{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 50,
  "global_step": 765,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06568144499178982,
      "grad_norm": 23.778915405273438,
      "learning_rate": 1.1688311688311688e-05,
      "loss": 11.1593,
      "step": 10
    },
    {
      "epoch": 0.13136288998357964,
      "grad_norm": 16.608760833740234,
      "learning_rate": 2.4675324675324678e-05,
      "loss": 8.0129,
      "step": 20
    },
    {
      "epoch": 0.19704433497536947,
      "grad_norm": 1.3987914323806763,
      "learning_rate": 3.7662337662337665e-05,
      "loss": 1.6387,
      "step": 30
    },
    {
      "epoch": 0.2627257799671593,
      "grad_norm": 1.8992557525634766,
      "learning_rate": 5.064935064935065e-05,
      "loss": 0.4754,
      "step": 40
    },
    {
      "epoch": 0.3284072249589491,
      "grad_norm": 0.8001347780227661,
      "learning_rate": 6.363636363636364e-05,
      "loss": 0.4542,
      "step": 50
    },
    {
      "epoch": 0.3284072249589491,
      "eval_loss": 0.4499332308769226,
      "eval_runtime": 758.3241,
      "eval_samples_per_second": 20.188,
      "eval_steps_per_second": 0.316,
      "step": 50
    },
    {
      "epoch": 0.39408866995073893,
      "grad_norm": 0.47190120816230774,
      "learning_rate": 7.662337662337662e-05,
      "loss": 0.4372,
      "step": 60
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.5638025403022766,
      "learning_rate": 8.961038961038961e-05,
      "loss": 0.4289,
      "step": 70
    },
    {
      "epoch": 0.5254515599343186,
      "grad_norm": 0.4038947522640228,
      "learning_rate": 9.999791493378921e-05,
      "loss": 0.4059,
      "step": 80
    },
    {
      "epoch": 0.5911330049261084,
      "grad_norm": 0.17405395209789276,
      "learning_rate": 9.992495587429129e-05,
      "loss": 0.4065,
      "step": 90
    },
    {
      "epoch": 0.6568144499178982,
      "grad_norm": 0.2470746487379074,
      "learning_rate": 9.97479173391143e-05,
      "loss": 0.392,
      "step": 100
    },
    {
      "epoch": 0.6568144499178982,
      "eval_loss": 0.3926284909248352,
      "eval_runtime": 757.2757,
      "eval_samples_per_second": 20.216,
      "eval_steps_per_second": 0.317,
      "step": 100
    },
    {
      "epoch": 0.722495894909688,
      "grad_norm": 0.3697839081287384,
      "learning_rate": 9.946716840375551e-05,
      "loss": 0.4104,
      "step": 110
    },
    {
      "epoch": 0.7881773399014779,
      "grad_norm": 0.4284563362598419,
      "learning_rate": 9.908329435069495e-05,
      "loss": 0.4002,
      "step": 120
    },
    {
      "epoch": 0.8538587848932676,
      "grad_norm": 0.23430955410003662,
      "learning_rate": 9.859709544924624e-05,
      "loss": 0.3943,
      "step": 130
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.2440187633037567,
      "learning_rate": 9.800958528722036e-05,
      "loss": 0.3891,
      "step": 140
    },
    {
      "epoch": 0.9852216748768473,
      "grad_norm": 0.1979193389415741,
      "learning_rate": 9.732198865788047e-05,
      "loss": 0.3933,
      "step": 150
    },
    {
      "epoch": 0.9852216748768473,
      "eval_loss": 0.3885968029499054,
      "eval_runtime": 757.125,
      "eval_samples_per_second": 20.22,
      "eval_steps_per_second": 0.317,
      "step": 150
    },
    {
      "epoch": 1.0459770114942528,
