{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1467,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.163015019595623,
      "epoch": 0.01704448610874382,
      "grad_norm": 0.050537109375,
      "learning_rate": 0.00010666666666666667,
      "loss": 2.4524998474121094,
      "mean_token_accuracy": 0.5153323707729578,
      "num_tokens": 132137.0,
      "step": 25
    },
    {
      "entropy": 2.2185084584355352,
      "epoch": 0.03408897221748764,
      "grad_norm": 0.068359375,
      "learning_rate": 0.000199996095299633,
      "loss": 2.173491668701172,
      "mean_token_accuracy": 0.5537562420964242,
      "num_tokens": 261404.0,
      "step": 50
    },
    {
      "entropy": 1.9352111649513244,
      "epoch": 0.051133458326231465,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.0001997948280489004,
      "loss": 1.8705439758300781,
      "mean_token_accuracy": 0.5959619037806988,
      "num_tokens": 396323.0,
      "step": 75
    },
    {
      "entropy": 1.76205413736403,
      "epoch": 0.06817794443497528,
      "grad_norm": 0.09423828125,
      "learning_rate": 0.00019928920737019733,
      "loss": 1.6511923217773437,
      "mean_token_accuracy": 0.6337622724473476,
      "num_tokens": 527358.0,
      "step": 100
    },
    {
      "entropy": 1.4949746784381568,
      "epoch": 0.08522243054371911,
      "grad_norm": 0.0625,
      "learning_rate": 0.00019848077530122083,
      "loss": 1.3673600769042968,
      "mean_token_accuracy": 0.6848376739770174,
      "num_tokens": 655447.0,
      "step": 125
    },
    {
      "entropy": 1.4837670608237385,
      "epoch": 0.10226691665246293,
      "grad_norm": 0.08837890625,
      "learning_rate": 0.0001973719973913012,
      "loss": 1.4141250610351563,
      "mean_token_accuracy": 0.6836329951137304,
      "num_tokens": 790490.0,
      "step": 150
    },
    {
      "entropy": 1.3150090302992612,
      "epoch": 0.11931140276120675,
      "grad_norm": 0.091796875,
      "learning_rate": 0.00019596625518199077,
      "loss": 1.2282736968994141,
      "mean_token_accuracy": 0.7172392677515745,
      "num_tokens": 925069.0,
      "step": 175
    },
    {
      "entropy": 1.3966816508676856,
      "epoch": 0.13635588886995056,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.00019426783589406642,
      "loss": 1.3173268127441407,
      "mean_token_accuracy": 0.7017394271492958,
      "num_tokens": 1052891.0,
      "step": 200
    },
    {
      "entropy": 1.2435051177628338,
      "epoch": 0.15340037497869438,
      "grad_norm": 0.042236328125,
      "learning_rate": 0.00019228191935240007,
      "loss": 1.115916519165039,
      "mean_token_accuracy": 0.7378023877739907,
      "num_tokens": 1184065.0,
      "step": 225
    },
    {
      "entropy": 1.3924890440981834,
      "epoch": 0.17044486108743823,
      "grad_norm": 0.072265625,
      "learning_rate": 0.00019001456218857208,
      "loss": 1.288157958984375,
      "mean_token_accuracy": 0.7004149560630322,
      "num_tokens": 1313612.0,
      "step": 250
    },
    {
      "entropy": 1.3593159637693315,
      "epoch": 0.18748934719618204,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.00018747267936940722,
      "loss": 1.2840399169921874,
      "mean_token_accuracy": 0.7129683202505112,
      "num_tokens": 1444287.0,
      "step": 275
    },
    {
      "entropy": 1.2371155321784317,
      "epoch": 0.20453383330492586,
      "grad_norm": 0.09130859375,
      "learning_rate": 0.00018466402310776708,
