{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 951,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.031545741324921134,
      "grad_norm": 4.99587251526554,
      "learning_rate": 5e-06,
      "loss": 0.7944,
      "step": 10
    },
    {
      "epoch": 0.06309148264984227,
      "grad_norm": 4.271464380765346,
      "learning_rate": 5e-06,
      "loss": 0.6536,
      "step": 20
    },
    {
      "epoch": 0.0946372239747634,
      "grad_norm": 1.1461490048117564,
      "learning_rate": 5e-06,
      "loss": 0.6254,
      "step": 30
    },
    {
      "epoch": 0.12618296529968454,
      "grad_norm": 0.981971583786276,
      "learning_rate": 5e-06,
      "loss": 0.6012,
      "step": 40
    },
    {
      "epoch": 0.15772870662460567,
      "grad_norm": 0.9522065058873813,
      "learning_rate": 5e-06,
      "loss": 0.573,
      "step": 50
    },
    {
      "epoch": 0.1892744479495268,
      "grad_norm": 0.8622864722897896,
      "learning_rate": 5e-06,
      "loss": 0.5621,
      "step": 60
    },
    {
      "epoch": 0.22082018927444794,
      "grad_norm": 0.6621715224692672,
      "learning_rate": 5e-06,
      "loss": 0.5434,
      "step": 70
    },
    {
      "epoch": 0.25236593059936907,
      "grad_norm": 0.5858137328472353,
      "learning_rate": 5e-06,
      "loss": 0.5307,
      "step": 80
    },
    {
      "epoch": 0.28391167192429023,
      "grad_norm": 0.6162490995497564,
      "learning_rate": 5e-06,
      "loss": 0.5299,
      "step": 90
    },
    {
      "epoch": 0.31545741324921134,
      "grad_norm": 0.7263871575915639,
      "learning_rate": 5e-06,
      "loss": 0.5234,
      "step": 100
    },
    {
      "epoch": 0.3470031545741325,
      "grad_norm": 0.7173526345718089,
      "learning_rate": 5e-06,
      "loss": 0.5149,
      "step": 110
    },
    {
      "epoch": 0.3785488958990536,
      "grad_norm": 0.5402852521407366,
      "learning_rate": 5e-06,
      "loss": 0.5158,
      "step": 120
    },
    {
      "epoch": 0.41009463722397477,
      "grad_norm": 0.5272811866546335,
      "learning_rate": 5e-06,
      "loss": 0.5101,
      "step": 130
    },
    {
      "epoch": 0.4416403785488959,
      "grad_norm": 0.8081780735356934,
      "learning_rate": 5e-06,
      "loss": 0.5067,
      "step": 140
    },
    {
      "epoch": 0.47318611987381703,
      "grad_norm": 0.7298094681047617,
      "learning_rate": 5e-06,
      "loss": 0.5042,
      "step": 150
    },
    {
      "epoch": 0.5047318611987381,
      "grad_norm": 0.5246448731879264,
      "learning_rate": 5e-06,
      "loss": 0.5022,
      "step": 160
    },
    {
      "epoch": 0.5362776025236593,
      "grad_norm": 0.642083232942298,
      "learning_rate": 5e-06,
      "loss": 0.5002,
      "step": 170
    },
    {
      "epoch": 0.5678233438485805,
      "grad_norm": 0.6059370520825695,
      "learning_rate": 5e-06,
      "loss": 0.4969,
      "step": 180
    },
    {
      "epoch": 0.5993690851735016,
      "grad_norm": 0.48161916454770276,
      "learning_rate": 5e-06,
      "loss": 0.4953,
      "step": 190
    },
    {
      "epoch": 0.6309148264984227,
      "grad_norm": 0.6810369029107554,
      "learning_rate": 5e-06,
      "loss": 0.4893,
      "step": 200
    },
    {
      "epoch": 0.6624605678233438,
      "grad_norm": 0.5503177401173632,
      "learning_rate": 5e-06,
      "loss": 0.4897,
      "step": 210
    },
    {
      "epoch": 0.694006309148265,
      "grad_norm": 0.6046372979618468,
      "learning_rate": 5e-06,
      "loss": 0.4874,
      "step": 220
    },
    {
      "epoch": 0.7255520504731862,
      "grad_norm": 0.5662755567010622,
      "learning_rate": 5e-06,
      "loss": 0.4846,
      "step": 230
    },
    {
      "epoch": 0.7570977917981072,
      "grad_norm": 0.4846646736673193,
      "learning_rate": 5e-06,
      "loss": 0.4863,
      "step": 240
    },
    {
      "epoch": 0.7886435331230284,
      "grad_norm": 0.4875951863435466,
      "learning_rate": 5e-06,
      "loss": 0.4909,
