{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 15,
  "global_step": 440,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.5958565310637156,
      "epoch": 0.034135518006485746,
      "grad_norm": 0.9453125,
      "learning_rate": 4.998932514657232e-05,
      "loss": 0.5431194941202799,
      "mean_token_accuracy": 0.8273607390125592,
      "num_tokens": 9493217.0,
      "step": 15
    },
    {
      "epoch": 0.034135518006485746,
      "eval_entropy": 0.5626228135913166,
      "eval_loss": 0.5182297825813293,
      "eval_mean_token_accuracy": 0.8334106785147938,
      "eval_num_tokens": 9493217.0,
      "eval_runtime": 116.2811,
      "eval_samples_per_second": 4.61,
      "eval_steps_per_second": 2.305,
      "step": 15
    },
    {
      "entropy": 0.5794778840243816,
      "epoch": 0.06827103601297149,
      "grad_norm": 1.0703125,
      "learning_rate": 4.975951801318083e-05,
      "loss": 0.5219340960184733,
      "mean_token_accuracy": 0.8325201846162478,
      "num_tokens": 18845682.0,
      "step": 30
    },
    {
      "epoch": 0.06827103601297149,
      "eval_entropy": 0.5626906700321098,
      "eval_loss": 0.5120116472244263,
      "eval_mean_token_accuracy": 0.8348900035246095,
      "eval_num_tokens": 18845682.0,
      "eval_runtime": 116.7506,
      "eval_samples_per_second": 4.591,
      "eval_steps_per_second": 2.295,
      "step": 30
    },
    {
      "entropy": 0.5509146493176619,
      "epoch": 0.10240655401945725,
      "grad_norm": 0.78515625,
      "learning_rate": 4.923264495657319e-05,
      "loss": 0.5065197944641113,
      "mean_token_accuracy": 0.8363819032907486,
      "num_tokens": 28538625.0,
      "step": 45
    },
    {
      "epoch": 0.10240655401945725,
      "eval_entropy": 0.530716907733412,
      "eval_loss": 0.5000724196434021,
      "eval_mean_token_accuracy": 0.8370915471173045,
      "eval_num_tokens": 28538625.0,
      "eval_runtime": 116.7765,
      "eval_samples_per_second": 4.59,
      "eval_steps_per_second": 2.295,
      "step": 45
    },
    {
      "entropy": 0.5471789278586705,
      "epoch": 0.13654207202594298,
      "grad_norm": 0.68359375,
      "learning_rate": 4.8415027425773386e-05,
      "loss": 0.5041735967000326,
      "mean_token_accuracy": 0.8362820886572202,
      "num_tokens": 37904749.0,
      "step": 60
    },
    {
      "epoch": 0.13654207202594298,
      "eval_entropy": 0.5335152440551502,
      "eval_loss": 0.49730995297431946,
      "eval_mean_token_accuracy": 0.8383040488211077,
      "eval_num_tokens": 37904749.0,
      "eval_runtime": 116.8049,
      "eval_samples_per_second": 4.589,
      "eval_steps_per_second": 2.294,
      "step": 60
    },
    {
      "entropy": 0.5440506326407194,
      "epoch": 0.17067759003242874,
      "grad_norm": 0.75,
      "learning_rate": 4.7316475236404454e-05,
      "loss": 0.5034982045491536,
      "mean_token_accuracy": 0.8362049227952957,
      "num_tokens": 47572704.0,
      "step": 75
    },
    {
      "epoch": 0.17067759003242874,
      "eval_entropy": 0.5288645684051869,
      "eval_loss": 0.4952949285507202,
      "eval_mean_token_accuracy": 0.8390978541391999,
      "eval_num_tokens": 47572704.0,
      "eval_runtime": 116.8116,
      "eval_samples_per_second": 4.589,
      "eval_steps_per_second": 2.294,
      "step": 75
    },
    {
      "entropy": 0.5435298949480056,
      "epoch": 0.2048131080389145,
      "grad_norm": 0.75,
      "learning_rate": 4.5950168872034885e-05,
      "loss": 0.501694933573405,
      "mean_token_accuracy": 0.836874247888724,
      "num_tokens": 56993911.0,
      "step": 90
    },
    {
      "epoch": 0.2048131080389145,
      "eval_entropy": 0.5174737074481908,
      "eval_loss": 0.49249210953712463,
      "eval_mean_token_accuracy": 0.8395018682106218,
      "eval_num_tokens": 56993911.0,
      "eval_runtime": 116.7564,