      "loss": 1.1687215423583985,
      "mean_token_accuracy": 0.73335683144629,
      "num_tokens": 1577307.0,
      "step": 300
    },
    {
      "entropy": 1.2478438840154558,
      "epoch": 0.22157831941366968,
      "grad_norm": 0.06005859375,
      "learning_rate": 0.00018159715921991612,
      "loss": 1.20022216796875,
      "mean_token_accuracy": 0.732666322439909,
      "num_tokens": 1711348.0,
      "step": 325
    },
    {
      "entropy": 1.1931916671898215,
      "epoch": 0.2386228055224135,
      "grad_norm": 0.103515625,
      "learning_rate": 0.00017828144100156754,
      "loss": 1.082593002319336,
      "mean_token_accuracy": 0.7456949570775032,
      "num_tokens": 1842011.0,
      "step": 350
    },
    {
      "entropy": 1.1265690931119026,
      "epoch": 0.2556672916311573,
      "grad_norm": 0.0888671875,
      "learning_rate": 0.00017472698070228047,
      "loss": 1.0630721282958984,
      "mean_token_accuracy": 0.7553409574180842,
      "num_tokens": 1976056.0,
      "step": 375
    },
    {
      "entropy": 1.1145534914545714,
      "epoch": 0.27271177773990113,
      "grad_norm": 0.06787109375,
      "learning_rate": 0.00017094461868520622,
      "loss": 1.0225981903076171,
      "mean_token_accuracy": 0.7580388783663511,
      "num_tokens": 2109389.0,
      "step": 400
    },
    {
      "entropy": 1.25198206695728,
      "epoch": 0.28975626384864495,
      "grad_norm": 0.048828125,
      "learning_rate": 0.00016694589036623988,
      "loss": 1.194885482788086,
      "mean_token_accuracy": 0.7303700438886881,
      "num_tokens": 2240117.0,
      "step": 425
    },
    {
      "entropy": 1.1865128787327557,
      "epoch": 0.30680074995738876,
      "grad_norm": 0.0556640625,
      "learning_rate": 0.0001627429910334058,
      "loss": 1.118544158935547,
      "mean_token_accuracy": 0.7400260034948587,
      "num_tokens": 2364080.0,
      "step": 450
    },
    {
      "entropy": 1.274649427868426,
      "epoch": 0.3238452360661326,
      "grad_norm": 0.06591796875,
      "learning_rate": 0.00015834873865377077,
      "loss": 1.1928933715820313,
      "mean_token_accuracy": 0.7288886807858944,
      "num_tokens": 2497166.0,
      "step": 475
    },
    {
      "entropy": 1.1076703028334305,
      "epoch": 0.34088972217487645,
      "grad_norm": 0.0693359375,
      "learning_rate": 0.00015377653478131556,
      "loss": 1.0115528869628907,
      "mean_token_accuracy": 0.7560868422687054,
      "num_tokens": 2632559.0,
      "step": 500
    },
    {
      "entropy": 1.1313659130921587,
      "epoch": 0.35793420828362027,
      "grad_norm": 0.08935546875,
      "learning_rate": 0.00014904032368498856,
      "loss": 1.0595830535888673,
      "mean_token_accuracy": 0.756401935890317,
      "num_tokens": 2764478.0,
      "step": 525
    },
    {
      "entropy": 1.0654471471440048,
      "epoch": 0.3749786943923641,
      "grad_norm": 0.0771484375,
      "learning_rate": 0.0001441545498215912,
      "loss": 1.0119430541992187,
      "mean_token_accuracy": 0.768122655749321,
      "num_tokens": 2901372.0,
      "step": 550
    },
    {
      "entropy": 1.2444889663858338,
      "epoch": 0.3920231805011079,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.00013913411378319504,
      "loss": 1.2215270233154296,
      "mean_token_accuracy": 0.729380588978529,
      "num_tokens": 3031999.0,
      "step": 575
    },
    {
      "entropy": 1.0762281655799597,