      "step": 250
    },
    {
      "epoch": 0.8201892744479495,
      "grad_norm": 0.5146296136980277,
      "learning_rate": 5e-06,
      "loss": 0.4826,
      "step": 260
    },
    {
      "epoch": 0.8517350157728707,
      "grad_norm": 0.4654893350187648,
      "learning_rate": 5e-06,
      "loss": 0.4722,
      "step": 270
    },
    {
      "epoch": 0.8832807570977917,
      "grad_norm": 0.4681820584043545,
      "learning_rate": 5e-06,
      "loss": 0.4787,
      "step": 280
    },
    {
      "epoch": 0.9148264984227129,
      "grad_norm": 0.4735053003732251,
      "learning_rate": 5e-06,
      "loss": 0.4769,
      "step": 290
    },
    {
      "epoch": 0.9463722397476341,
      "grad_norm": 0.5851272182332038,
      "learning_rate": 5e-06,
      "loss": 0.479,
      "step": 300
    },
    {
      "epoch": 0.9779179810725552,
      "grad_norm": 0.6125017656961282,
      "learning_rate": 5e-06,
      "loss": 0.4745,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.47562769055366516,
      "eval_runtime": 170.473,
      "eval_samples_per_second": 50.084,
      "eval_steps_per_second": 0.393,
      "step": 317
    },
    {
      "epoch": 1.0094637223974763,
      "grad_norm": 0.7703225863849302,
      "learning_rate": 5e-06,
      "loss": 0.4658,
      "step": 320
    },
    {
      "epoch": 1.0410094637223974,
      "grad_norm": 0.5658238656624024,
      "learning_rate": 5e-06,
      "loss": 0.4369,
      "step": 330
    },
    {
      "epoch": 1.0725552050473186,
      "grad_norm": 0.6051214020937586,
      "learning_rate": 5e-06,
      "loss": 0.4401,
      "step": 340
    },
    {
      "epoch": 1.1041009463722398,
      "grad_norm": 0.7172129607524986,
      "learning_rate": 5e-06,
      "loss": 0.4359,
      "step": 350
    },
    {
      "epoch": 1.135646687697161,
      "grad_norm": 0.5578022584419556,
      "learning_rate": 5e-06,
      "loss": 0.4349,
      "step": 360
    },
    {
      "epoch": 1.167192429022082,
      "grad_norm": 0.49543679997471873,
      "learning_rate": 5e-06,
      "loss": 0.4394,
      "step": 370
    },
    {
      "epoch": 1.1987381703470033,
      "grad_norm": 0.6068336570614361,
      "learning_rate": 5e-06,
      "loss": 0.4383,
      "step": 380
    },
    {
      "epoch": 1.2302839116719242,
      "grad_norm": 0.5054022564611206,
      "learning_rate": 5e-06,
      "loss": 0.4311,
      "step": 390
    },
    {
      "epoch": 1.2618296529968454,
      "grad_norm": 0.5339101451875964,
      "learning_rate": 5e-06,
      "loss": 0.4313,
      "step": 400
    },
    {
      "epoch": 1.2933753943217665,
      "grad_norm": 0.4854999161353839,
      "learning_rate": 5e-06,
      "loss": 0.4337,
      "step": 410
    },
    {
      "epoch": 1.3249211356466877,
      "grad_norm": 0.4776808074651913,
      "learning_rate": 5e-06,
      "loss": 0.4284,
      "step": 420
    },
    {
      "epoch": 1.3564668769716088,
      "grad_norm": 0.571848423035293,
      "learning_rate": 5e-06,
      "loss": 0.4333,
      "step": 430
    },
    {
      "epoch": 1.38801261829653,
      "grad_norm": 0.5673583129436841,
      "learning_rate": 5e-06,
      "loss": 0.4323,
      "step": 440
    },
    {
      "epoch": 1.4195583596214512,
      "grad_norm": 0.5703308592011652,
      "learning_rate": 5e-06,
      "loss": 0.4284,
      "step": 450
    },
    {
      "epoch": 1.4511041009463723,
      "grad_norm": 0.5235306886956671,
      "learning_rate": 5e-06,
      "loss": 0.4285,
      "step": 460
    },
    {
      "epoch": 1.4826498422712935,
      "grad_norm": 0.5621748063203227,
      "learning_rate": 5e-06,
      "loss": 0.4256,
      "step": 470
    },
    {
      "epoch": 1.5141955835962144,
      "grad_norm": 0.5709758620722418,
      "learning_rate": 5e-06,
      "loss": 0.4229,
      "step": 480
    },
    {
      "epoch": 1.5457413249211358,
      "grad_norm": 0.7440083314774995,
      "learning_rate": 5e-06,
      "loss": 0.4223,
      "step": 490
    },
    {
      "epoch": 1.5772870662460567,