      "grad_norm": 0.2124597281217575,
      "learning_rate": 9.653573900659292e-05,
      "loss": 0.3817,
      "step": 160
    },
    {
      "epoch": 1.1116584564860428,
      "grad_norm": 0.14501869678497314,
      "learning_rate": 9.56524754424971e-05,
      "loss": 0.3808,
      "step": 170
    },
    {
      "epoch": 1.1773399014778325,
      "grad_norm": 0.17397305369377136,
      "learning_rate": 9.467403932142452e-05,
      "loss": 0.3836,
      "step": 180
    },
    {
      "epoch": 1.2430213464696223,
      "grad_norm": 0.2558547556400299,
      "learning_rate": 9.360247040719039e-05,
      "loss": 0.3953,
      "step": 190
    },
    {
      "epoch": 1.3087027914614122,
      "grad_norm": 0.2716054618358612,
      "learning_rate": 9.24400026192603e-05,
      "loss": 0.3696,
      "step": 200
    },
    {
      "epoch": 1.3087027914614122,
      "eval_loss": 0.38374823331832886,
      "eval_runtime": 757.3043,
      "eval_samples_per_second": 20.215,
      "eval_steps_per_second": 0.317,
      "step": 200
    },
    {
      "epoch": 1.374384236453202,
      "grad_norm": 0.36536920070648193,
      "learning_rate": 9.118905937565722e-05,
      "loss": 0.3876,
      "step": 210
    },
    {
      "epoch": 1.4400656814449917,
      "grad_norm": 0.390596479177475,
      "learning_rate": 8.985224854081726e-05,
      "loss": 0.3896,
      "step": 220
    },
    {
      "epoch": 1.5057471264367817,
      "grad_norm": 0.31378212571144104,
      "learning_rate": 8.84323569889266e-05,
      "loss": 0.3921,
      "step": 230
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.21976451575756073,
      "learning_rate": 8.693234479407353e-05,
      "loss": 0.3734,
      "step": 240
    },
    {
      "epoch": 1.6371100164203614,
      "grad_norm": 0.20274367928504944,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.3711,
      "step": 250
    },
    {
      "epoch": 1.6371100164203614,
      "eval_loss": 0.3857766389846802,
      "eval_runtime": 757.2508,
      "eval_samples_per_second": 20.217,
      "eval_steps_per_second": 0.317,
      "step": 250
    },
    {
      "epoch": 1.702791461412151,
      "grad_norm": 0.3137906491756439,
      "learning_rate": 8.370462739760923e-05,
      "loss": 0.3761,
      "step": 260
    },
    {
      "epoch": 1.7684729064039408,
      "grad_norm": 0.5217385292053223,
      "learning_rate": 8.198365107794457e-05,
      "loss": 0.3783,
      "step": 270
    },
    {
      "epoch": 1.8341543513957306,
      "grad_norm": 0.21547040343284607,
      "learning_rate": 8.019599785138635e-05,
      "loss": 0.379,
      "step": 280
    },
    {
      "epoch": 1.8998357963875205,
      "grad_norm": 0.22457750141620636,
      "learning_rate": 7.834539447156424e-05,
      "loss": 0.3797,
      "step": 290
    },
    {
      "epoch": 1.9655172413793105,
      "grad_norm": 0.24211543798446655,
      "learning_rate": 7.643569892545267e-05,
      "loss": 0.3776,
      "step": 300
    },
    {
      "epoch": 1.9655172413793105,
      "eval_loss": 0.3805686831474304,
      "eval_runtime": 757.2414,
      "eval_samples_per_second": 20.217,
      "eval_steps_per_second": 0.317,
      "step": 300
    },
    {
      "epoch": 2.026272577996716,
      "grad_norm": 0.4717698097229004,
      "learning_rate": 7.447089239055428e-05,
      "loss": 0.3699,
      "step": 310
    },
    {
