{
  "best_metric": 0.21462436020374298,
  "best_model_checkpoint": "finetuned-indian-food/checkpoint-1200",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 1332,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03003003003003003,
      "grad_norm": 2.086472749710083,
      "learning_rate": 0.0001984984984984985,
      "loss": 2.8715,
      "step": 10
    },
    {
      "epoch": 0.06006006006006006,
      "grad_norm": 2.6774790287017822,
      "learning_rate": 0.00019699699699699701,
      "loss": 2.5719,
      "step": 20
    },
    {
      "epoch": 0.09009009009009009,
      "grad_norm": 2.241564989089966,
      "learning_rate": 0.0001954954954954955,
      "loss": 2.2151,
      "step": 30
    },
    {
      "epoch": 0.12012012012012012,
      "grad_norm": 2.272333860397339,
      "learning_rate": 0.00019399399399399402,
      "loss": 2.0446,
      "step": 40
    },
    {
      "epoch": 0.15015015015015015,
      "grad_norm": 2.3084256649017334,
      "learning_rate": 0.0001924924924924925,
      "loss": 1.6935,
      "step": 50
    },
    {
      "epoch": 0.18018018018018017,
      "grad_norm": 2.6124863624572754,
      "learning_rate": 0.000190990990990991,
      "loss": 1.4559,
      "step": 60
    },
    {
      "epoch": 0.21021021021021022,
      "grad_norm": 1.7842832803726196,
      "learning_rate": 0.0001894894894894895,
      "loss": 1.2871,
      "step": 70
    },
    {
      "epoch": 0.24024024024024024,
      "grad_norm": 1.8386505842208862,
      "learning_rate": 0.000187987987987988,
      "loss": 1.244,
      "step": 80
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 3.2887415885925293,
      "learning_rate": 0.0001864864864864865,
      "loss": 1.2332,
      "step": 90
    },
    {
      "epoch": 0.3003003003003003,
      "grad_norm": 1.8092305660247803,
      "learning_rate": 0.000184984984984985,
      "loss": 1.0574,
      "step": 100
    },
    {
      "epoch": 0.3003003003003003,
      "eval_accuracy": 0.8575982996811902,
      "eval_loss": 0.9444788694381714,
      "eval_runtime": 26.2853,
      "eval_samples_per_second": 35.8,
      "eval_steps_per_second": 4.489,
      "step": 100
    },
    {
      "epoch": 0.3303303303303303,
      "grad_norm": 2.07977557182312,
      "learning_rate": 0.0001834834834834835,
      "loss": 1.0456,
      "step": 110
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 1.9540412425994873,
      "learning_rate": 0.000181981981981982,
      "loss": 1.0319,
      "step": 120
    },
    {
      "epoch": 0.39039039039039036,
      "grad_norm": 1.9397859573364258,
      "learning_rate": 0.0001804804804804805,
      "loss": 0.7043,
      "step": 130
    },
    {
      "epoch": 0.42042042042042044,
      "grad_norm": 3.5629427433013916,
      "learning_rate": 0.00017897897897897898,
      "loss": 0.8621,
      "step": 140
    },
    {
      "epoch": 0.45045045045045046,
      "grad_norm": 2.1689531803131104,
      "learning_rate": 0.0001774774774774775,
      "loss": 0.7846,
      "step": 150
    },
    {
      "epoch": 0.4804804804804805,
      "grad_norm": 5.137659549713135,
      "learning_rate": 0.000175975975975976,
      "loss": 0.8796,
      "step": 160
    },
    {
      "epoch": 0.5105105105105106,
      "grad_norm": 3.2067863941192627,
      "learning_rate": 0.0001744744744744745,
      "loss": 0.7393,
      "step": 170
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 3.543780565261841,
      "learning_rate": 0.000172972972972973,