      "grad_norm": 0.5274171243956395,
      "learning_rate": 5e-06,
      "loss": 0.4203,
      "step": 500
    },
    {
      "epoch": 1.608832807570978,
      "grad_norm": 0.5573889566155682,
      "learning_rate": 5e-06,
      "loss": 0.4246,
      "step": 510
    },
    {
      "epoch": 1.640378548895899,
      "grad_norm": 0.5089409762486478,
      "learning_rate": 5e-06,
      "loss": 0.422,
      "step": 520
    },
    {
      "epoch": 1.6719242902208202,
      "grad_norm": 0.5487128479946805,
      "learning_rate": 5e-06,
      "loss": 0.4327,
      "step": 530
    },
    {
      "epoch": 1.7034700315457414,
      "grad_norm": 0.6447340995233556,
      "learning_rate": 5e-06,
      "loss": 0.4223,
      "step": 540
    },
    {
      "epoch": 1.7350157728706623,
      "grad_norm": 0.4778520303774798,
      "learning_rate": 5e-06,
      "loss": 0.4249,
      "step": 550
    },
    {
      "epoch": 1.7665615141955837,
      "grad_norm": 0.5277561759622408,
      "learning_rate": 5e-06,
      "loss": 0.4231,
      "step": 560
    },
    {
      "epoch": 1.7981072555205047,
      "grad_norm": 0.507228211837066,
      "learning_rate": 5e-06,
      "loss": 0.4208,
      "step": 570
    },
    {
      "epoch": 1.8296529968454258,
      "grad_norm": 0.573778603365111,
      "learning_rate": 5e-06,
      "loss": 0.4246,
      "step": 580
    },
    {
      "epoch": 1.861198738170347,
      "grad_norm": 0.554076788906782,
      "learning_rate": 5e-06,
      "loss": 0.4178,
      "step": 590
    },
    {
      "epoch": 1.8927444794952681,
      "grad_norm": 0.48057226381097573,
      "learning_rate": 5e-06,
      "loss": 0.4225,
      "step": 600
    },
    {
      "epoch": 1.9242902208201893,
      "grad_norm": 0.5626542258956102,
      "learning_rate": 5e-06,
      "loss": 0.4241,
      "step": 610
    },
    {
      "epoch": 1.9558359621451105,
      "grad_norm": 0.49935383563605856,
      "learning_rate": 5e-06,
      "loss": 0.4199,
      "step": 620
    },
    {
      "epoch": 1.9873817034700316,
      "grad_norm": 0.5226498960785609,
      "learning_rate": 5e-06,
      "loss": 0.4181,
      "step": 630
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.44654515385627747,
      "eval_runtime": 170.7254,
      "eval_samples_per_second": 50.01,
      "eval_steps_per_second": 0.392,
      "step": 634
    },
    {
      "epoch": 2.0189274447949526,
      "grad_norm": 0.6088532387627132,
      "learning_rate": 5e-06,
      "loss": 0.3953,
      "step": 640
    },
    {
      "epoch": 2.050473186119874,
      "grad_norm": 0.6170163085689537,
      "learning_rate": 5e-06,
      "loss": 0.38,
      "step": 650
    },
    {
      "epoch": 2.082018927444795,
      "grad_norm": 0.6044774577568809,
      "learning_rate": 5e-06,
      "loss": 0.3821,
      "step": 660
    },
    {
      "epoch": 2.1135646687697163,
      "grad_norm": 0.7001653617975729,
      "learning_rate": 5e-06,
      "loss": 0.386,
      "step": 670
    },
    {
      "epoch": 2.145110410094637,
      "grad_norm": 0.5560077663575217,
      "learning_rate": 5e-06,
      "loss": 0.3787,
      "step": 680
    },
    {
      "epoch": 2.176656151419558,
      "grad_norm": 0.5213047969133517,
      "learning_rate": 5e-06,
      "loss": 0.3838,
      "step": 690
    },
    {
      "epoch": 2.2082018927444795,
      "grad_norm": 0.5147380409958507,
      "learning_rate": 5e-06,
      "loss": 0.3851,
      "step": 700
    },
    {
      "epoch": 2.2397476340694005,
      "grad_norm": 0.5497461937255633,
      "learning_rate": 5e-06,
      "loss": 0.3864,
      "step": 710
    },
    {
      "epoch": 2.271293375394322,
      "grad_norm": 0.620646304852149,
      "learning_rate": 5e-06,
      "loss": 0.3814,
      "step": 720
    },
    {
      "epoch": 2.302839116719243,
      "grad_norm": 0.6036500927422149,
      "learning_rate": 5e-06,
      "loss": 0.3776,
      "step": 730
    },
    {
      "epoch": 2.334384858044164,
      "grad_norm": 0.5481492981732653,