      "epoch": 2.0919540229885056,
      "grad_norm": 0.1722123920917511,
      "learning_rate": 7.245507093526574e-05,
      "loss": 0.376,
      "step": 320
    },
    {
      "epoch": 2.1576354679802954,
      "grad_norm": 0.23734314739704132,
      "learning_rate": 7.039243697972857e-05,
      "loss": 0.3661,
      "step": 330
    },
    {
      "epoch": 2.2233169129720856,
      "grad_norm": 0.2563398778438568,
      "learning_rate": 6.828729053496629e-05,
      "loss": 0.3654,
      "step": 340
    },
    {
      "epoch": 2.2889983579638753,
      "grad_norm": 0.4312366247177124,
      "learning_rate": 6.614402023857232e-05,
      "loss": 0.363,
      "step": 350
    },
    {
      "epoch": 2.2889983579638753,
      "eval_loss": 0.39721259474754333,
      "eval_runtime": 758.0462,
      "eval_samples_per_second": 20.195,
      "eval_steps_per_second": 0.317,
      "step": 350
    },
    {
      "epoch": 2.354679802955665,
      "grad_norm": 0.3123910427093506,
      "learning_rate": 6.39670942056362e-05,
      "loss": 0.3656,
      "step": 360
    },
    {
      "epoch": 2.4203612479474548,
      "grad_norm": 0.43575626611709595,
      "learning_rate": 6.17610507139818e-05,
      "loss": 0.3662,
      "step": 370
    },
    {
      "epoch": 2.4860426929392445,
      "grad_norm": 0.2875679135322571,
      "learning_rate": 5.9530488743135746e-05,
      "loss": 0.3648,
      "step": 380
    },
    {
      "epoch": 2.5517241379310347,
      "grad_norm": 0.2927298843860626,
      "learning_rate": 5.728005838675025e-05,
      "loss": 0.3704,
      "step": 390
    },
    {
      "epoch": 2.6174055829228244,
      "grad_norm": 0.20038872957229614,
      "learning_rate": 5.5014451158466975e-05,
      "loss": 0.3725,
      "step": 400
    },
    {
      "epoch": 2.6174055829228244,
      "eval_loss": 0.39082425832748413,
      "eval_runtime": 757.7828,
      "eval_samples_per_second": 20.202,
      "eval_steps_per_second": 0.317,
      "step": 400
    },
    {
      "epoch": 2.683087027914614,
      "grad_norm": 0.3112049996852875,
      "learning_rate": 5.273839021143218e-05,
      "loss": 0.3534,
      "step": 410
    },
    {
      "epoch": 2.748768472906404,
      "grad_norm": 0.3891284167766571,
      "learning_rate": 5.045662049185229e-05,
      "loss": 0.3662,
      "step": 420
    },
    {
      "epoch": 2.8144499178981937,
      "grad_norm": 0.25988757610321045,
      "learning_rate": 4.817389884711705e-05,
      "loss": 0.37,
      "step": 430
    },
    {
      "epoch": 2.8801313628899834,
      "grad_norm": 0.23573455214500427,
      "learning_rate": 4.589498410911215e-05,
      "loss": 0.3638,
      "step": 440
    },
    {
      "epoch": 2.945812807881773,
      "grad_norm": 0.4288446605205536,
      "learning_rate": 4.362462717339461e-05,
      "loss": 0.3638,
      "step": 450
    },
    {
      "epoch": 2.945812807881773,
      "eval_loss": 0.38178473711013794,
      "eval_runtime": 757.5923,
      "eval_samples_per_second": 20.207,
      "eval_steps_per_second": 0.317,
      "step": 450
    },
    {
      "epoch": 3.006568144499179,
      "grad_norm": 0.38512226939201355,
      "learning_rate": 4.1367561094913335e-05,
      "loss": 0.3559,
      "step": 460
    },
    {
      "epoch": 3.0722495894909687,
      "grad_norm": 0.44673025608062744,
      "learning_rate": 3.9128491220922156e-05,
      "loss": 0.345,