      "loss": 0.7776,
      "step": 180
    },
    {
      "epoch": 0.5705705705705706,
      "grad_norm": 5.8286333084106445,
      "learning_rate": 0.00017147147147147148,
      "loss": 0.7427,
      "step": 190
    },
    {
      "epoch": 0.6006006006006006,
      "grad_norm": 2.206890106201172,
      "learning_rate": 0.00016996996996997,
      "loss": 0.8399,
      "step": 200
    },
    {
      "epoch": 0.6006006006006006,
      "eval_accuracy": 0.8862911795961743,
      "eval_loss": 0.5542067885398865,
      "eval_runtime": 26.2959,
      "eval_samples_per_second": 35.785,
      "eval_steps_per_second": 4.487,
      "step": 200
    },
    {
      "epoch": 0.6306306306306306,
      "grad_norm": 5.071735382080078,
      "learning_rate": 0.00016846846846846846,
      "loss": 0.4854,
      "step": 210
    },
    {
      "epoch": 0.6606606606606606,
      "grad_norm": 1.8520969152450562,
      "learning_rate": 0.00016696696696696697,
      "loss": 0.5468,
      "step": 220
    },
    {
      "epoch": 0.6906906906906907,
      "grad_norm": 2.1558125019073486,
      "learning_rate": 0.00016546546546546546,
      "loss": 0.5233,
      "step": 230
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 4.998787879943848,
      "learning_rate": 0.00016396396396396395,
      "loss": 0.6996,
      "step": 240
    },
    {
      "epoch": 0.7507507507507507,
      "grad_norm": 2.3446874618530273,
      "learning_rate": 0.00016246246246246247,
      "loss": 0.6156,
      "step": 250
    },
    {
      "epoch": 0.7807807807807807,
      "grad_norm": 9.410783767700195,
      "learning_rate": 0.00016096096096096096,
      "loss": 0.7073,
      "step": 260
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 1.848683476448059,
      "learning_rate": 0.00015945945945945947,
      "loss": 0.5893,
      "step": 270
    },
    {
      "epoch": 0.8408408408408409,
      "grad_norm": 4.705769062042236,
      "learning_rate": 0.00015795795795795796,
      "loss": 0.617,
      "step": 280
    },
    {
      "epoch": 0.8708708708708709,
      "grad_norm": 3.885349988937378,
      "learning_rate": 0.00015645645645645645,
      "loss": 0.5976,
      "step": 290
    },
    {
      "epoch": 0.9009009009009009,
      "grad_norm": 2.5080394744873047,
      "learning_rate": 0.00015495495495495496,
      "loss": 0.6418,
      "step": 300
    },
    {
      "epoch": 0.9009009009009009,
      "eval_accuracy": 0.8671625929861849,
      "eval_loss": 0.5741079449653625,
      "eval_runtime": 26.4803,
      "eval_samples_per_second": 35.536,
      "eval_steps_per_second": 4.456,
      "step": 300
    },
    {
      "epoch": 0.9309309309309309,
      "grad_norm": 5.160592555999756,
      "learning_rate": 0.00015345345345345345,
      "loss": 0.6029,
      "step": 310
    },
    {
      "epoch": 0.960960960960961,
      "grad_norm": 3.9884884357452393,
      "learning_rate": 0.00015195195195195194,
      "loss": 0.6916,
      "step": 320
    },
    {
      "epoch": 0.990990990990991,
      "grad_norm": 4.166062355041504,
      "learning_rate": 0.00015045045045045046,
      "loss": 0.7854,
      "step": 330
    },
    {
      "epoch": 1.021021021021021,
      "grad_norm": 4.180619716644287,
      "learning_rate": 0.00014894894894894895,
      "loss": 0.5521,
      "step": 340
    },
    {
      "epoch": 1.0510510510510511,
      "grad_norm": 4.4930315017700195,
      "learning_rate": 0.00014744744744744746,
      "loss": 0.4041,
