
      "num_tokens": 39307927.0,
      "step": 640
    },
    {
      "entropy": 0.5934834271669388,
      "epoch": 1.8465909090909092,
      "grad_norm": 0.36483898758888245,
      "learning_rate": 1.1290092678163018e-05,
      "loss": 0.1249,
      "mean_token_accuracy": 0.9580485701560975,
      "num_tokens": 39913713.0,
      "step": 650
    },
    {
      "entropy": 0.5831298351287841,
      "epoch": 1.875,
      "grad_norm": 0.9556809663772583,
      "learning_rate": 1.1113862010784877e-05,
      "loss": 0.1257,
      "mean_token_accuracy": 0.9594966471195221,
      "num_tokens": 40529309.0,
      "step": 660
    },
    {
      "entropy": 0.6056891679763794,
      "epoch": 1.9034090909090908,
      "grad_norm": 0.5931299924850464,
      "learning_rate": 1.0972408940279234e-05,
      "loss": 0.145,
      "mean_token_accuracy": 0.9528475373983383,
      "num_tokens": 41128227.0,
      "step": 670
    },
    {
      "entropy": 0.5745501488447189,
      "epoch": 1.9318181818181817,
      "grad_norm": 0.9312835931777954,
      "learning_rate": 1.086604534129833e-05,
      "loss": 0.1022,
      "mean_token_accuracy": 0.9656643062829972,
      "num_tokens": 41760880.0,
      "step": 680
    },
    {
      "entropy": 0.5762614399194718,
      "epoch": 1.9602272727272727,
      "grad_norm": 0.6579990983009338,
      "learning_rate": 1.0795005723494907e-05,
      "loss": 0.0908,
      "mean_token_accuracy": 0.9674515962600708,
      "num_tokens": 42376978.0,
      "step": 690
    },
    {
      "entropy": 0.5801126807928085,
      "epoch": 1.9886363636363638,
      "grad_norm": 0.7780250906944275,
      "learning_rate": 1.07594467144771e-05,
      "loss": 0.0792,
      "mean_token_accuracy": 0.9736239582300186,
      "num_tokens": 42981391.0,
      "step": 700
    }
  ],
  "logging_steps": 10,
  "max_steps": 704,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 2.5804263103922176e+18,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}