      "step": 470
    },
    {
      "epoch": 3.1379310344827585,
      "grad_norm": 0.36454498767852783,
      "learning_rate": 3.6912085381655734e-05,
      "loss": 0.3487,
      "step": 480
    },
    {
      "epoch": 3.2036124794745486,
      "grad_norm": 0.59967041015625,
      "learning_rate": 3.472296415921783e-05,
      "loss": 0.3549,
      "step": 490
    },
    {
      "epoch": 3.2692939244663384,
      "grad_norm": 0.3978519141674042,
      "learning_rate": 3.256569125496858e-05,
      "loss": 0.351,
      "step": 500
    },
    {
      "epoch": 3.2692939244663384,
      "eval_loss": 0.3886831998825073,
      "eval_runtime": 757.7412,
      "eval_samples_per_second": 20.203,
      "eval_steps_per_second": 0.317,
      "step": 500
    },
    {
      "epoch": 3.334975369458128,
      "grad_norm": 0.4203931987285614,
      "learning_rate": 3.0444763975492208e-05,
      "loss": 0.3495,
      "step": 510
    },
    {
      "epoch": 3.400656814449918,
      "grad_norm": 0.5428864359855652,
      "learning_rate": 2.836460385697911e-05,
      "loss": 0.3498,
      "step": 520
    },
    {
      "epoch": 3.4663382594417076,
      "grad_norm": 0.45395535230636597,
      "learning_rate": 2.6329547447567836e-05,
      "loss": 0.3479,
      "step": 530
    },
    {
      "epoch": 3.5320197044334973,
      "grad_norm": 0.5014253854751587,
      "learning_rate": 2.4343837266863246e-05,
      "loss": 0.3393,
      "step": 540
    },
    {
      "epoch": 3.5977011494252875,
      "grad_norm": 0.6780768632888794,
      "learning_rate": 2.24116129614777e-05,
      "loss": 0.3458,
      "step": 550
    },
    {
      "epoch": 3.5977011494252875,
      "eval_loss": 0.39063215255737305,
      "eval_runtime": 757.4915,
      "eval_samples_per_second": 20.21,
      "eval_steps_per_second": 0.317,
      "step": 550
    },
    {
      "epoch": 3.6633825944170773,
      "grad_norm": 0.40080639719963074,
      "learning_rate": 2.0536902675033548e-05,
      "loss": 0.3509,
      "step": 560
    },
    {
      "epoch": 3.729064039408867,
      "grad_norm": 0.38429707288742065,
      "learning_rate": 1.8723614650617723e-05,
      "loss": 0.3429,
      "step": 570
    },
    {
      "epoch": 3.7947454844006567,
      "grad_norm": 0.46875861287117004,
      "learning_rate": 1.697552908319553e-05,
      "loss": 0.3433,
      "step": 580
    },
    {
      "epoch": 3.8604269293924465,
      "grad_norm": 0.35113444924354553,
      "learning_rate": 1.5296290238968303e-05,
      "loss": 0.341,
      "step": 590
    },
    {
      "epoch": 3.9261083743842367,
      "grad_norm": 0.4000733494758606,
      "learning_rate": 1.3689398858104751e-05,
      "loss": 0.3426,
      "step": 600
    },
    {
      "epoch": 3.9261083743842367,
      "eval_loss": 0.38916587829589844,
      "eval_runtime": 757.8418,
      "eval_samples_per_second": 20.201,
      "eval_steps_per_second": 0.317,
      "step": 600
    },
    {
      "epoch": 3.9917898193760264,
      "grad_norm": 0.5007103681564331,
      "learning_rate": 1.2158204856683176e-05,
      "loss": 0.3432,
      "step": 610
    },
    {
      "epoch": 4.052545155993432,
      "grad_norm": 0.5261009931564331,
      "learning_rate": 1.0705900343059855e-05,
      "loss": 0.32,
      "step": 620
    },
    {
      "epoch": 4.1182266009852215,
      "grad_norm": 0.5950102806091309,