      "epoch": 0.4090676666098517,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.0001339943268534414,
      "loss": 0.9998902893066406,
      "mean_token_accuracy": 0.768867435157299,
      "num_tokens": 3164064.0,
      "step": 600
    },
    {
      "entropy": 1.0626403710339218,
      "epoch": 0.42611215271859554,
      "grad_norm": 0.134765625,
      "learning_rate": 0.00012875086431131716,
      "loss": 0.9624679565429688,
      "mean_token_accuracy": 0.771867539063096,
      "num_tokens": 3288692.0,
      "step": 625
    },
    {
      "entropy": 0.9459870021557435,
      "epoch": 0.44315663882733936,
      "grad_norm": 0.072265625,
      "learning_rate": 0.00012341971762482043,
      "loss": 0.9064933776855468,
      "mean_token_accuracy": 0.7910215198248625,
      "num_tokens": 3418627.0,
      "step": 650
    },
    {
      "entropy": 1.045862675276585,
      "epoch": 0.4602011249360832,
      "grad_norm": 0.06494140625,
      "learning_rate": 0.00011801714568031615,
      "loss": 0.9381060791015625,
      "mean_token_accuracy": 0.770117818787694,
      "num_tokens": 3545140.0,
      "step": 675
    },
    {
      "entropy": 0.9516787563310936,
      "epoch": 0.477245611044827,
      "grad_norm": 0.07666015625,
      "learning_rate": 0.00011255962519632081,
      "loss": 0.8808039093017578,
      "mean_token_accuracy": 0.7909089859575034,
      "num_tokens": 3679692.0,
      "step": 700
    },
    {
      "entropy": 0.9953352324664593,
      "epoch": 0.4942900971535708,
      "grad_norm": 0.051513671875,
      "learning_rate": 0.00010706380047294496,
      "loss": 0.9232154846191406,
      "mean_token_accuracy": 0.7844866032898427,
      "num_tokens": 3810488.0,
      "step": 725
    },
    {
      "entropy": 0.92748022117652,
      "epoch": 0.5113345832623146,
      "grad_norm": 0.07421875,
      "learning_rate": 0.00010154643263024644,
      "loss": 0.8504108428955078,
      "mean_token_accuracy": 0.7934545055776835,
      "num_tokens": 3937101.0,
      "step": 750
    },
    {
      "entropy": 0.9793470423854888,
      "epoch": 0.5283790693710585,
      "grad_norm": 0.0771484375,
      "learning_rate": 9.602434849030745e-05,
      "loss": 0.8816374969482422,
      "mean_token_accuracy": 0.786624568477273,
      "num_tokens": 4069539.0,
      "step": 775
    },
    {
      "entropy": 0.9201436020294205,
      "epoch": 0.5454235554798023,
      "grad_norm": 0.06005859375,
      "learning_rate": 9.051438925893384e-05,
      "loss": 0.8456056976318359,
      "mean_token_accuracy": 0.7975382874906063,
      "num_tokens": 4196197.0,
      "step": 800
    },
    {
      "entropy": 0.8780001322366298,
      "epoch": 0.5624680415885461,
      "grad_norm": 0.06787109375,
      "learning_rate": 8.503335916348721e-05,
      "loss": 0.7664035034179687,
      "mean_token_accuracy": 0.8094721902906895,
      "num_tokens": 4328551.0,
      "step": 825
    },
    {
      "entropy": 1.0318437330564485,
      "epoch": 0.5795125276972899,
      "grad_norm": 0.0625,
      "learning_rate": 7.959797420349355e-05,
      "loss": 0.9603571319580078,
      "mean_token_accuracy": 0.7762538179755211,
      "num_tokens": 4461738.0,
      "step": 850
    },
    {
      "entropy": 1.001236357828602,
      "epoch": 0.5965570138060338,
      "grad_norm": 0.05322265625,
      "learning_rate": 7.422481117032823e-05,
      "loss": 0.9705011749267578,
      "mean_token_accuracy": 0.7822855979949236,
      "num_tokens": 4595951.0,
      "step": 875
    },
    {
      "entropy": 0.9190932653006166,