      "eval_samples_per_second": 4.591,
      "eval_steps_per_second": 2.295,
      "step": 90
    },
    {
      "entropy": 0.5275995331257581,
      "epoch": 0.23894862604540024,
      "grad_norm": 0.74609375,
      "learning_rate": 4.433250134408401e-05,
      "loss": 0.4931917190551758,
      "mean_token_accuracy": 0.8394653130571047,
      "num_tokens": 66680612.0,
      "step": 105
    },
    {
      "epoch": 0.23894862604540024,
      "eval_entropy": 0.5392143717452661,
      "eval_loss": 0.4933963716030121,
      "eval_mean_token_accuracy": 0.8397051783195183,
      "eval_num_tokens": 66680612.0,
      "eval_runtime": 116.445,
      "eval_samples_per_second": 4.603,
      "eval_steps_per_second": 2.302,
      "step": 105
    },
    {
      "entropy": 0.5234112652142843,
      "epoch": 0.27308414405188597,
      "grad_norm": 0.67578125,
      "learning_rate": 4.248288150765925e-05,
      "loss": 0.492865784962972,
      "mean_token_accuracy": 0.839345374206702,
      "num_tokens": 76198232.0,
      "step": 120
    },
    {
      "epoch": 0.27308414405188597,
      "eval_entropy": 0.5288595106841912,
      "eval_loss": 0.4941500127315521,
      "eval_mean_token_accuracy": 0.8389813977835784,
      "eval_num_tokens": 76198232.0,
      "eval_runtime": 116.2097,
      "eval_samples_per_second": 4.612,
      "eval_steps_per_second": 2.306,
      "step": 120
    },
    {
      "entropy": 0.5145147764434417,
      "epoch": 0.30721966205837176,
      "grad_norm": 0.7109375,
      "learning_rate": 4.0423501193151416e-05,
      "loss": 0.4943487485249837,
      "mean_token_accuracy": 0.8380054174860319,
      "num_tokens": 85638789.0,
      "step": 135
    },
    {
      "epoch": 0.30721966205837176,
      "eval_entropy": 0.5192181238932396,
      "eval_loss": 0.48689964413642883,
      "eval_mean_token_accuracy": 0.8401346262266387,
      "eval_num_tokens": 85638789.0,
      "eval_runtime": 116.0494,
      "eval_samples_per_second": 4.619,
      "eval_steps_per_second": 2.309,
      "step": 135
    },
    {
      "entropy": 0.5169518480201563,
      "epoch": 0.3413551800648575,
      "grad_norm": 0.77734375,
      "learning_rate": 3.8179068947554705e-05,
      "loss": 0.4929698626200358,
      "mean_token_accuracy": 0.8384132486581802,
      "num_tokens": 95138318.0,
      "step": 150
    },
    {
      "epoch": 0.3413551800648575,
      "eval_entropy": 0.5390372970210973,
      "eval_loss": 0.490721732378006,
      "eval_mean_token_accuracy": 0.8402899370264652,
      "eval_num_tokens": 95138318.0,
      "eval_runtime": 116.1308,
      "eval_samples_per_second": 4.615,
      "eval_steps_per_second": 2.308,
      "step": 150
    },
    {
      "entropy": 0.5005746755748988,
      "epoch": 0.3754906980713432,
      "grad_norm": 0.77734375,
      "learning_rate": 3.5776513580096315e-05,
      "loss": 0.4819363911946615,
      "mean_token_accuracy": 0.8415288893381755,
      "num_tokens": 104943003.0,
      "step": 165
    },
    {
      "epoch": 0.3754906980713432,
      "eval_entropy": 0.508869458601546,
      "eval_loss": 0.4827399253845215,
      "eval_mean_token_accuracy": 0.8416397571563721,
      "eval_num_tokens": 104943003.0,
      "eval_runtime": 116.0811,
      "eval_samples_per_second": 4.617,
      "eval_steps_per_second": 2.309,
      "step": 165
    },
    {
      "entropy": 0.5018329432606697,
      "epoch": 0.409626216077829,
      "grad_norm": 0.69140625,
      "learning_rate": 3.324466106904981e-05,
      "loss": 0.4793397585550944,
      "mean_token_accuracy": 0.8424180436134339,
      "num_tokens": 114788577.0,
      "step": 180
    },
    {
      "epoch": 0.409626216077829,
      "eval_entropy": 0.5023460994794298,
      "eval_loss": 0.48032286763191223,
      "eval_mean_token_accuracy": 0.8421972496296043,
      "eval_num_tokens": 114788577.0,
      "eval_runtime": 115.7868,