      "step": 350
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 4.935527324676514,
      "learning_rate": 0.00014594594594594595,
      "loss": 0.4967,
      "step": 360
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 12.9185791015625,
      "learning_rate": 0.00014444444444444444,
      "loss": 0.4248,
      "step": 370
    },
    {
      "epoch": 1.1411411411411412,
      "grad_norm": 3.98164439201355,
      "learning_rate": 0.00014294294294294295,
      "loss": 0.5054,
      "step": 380
    },
    {
      "epoch": 1.1711711711711712,
      "grad_norm": 4.026088237762451,
      "learning_rate": 0.00014144144144144144,
      "loss": 0.4668,
      "step": 390
    },
    {
      "epoch": 1.2012012012012012,
      "grad_norm": 5.09647798538208,
      "learning_rate": 0.00013993993993993996,
      "loss": 0.3785,
      "step": 400
    },
    {
      "epoch": 1.2012012012012012,
      "eval_accuracy": 0.8841657810839533,
      "eval_loss": 0.4702335000038147,
      "eval_runtime": 26.6715,
      "eval_samples_per_second": 35.281,
      "eval_steps_per_second": 4.424,
      "step": 400
    },
    {
      "epoch": 1.2312312312312312,
      "grad_norm": 0.3809890151023865,
      "learning_rate": 0.00013843843843843845,
      "loss": 0.37,
      "step": 410
    },
    {
      "epoch": 1.2612612612612613,
      "grad_norm": 5.375821113586426,
      "learning_rate": 0.00013693693693693693,
      "loss": 0.4845,
      "step": 420
    },
    {
      "epoch": 1.2912912912912913,
      "grad_norm": 1.7599005699157715,
      "learning_rate": 0.00013543543543543545,
      "loss": 0.6018,
      "step": 430
    },
    {
      "epoch": 1.3213213213213213,
      "grad_norm": 1.0588518381118774,
      "learning_rate": 0.00013393393393393394,
      "loss": 0.4875,
      "step": 440
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 2.4473206996917725,
      "learning_rate": 0.00013243243243243243,
      "loss": 0.5371,
      "step": 450
    },
    {
      "epoch": 1.3813813813813813,
      "grad_norm": 9.87109661102295,
      "learning_rate": 0.00013093093093093094,
      "loss": 0.4605,
      "step": 460
    },
    {
      "epoch": 1.4114114114114114,
      "grad_norm": 2.05519700050354,
      "learning_rate": 0.00012942942942942943,
      "loss": 0.4135,
      "step": 470
    },
    {
      "epoch": 1.4414414414414414,
      "grad_norm": 3.4596476554870605,
      "learning_rate": 0.00012792792792792795,
      "loss": 0.543,
      "step": 480
    },
    {
      "epoch": 1.4714714714714714,
      "grad_norm": 2.090067148208618,
      "learning_rate": 0.00012642642642642644,
      "loss": 0.365,
      "step": 490
    },
    {
      "epoch": 1.5015015015015014,
      "grad_norm": 1.422653317451477,
      "learning_rate": 0.00012492492492492492,
      "loss": 0.4451,
      "step": 500
    },
    {
      "epoch": 1.5015015015015014,
      "eval_accuracy": 0.9117959617428267,
      "eval_loss": 0.3684743642807007,
      "eval_runtime": 26.6937,
      "eval_samples_per_second": 35.252,
      "eval_steps_per_second": 4.421,
      "step": 500
    },
    {
      "epoch": 1.5315315315315314,
      "grad_norm": 3.5416722297668457,
      "learning_rate": 0.00012342342342342344,
      "loss": 0.3413,
      "step": 510
    },
    {
      "epoch": 1.5615615615615615,
      "grad_norm": 0.6044331789016724,
      "learning_rate": 0.00012192192192192193,
      "loss": 0.3424,
      "step": 520
    },
    {
      "epoch": 1.5915915915915915,