      "epoch": 0.6136014999147775,
      "grad_norm": 0.1259765625,
      "learning_rate": 6.893025709145718e-05,
      "loss": 0.7840510559082031,
      "mean_token_accuracy": 0.7955148361623288,
      "num_tokens": 4721963.0,
      "step": 900
    },
    {
      "entropy": 0.9409526876406744,
      "epoch": 0.6306459860235214,
      "grad_norm": 0.07421875,
      "learning_rate": 6.373045925341794e-05,
      "loss": 0.8891078948974609,
      "mean_token_accuracy": 0.797613599896431,
      "num_tokens": 4857248.0,
      "step": 925
    },
    {
      "entropy": 0.9927360657276586,
      "epoch": 0.6476904721322652,
      "grad_norm": 0.05859375,
      "learning_rate": 5.864127595596118e-05,
      "loss": 0.9160787963867187,
      "mean_token_accuracy": 0.7845662014186382,
      "num_tokens": 4990977.0,
      "step": 950
    },
    {
      "entropy": 1.0460572233749554,
      "epoch": 0.664734958241009,
      "grad_norm": 0.048583984375,
      "learning_rate": 5.367822814754124e-05,
      "loss": 1.0131810760498048,
      "mean_token_accuracy": 0.7867030793055892,
      "num_tokens": 5126029.0,
      "step": 975
    },
    {
      "entropy": 0.9399900265224278,
      "epoch": 0.6817794443497529,
      "grad_norm": 0.0810546875,
      "learning_rate": 4.885645208965779e-05,
      "loss": 0.8525303649902344,
      "mean_token_accuracy": 0.7958442622423172,
      "num_tokens": 5258088.0,
      "step": 1000
    },
    {
      "entropy": 0.9254711728263646,
      "epoch": 0.6988239304584967,
      "grad_norm": 0.06298828125,
      "learning_rate": 4.419065319441306e-05,
      "loss": 0.8452438354492188,
      "mean_token_accuracy": 0.8007816936075688,
      "num_tokens": 5389128.0,
      "step": 1025
    },
    {
      "entropy": 0.9583898686571046,
      "epoch": 0.7158684165672405,
      "grad_norm": 0.0615234375,
      "learning_rate": 3.969506117606927e-05,
      "loss": 0.910721435546875,
      "mean_token_accuracy": 0.7937231123447418,
      "num_tokens": 5518384.0,
      "step": 1050
    },
    {
      "entropy": 0.9213999844761566,
      "epoch": 0.7329129026759843,
      "grad_norm": 0.0625,
      "learning_rate": 3.538338665338589e-05,
      "loss": 0.8783349609375,
      "mean_token_accuracy": 0.7985107507556677,
      "num_tokens": 5649311.0,
      "step": 1075
    },
    {
      "entropy": 1.0327887642104179,
      "epoch": 0.7499573887847282,
      "grad_norm": 0.0810546875,
      "learning_rate": 3.126877933508957e-05,
      "loss": 0.9161809539794922,
      "mean_token_accuracy": 0.7725442545861005,
      "num_tokens": 5777733.0,
      "step": 1100
    },
    {
      "entropy": 1.0169112877128645,
      "epoch": 0.7670018748934719,
      "grad_norm": 0.0595703125,
      "learning_rate": 2.7363787916003037e-05,
      "loss": 0.9725022888183594,
      "mean_token_accuracy": 0.777765890210867,
      "num_tokens": 5909211.0,
      "step": 1125
    },
    {
      "entropy": 0.9282390323840082,
      "epoch": 0.7840463610022158,
      "grad_norm": 0.06298828125,
      "learning_rate": 2.368032180614118e-05,
      "loss": 0.8365528869628907,
      "mean_token_accuracy": 0.7959657093882561,
      "num_tokens": 6044115.0,
      "step": 1150
    },
    {
      "entropy": 0.9648981914808974,
      "epoch": 0.8010908471109596,
      "grad_norm": 0.05712890625,
      "learning_rate": 2.0229614809492303e-05,
      "loss": 0.8802178955078125,
      "mean_token_accuracy": 0.7891545896232128,
      "num_tokens": 6175533.0,