      "eval_samples_per_second": 4.629,
      "eval_steps_per_second": 2.315,
      "step": 180
    },
    {
      "entropy": 0.5232079241176446,
      "epoch": 0.44376173408431474,
      "grad_norm": 0.70703125,
      "learning_rate": 3.0613888706220336e-05,
      "loss": 0.4860788981119792,
      "mean_token_accuracy": 0.8409416891137759,
      "num_tokens": 124143129.0,
      "step": 195
    },
    {
      "epoch": 0.44376173408431474,
      "eval_entropy": 0.49868492315064616,
      "eval_loss": 0.4805947542190552,
      "eval_mean_token_accuracy": 0.8426268127427172,
      "eval_num_tokens": 124143129.0,
      "eval_runtime": 115.6825,
      "eval_samples_per_second": 4.633,
      "eval_steps_per_second": 2.317,
      "step": 195
    },
    {
      "entropy": 0.49873940070470174,
      "epoch": 0.4778972520908005,
      "grad_norm": 0.671875,
      "learning_rate": 2.7915760628693256e-05,
      "loss": 0.47950067520141604,
      "mean_token_accuracy": 0.8420795334378878,
      "num_tokens": 133487337.0,
      "step": 210
    },
    {
      "epoch": 0.4778972520908005,
      "eval_entropy": 0.504599145210501,
      "eval_loss": 0.4775504469871521,
      "eval_mean_token_accuracy": 0.8426278415900558,
      "eval_num_tokens": 133487337.0,
      "eval_runtime": 116.2085,
      "eval_samples_per_second": 4.612,
      "eval_steps_per_second": 2.306,
      "step": 210
    },
    {
      "entropy": 0.5007210767020782,
      "epoch": 0.5120327700972862,
      "grad_norm": 0.58203125,
      "learning_rate": 2.5182649110754324e-05,
      "loss": 0.4816947301228841,
      "mean_token_accuracy": 0.8411219593882561,
      "num_tokens": 143380026.0,
      "step": 225
    },
    {
      "epoch": 0.5120327700972862,
      "eval_entropy": 0.4996034526335659,
      "eval_loss": 0.4734613299369812,
      "eval_mean_token_accuracy": 0.8439789961523084,
      "eval_num_tokens": 143380026.0,
      "eval_runtime": 115.8579,
      "eval_samples_per_second": 4.626,
      "eval_steps_per_second": 2.313,
      "step": 225
    },
    {
      "entropy": 0.5038241323580345,
      "epoch": 0.5461682881037719,
      "grad_norm": 0.66796875,
      "learning_rate": 2.2447346159739772e-05,
      "loss": 0.47998129526774086,
      "mean_token_accuracy": 0.8417085768779119,
      "num_tokens": 152823284.0,
      "step": 240
    },
    {
      "epoch": 0.5461682881037719,
      "eval_entropy": 0.5027493052740595,
      "eval_loss": 0.47307440638542175,
      "eval_mean_token_accuracy": 0.8440539672303555,
      "eval_num_tokens": 152823284.0,
      "eval_runtime": 115.8381,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 2.314,
      "step": 240
    },
    {
      "entropy": 0.5137605111300946,
      "epoch": 0.5803038061102577,
      "grad_norm": 0.7109375,
      "learning_rate": 1.974267007590835e-05,
      "loss": 0.467722225189209,
      "mean_token_accuracy": 0.8469271834691365,
      "num_tokens": 162394249.0,
      "step": 255
    },
    {
      "epoch": 0.5803038061102577,
      "eval_entropy": 0.5124658280566557,
      "eval_loss": 0.47308412194252014,
      "eval_mean_token_accuracy": 0.8451244248383081,
      "eval_num_tokens": 162394249.0,
      "eval_runtime": 115.8497,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 2.313,
      "step": 255
    },
    {
      "entropy": 0.48645271569490434,
      "epoch": 0.6144393241167435,
      "grad_norm": 0.62109375,
      "learning_rate": 1.710107169684972e-05,
      "loss": 0.46386197408040364,
      "mean_token_accuracy": 0.8466676217317581,
      "num_tokens": 171964006.0,
      "step": 270
    },
    {
      "epoch": 0.6144393241167435,
      "eval_entropy": 0.4833642232662706,
      "eval_loss": 0.46884289383888245,
      "eval_mean_token_accuracy": 0.8448901060801833,
      "eval_num_tokens": 171964006.0,
      "eval_runtime": 116.0004,