      "learning_rate": 9.335512963221732e-06,
      "loss": 0.3169,
      "step": 630
    },
    {
      "epoch": 4.183908045977011,
      "grad_norm": 0.8160276412963867,
      "learning_rate": 8.049899588997244e-06,
      "loss": 0.3324,
      "step": 640
    },
    {
      "epoch": 4.249589490968801,
      "grad_norm": 0.6215735673904419,
      "learning_rate": 6.851740362282788e-06,
      "loss": 0.3216,
      "step": 650
    },
    {
      "epoch": 4.249589490968801,
      "eval_loss": 0.4052900969982147,
      "eval_runtime": 757.6371,
      "eval_samples_per_second": 20.206,
      "eval_steps_per_second": 0.317,
      "step": 650
    },
    {
      "epoch": 4.315270935960591,
      "grad_norm": 0.7963327169418335,
      "learning_rate": 5.743533107701593e-06,
      "loss": 0.3218,
      "step": 660
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 0.5660962462425232,
      "learning_rate": 4.727588125342669e-06,
      "loss": 0.3213,
      "step": 670
    },
    {
      "epoch": 4.446633825944171,
      "grad_norm": 0.7453513145446777,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.326,
      "step": 680
    },
    {
      "epoch": 4.512315270935961,
      "grad_norm": 0.7786881923675537,
      "learning_rate": 2.9807600580021634e-06,
      "loss": 0.3132,
      "step": 690
    },
    {
      "epoch": 4.577996715927751,
      "grad_norm": 0.8105347156524658,
      "learning_rate": 2.253518617688377e-06,
      "loss": 0.3219,
      "step": 700
    },
    {
      "epoch": 4.577996715927751,
      "eval_loss": 0.4105878174304962,
      "eval_runtime": 757.7357,
      "eval_samples_per_second": 20.204,
      "eval_steps_per_second": 0.317,
      "step": 700
    },
    {
      "epoch": 4.64367816091954,
      "grad_norm": 0.7195111513137817,
      "learning_rate": 1.6258151471287396e-06,
      "loss": 0.3172,
      "step": 710
    },
    {
      "epoch": 4.70935960591133,
      "grad_norm": 0.8316687345504761,
      "learning_rate": 1.0989582313175374e-06,
      "loss": 0.3301,
      "step": 720
    },
    {
      "epoch": 4.77504105090312,
      "grad_norm": 0.8696787357330322,
      "learning_rate": 6.74046218577673e-07,
      "loss": 0.3231,
      "step": 730
    },
    {
      "epoch": 4.8407224958949095,
      "grad_norm": 0.7424209117889404,
      "learning_rate": 3.5196493081366967e-07,
      "loss": 0.3222,
      "step": 740
    },
    {
      "epoch": 4.906403940886699,
      "grad_norm": 0.7921685576438904,
      "learning_rate": 1.333858168224178e-07,
      "loss": 0.318,
      "step": 750
    },
    {
      "epoch": 4.906403940886699,
      "eval_loss": 0.40994852781295776,
      "eval_runtime": 757.7888,
      "eval_samples_per_second": 20.202,
      "eval_steps_per_second": 0.317,
      "step": 750
    },
    {
      "epoch": 4.972085385878489,
      "grad_norm": 0.7382909655570984,
      "learning_rate": 1.8764552511485457e-08,
      "loss": 0.3227,
      "step": 760
    },
    {
      "epoch": 5.0,
      "step": 765,
      "total_flos": 1.0568745850559791e+19,
      "train_loss": 0.6214593920053221,
      "train_runtime": 42475.8289,
      "train_samples_per_second": 4.583,
      "train_steps_per_second": 0.018
    }
  ],
  "logging_steps": 10,
  "max_steps": 765,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0568745850559791e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}