      "step": 1175
    },
    {
      "entropy": 0.8922164059896022,
      "epoch": 0.8181353332197034,
      "grad_norm": 0.05859375,
      "learning_rate": 1.7022190863257183e-05,
      "loss": 0.8290106201171875,
      "mean_token_accuracy": 0.8064813084155321,
      "num_tokens": 6310296.0,
      "step": 1200
    },
    {
      "entropy": 0.9672217888385057,
      "epoch": 0.8351798193284472,
      "grad_norm": 0.06884765625,
      "learning_rate": 1.4067831942033904e-05,
      "loss": 0.908281478881836,
      "mean_token_accuracy": 0.7912823136150837,
      "num_tokens": 6436503.0,
      "step": 1225
    },
    {
      "entropy": 1.0066828985884786,
      "epoch": 0.8522243054371911,
      "grad_norm": 0.06494140625,
      "learning_rate": 1.137554822483381e-05,
      "loss": 0.9817295074462891,
      "mean_token_accuracy": 0.7799629791080952,
      "num_tokens": 6572817.0,
      "step": 1250
    },
    {
      "entropy": 0.8913562398357316,
      "epoch": 0.8692687915459348,
      "grad_norm": 0.07421875,
      "learning_rate": 8.953550615913397e-06,
      "loss": 0.7918276977539063,
      "mean_token_accuracy": 0.8021628592163325,
      "num_tokens": 6701599.0,
      "step": 1275
    },
    {
      "entropy": 0.9468374169385061,
      "epoch": 0.8863132776546787,
      "grad_norm": 0.07958984375,
      "learning_rate": 6.809225703227351e-06,
      "loss": 0.8823848724365234,
      "mean_token_accuracy": 0.7927402850985527,
      "num_tokens": 6838746.0,
      "step": 1300
    },
    {
      "entropy": 0.9219346580654383,
      "epoch": 0.9033577637634226,
      "grad_norm": 0.0869140625,
      "learning_rate": 4.949113230874847e-06,
      "loss": 0.8668992614746094,
      "mean_token_accuracy": 0.797563666626811,
      "num_tokens": 6971712.0,
      "step": 1325
    },
    {
      "entropy": 0.984245891966857,
      "epoch": 0.9204022498721663,
      "grad_norm": 0.07666015625,
      "learning_rate": 3.3788861542428553e-06,
      "loss": 0.9503571319580079,
      "mean_token_accuracy": 0.792408836260438,
      "num_tokens": 7106279.0,
      "step": 1350
    },
    {
      "entropy": 0.913267534552142,
      "epoch": 0.9374467359809102,
      "grad_norm": 0.115234375,
      "learning_rate": 2.1033333386744846e-06,
      "loss": 0.8369984436035156,
      "mean_token_accuracy": 0.803322741240263,
      "num_tokens": 7238150.0,
      "step": 1375
    },
    {
      "entropy": 0.8992303860466927,
      "epoch": 0.954491222089654,
      "grad_norm": 0.064453125,
      "learning_rate": 1.1263449544276938e-06,
      "loss": 0.8289890289306641,
      "mean_token_accuracy": 0.8079592730104923,
      "num_tokens": 7374135.0,
      "step": 1400
    },
    {
      "entropy": 0.9052433096244932,
      "epoch": 0.9715357081983979,
      "grad_norm": 0.0498046875,
      "learning_rate": 4.5090061246676475e-07,
      "loss": 0.7991035461425782,
      "mean_token_accuracy": 0.7997805786132812,
      "num_tokens": 7508742.0,
      "step": 1425
    },
    {
      "entropy": 0.9894003225583583,
      "epoch": 0.9885801943071416,
      "grad_norm": 0.10595703125,
      "learning_rate": 7.906027726981568e-08,
      "loss": 0.9240727233886719,
      "mean_token_accuracy": 0.786658907160163,
      "num_tokens": 7649672.0,
      "step": 1450
    }
  ],
  "logging_steps": 25,
  "max_steps": 1467,
  "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": 3.3015596632508314e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}