      "eval_samples_per_second": 4.621,
      "eval_steps_per_second": 2.31,
      "step": 270
    },
    {
      "entropy": 0.4950397784014543,
      "epoch": 0.6485748421232292,
      "grad_norm": 0.625,
      "learning_rate": 1.4554245050728085e-05,
      "loss": 0.4757593790690104,
      "mean_token_accuracy": 0.8425892086823782,
      "num_tokens": 181647775.0,
      "step": 285
    },
    {
      "epoch": 0.6485748421232292,
      "eval_entropy": 0.49345827452949625,
      "eval_loss": 0.4664832353591919,
      "eval_mean_token_accuracy": 0.8457899912079768,
      "eval_num_tokens": 181647775.0,
      "eval_runtime": 115.9239,
      "eval_samples_per_second": 4.624,
      "eval_steps_per_second": 2.312,
      "step": 285
    },
    {
      "entropy": 0.5087025686850151,
      "epoch": 0.682710360129715,
      "grad_norm": 0.56640625,
      "learning_rate": 1.213274708976271e-05,
      "loss": 0.4791340510050456,
      "mean_token_accuracy": 0.8421124185125033,
      "num_tokens": 190956252.0,
      "step": 300
    },
    {
      "epoch": 0.682710360129715,
      "eval_entropy": 0.4901320324904883,
      "eval_loss": 0.4655348062515259,
      "eval_mean_token_accuracy": 0.8461532561636683,
      "eval_num_tokens": 190956252.0,
      "eval_runtime": 116.1839,
      "eval_samples_per_second": 4.613,
      "eval_steps_per_second": 2.307,
      "step": 300
    },
    {
      "entropy": 0.4906736594935258,
      "epoch": 0.7168458781362007,
      "grad_norm": 0.58203125,
      "learning_rate": 9.865631066402137e-06,
      "loss": 0.46257413228352867,
      "mean_token_accuracy": 0.8474820895989736,
      "num_tokens": 200197707.0,
      "step": 315
    },
    {
      "epoch": 0.7168458781362007,
      "eval_entropy": 0.4901099623139225,
      "eval_loss": 0.46475765109062195,
      "eval_mean_token_accuracy": 0.8463392286603131,
      "eval_num_tokens": 200197707.0,
      "eval_runtime": 115.7593,
      "eval_samples_per_second": 4.63,
      "eval_steps_per_second": 2.315,
      "step": 315
    },
    {
      "entropy": 0.49689885343114537,
      "epoch": 0.7509813961426864,
      "grad_norm": 0.5234375,
      "learning_rate": 7.780097950962447e-06,
      "loss": 0.4677156766255697,
      "mean_token_accuracy": 0.845871776342392,
      "num_tokens": 209679362.0,
      "step": 330
    },
    {
      "epoch": 0.7509813961426864,
      "eval_entropy": 0.4896693821273633,
      "eval_loss": 0.4636322855949402,
      "eval_mean_token_accuracy": 0.8466639899051012,
      "eval_num_tokens": 209679362.0,
      "eval_runtime": 116.1007,
      "eval_samples_per_second": 4.617,
      "eval_steps_per_second": 2.308,
      "step": 330
    },
    {
      "entropy": 0.4864589220285416,
      "epoch": 0.7851169141491722,
      "grad_norm": 0.63671875,
      "learning_rate": 5.901170073038523e-06,
      "loss": 0.4569451649983724,
      "mean_token_accuracy": 0.8488482243816058,
      "num_tokens": 219483715.0,
      "step": 345
    },
    {
      "epoch": 0.7851169141491722,
      "eval_entropy": 0.5090843094373817,
      "eval_loss": 0.4672224521636963,
      "eval_mean_token_accuracy": 0.8463522233624956,
      "eval_num_tokens": 219483715.0,
      "eval_runtime": 115.7638,
      "eval_samples_per_second": 4.63,
      "eval_steps_per_second": 2.315,
      "step": 345
    },
    {
      "entropy": 0.5153097993135453,
      "epoch": 0.819252432155658,
      "grad_norm": 0.65625,
      "learning_rate": 4.251390902354413e-06,
      "loss": 0.4748752911885579,
      "mean_token_accuracy": 0.8440132855375608,
      "num_tokens": 229011040.0,
      "step": 360
    },
    {
      "epoch": 0.819252432155658,
      "eval_entropy": 0.49651382073982436,
      "eval_loss": 0.46521133184432983,
      "eval_mean_token_accuracy": 0.8467047855035582,
      "eval_num_tokens": 229011040.0,