      "learning_rate": 5e-06,
      "loss": 0.3858,
      "step": 740
    },
    {
      "epoch": 2.365930599369085,
      "grad_norm": 0.5654790601381019,
      "learning_rate": 5e-06,
      "loss": 0.3824,
      "step": 750
    },
    {
      "epoch": 2.3974763406940065,
      "grad_norm": 0.5811999163309192,
      "learning_rate": 5e-06,
      "loss": 0.3831,
      "step": 760
    },
    {
      "epoch": 2.4290220820189274,
      "grad_norm": 0.5244470729272699,
      "learning_rate": 5e-06,
      "loss": 0.3776,
      "step": 770
    },
    {
      "epoch": 2.4605678233438484,
      "grad_norm": 0.48851134130848284,
      "learning_rate": 5e-06,
      "loss": 0.3791,
      "step": 780
    },
    {
      "epoch": 2.4921135646687698,
      "grad_norm": 0.5371797697436069,
      "learning_rate": 5e-06,
      "loss": 0.3797,
      "step": 790
    },
    {
      "epoch": 2.5236593059936907,
      "grad_norm": 0.49066553328169765,
      "learning_rate": 5e-06,
      "loss": 0.3815,
      "step": 800
    },
    {
      "epoch": 2.555205047318612,
      "grad_norm": 0.5984827997828263,
      "learning_rate": 5e-06,
      "loss": 0.3756,
      "step": 810
    },
    {
      "epoch": 2.586750788643533,
      "grad_norm": 0.6179347432660328,
      "learning_rate": 5e-06,
      "loss": 0.3806,
      "step": 820
    },
    {
      "epoch": 2.6182965299684544,
      "grad_norm": 0.6394278462521181,
      "learning_rate": 5e-06,
      "loss": 0.3834,
      "step": 830
    },
    {
      "epoch": 2.6498422712933754,
      "grad_norm": 0.579789988236478,
      "learning_rate": 5e-06,
      "loss": 0.3869,
      "step": 840
    },
    {
      "epoch": 2.6813880126182967,
      "grad_norm": 0.5366527627741544,
      "learning_rate": 5e-06,
      "loss": 0.3848,
      "step": 850
    },
    {
      "epoch": 2.7129337539432177,
      "grad_norm": 0.6067239997780901,
      "learning_rate": 5e-06,
      "loss": 0.3768,
      "step": 860
    },
    {
      "epoch": 2.7444794952681386,
      "grad_norm": 0.5441160362260907,
      "learning_rate": 5e-06,
      "loss": 0.3883,
      "step": 870
    },
    {
      "epoch": 2.77602523659306,
      "grad_norm": 0.5231719514823525,
      "learning_rate": 5e-06,
      "loss": 0.3795,
      "step": 880
    },
    {
      "epoch": 2.807570977917981,
      "grad_norm": 0.6141826745417515,
      "learning_rate": 5e-06,
      "loss": 0.3771,
      "step": 890
    },
    {
      "epoch": 2.8391167192429023,
      "grad_norm": 0.542559242674366,
      "learning_rate": 5e-06,
      "loss": 0.3782,
      "step": 900
    },
    {
      "epoch": 2.8706624605678233,
      "grad_norm": 0.5043528521109601,
      "learning_rate": 5e-06,
      "loss": 0.3779,
      "step": 910
    },
    {
      "epoch": 2.9022082018927446,
      "grad_norm": 0.5168292506779991,
      "learning_rate": 5e-06,
      "loss": 0.3782,
      "step": 920
    },
    {
      "epoch": 2.9337539432176656,
      "grad_norm": 0.556675998633135,
      "learning_rate": 5e-06,
      "loss": 0.3802,
      "step": 930
    },
    {
      "epoch": 2.965299684542587,
      "grad_norm": 0.5534284489693292,
      "learning_rate": 5e-06,
      "loss": 0.3806,
      "step": 940
    },
    {
      "epoch": 2.996845425867508,
      "grad_norm": 0.5746396382994269,
      "learning_rate": 5e-06,
      "loss": 0.3794,
      "step": 950
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4385061264038086,
      "eval_runtime": 170.7773,
      "eval_samples_per_second": 49.995,
      "eval_steps_per_second": 0.392,
      "step": 951
    },
    {
      "epoch": 3.0,
      "step": 951,
      "total_flos": 1592541661102080.0,
      "train_loss": 0.4442997165220142,
      "train_runtime": 28675.07,
      "train_samples_per_second": 16.971,
      "train_steps_per_second": 0.033
    }
  ],
  "logging_steps": 10,
  "max_steps": 951,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1592541661102080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}