      "eval_runtime": 115.9049,
      "eval_samples_per_second": 4.624,
      "eval_steps_per_second": 2.312,
      "step": 360
    },
    {
      "entropy": 0.4948208748052518,
      "epoch": 0.8533879501621437,
      "grad_norm": 0.5234375,
      "learning_rate": 2.850554571097211e-06,
      "loss": 0.46651153564453124,
      "mean_token_accuracy": 0.8460746076703072,
      "num_tokens": 238591135.0,
      "step": 375
    },
    {
      "epoch": 0.8533879501621437,
      "eval_entropy": 0.49041391864641387,
      "eval_loss": 0.4638001322746277,
      "eval_mean_token_accuracy": 0.8467907998988877,
      "eval_num_tokens": 238591135.0,
      "eval_runtime": 115.8565,
      "eval_samples_per_second": 4.626,
      "eval_steps_per_second": 2.313,
      "step": 375
    },
    {
      "entropy": 0.49075523984928926,
      "epoch": 0.8875234681686295,
      "grad_norm": 0.53515625,
      "learning_rate": 1.7154683829389284e-06,
      "loss": 0.4624965667724609,
      "mean_token_accuracy": 0.847155034840107,
      "num_tokens": 248104183.0,
      "step": 390
    },
    {
      "epoch": 0.8875234681686295,
      "eval_entropy": 0.489827108805749,
      "eval_loss": 0.46358540654182434,
      "eval_mean_token_accuracy": 0.8467733839999384,
      "eval_num_tokens": 248104183.0,
      "eval_runtime": 115.9141,
      "eval_samples_per_second": 4.624,
      "eval_steps_per_second": 2.312,
      "step": 390
    },
    {
      "entropy": 0.48882623645166556,
      "epoch": 0.9216589861751152,
      "grad_norm": 0.5078125,
      "learning_rate": 8.597511581746626e-07,
      "loss": 0.46177399953206383,
      "mean_token_accuracy": 0.8472742224733035,
      "num_tokens": 257630164.0,
      "step": 405
    },
    {
      "epoch": 0.9216589861751152,
      "eval_entropy": 0.4897004203565085,
      "eval_loss": 0.4635615646839142,
      "eval_mean_token_accuracy": 0.8467840510072993,
      "eval_num_tokens": 257630164.0,
      "eval_runtime": 115.8871,
      "eval_samples_per_second": 4.625,
      "eval_steps_per_second": 2.313,
      "step": 405
    },
    {
      "entropy": 0.5017094193398952,
      "epoch": 0.955794504181601,
      "grad_norm": 0.5546875,
      "learning_rate": 2.936698344423505e-07,
      "loss": 0.4753507296244303,
      "mean_token_accuracy": 0.8430813058217367,
      "num_tokens": 267307871.0,
      "step": 420
    },
    {
      "epoch": 0.955794504181601,
      "eval_entropy": 0.48964151384225535,
      "eval_loss": 0.46354392170906067,
      "eval_mean_token_accuracy": 0.8468164723755708,
      "eval_num_tokens": 267307871.0,
      "eval_runtime": 115.8865,
      "eval_samples_per_second": 4.625,
      "eval_steps_per_second": 2.313,
      "step": 420
    },
    {
      "entropy": 0.4930376689632734,
      "epoch": 0.9899300221880867,
      "grad_norm": 0.5078125,
      "learning_rate": 2.4016283496544613e-08,
      "loss": 0.4677823702494303,
      "mean_token_accuracy": 0.8451833475629489,
      "num_tokens": 277123336.0,
      "step": 435
    },
    {
      "epoch": 0.9899300221880867,
      "eval_entropy": 0.48959980849454654,
      "eval_loss": 0.4635278582572937,
      "eval_mean_token_accuracy": 0.8468158480836384,
      "eval_num_tokens": 277123336.0,
      "eval_runtime": 115.8283,
      "eval_samples_per_second": 4.628,
      "eval_steps_per_second": 2.314,
      "step": 435
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.48961994735829867,
      "eval_loss": 0.4635235071182251,
      "eval_mean_token_accuracy": 0.8468412409967451,
      "eval_num_tokens": 279969617.0,
      "eval_runtime": 115.9238,
      "eval_samples_per_second": 4.624,
      "eval_steps_per_second": 2.312,
      "step": 440
    }
  ],
  "logging_steps": 15,
  "max_steps": 440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.904154761322397e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}