      "grad_norm": 8.660246849060059,
      "learning_rate": 0.00012042042042042043,
      "loss": 0.4565,
      "step": 530
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 4.744662761688232,
      "learning_rate": 0.00011891891891891893,
      "loss": 0.4159,
      "step": 540
    },
    {
      "epoch": 1.6516516516516515,
      "grad_norm": 2.420365333557129,
      "learning_rate": 0.00011741741741741743,
      "loss": 0.356,
      "step": 550
    },
    {
      "epoch": 1.6816816816816815,
      "grad_norm": 4.598628997802734,
      "learning_rate": 0.00011591591591591592,
      "loss": 0.4904,
      "step": 560
    },
    {
      "epoch": 1.7117117117117115,
      "grad_norm": 3.6854140758514404,
      "learning_rate": 0.00011441441441441443,
      "loss": 0.353,
      "step": 570
    },
    {
      "epoch": 1.7417417417417418,
      "grad_norm": 4.791896343231201,
      "learning_rate": 0.00011291291291291293,
      "loss": 0.4928,
      "step": 580
    },
    {
      "epoch": 1.7717717717717718,
      "grad_norm": 4.51821231842041,
      "learning_rate": 0.00011141141141141143,
      "loss": 0.365,
      "step": 590
    },
    {
      "epoch": 1.8018018018018018,
      "grad_norm": 4.391872882843018,
      "learning_rate": 0.00010990990990990993,
      "loss": 0.4535,
      "step": 600
    },
    {
      "epoch": 1.8018018018018018,
      "eval_accuracy": 0.9096705632306057,
      "eval_loss": 0.37811535596847534,
      "eval_runtime": 26.4312,
      "eval_samples_per_second": 35.602,
      "eval_steps_per_second": 4.464,
      "step": 600
    },
    {
      "epoch": 1.8318318318318318,
      "grad_norm": 2.613373279571533,
      "learning_rate": 0.00010840840840840842,
      "loss": 0.3374,
      "step": 610
    },
    {
      "epoch": 1.8618618618618619,
      "grad_norm": 2.3293354511260986,
      "learning_rate": 0.00010690690690690692,
      "loss": 0.2425,
      "step": 620
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 4.844559669494629,
      "learning_rate": 0.0001054054054054054,
      "loss": 0.4471,
      "step": 630
    },
    {
      "epoch": 1.921921921921922,
      "grad_norm": 4.249619960784912,
      "learning_rate": 0.0001039039039039039,
      "loss": 0.4115,
      "step": 640
    },
    {
      "epoch": 1.951951951951952,
      "grad_norm": 5.946290969848633,
      "learning_rate": 0.0001024024024024024,
      "loss": 0.4435,
      "step": 650
    },
    {
      "epoch": 1.981981981981982,
      "grad_norm": 1.4038317203521729,
      "learning_rate": 0.00010090090090090089,
      "loss": 0.2987,
      "step": 660
    },
    {
      "epoch": 2.012012012012012,
      "grad_norm": 2.397927761077881,
      "learning_rate": 9.93993993993994e-05,
      "loss": 0.224,
      "step": 670
    },
    {
      "epoch": 2.042042042042042,
      "grad_norm": 3.2759602069854736,
      "learning_rate": 9.789789789789791e-05,
      "loss": 0.3491,
      "step": 680
    },
    {
      "epoch": 2.0720720720720722,
      "grad_norm": 0.820797324180603,
      "learning_rate": 9.639639639639641e-05,
      "loss": 0.3682,
      "step": 690
    },
    {
      "epoch": 2.1021021021021022,
      "grad_norm": 1.0604861974716187,
      "learning_rate": 9.48948948948949e-05,
      "loss": 0.4618,
      "step": 700
    },
    {
      "epoch": 2.1021021021021022,
      "eval_accuracy": 0.9287991498405951,
      "eval_loss": 0.3000124990940094,
      "eval_runtime": 26.5936,
      "eval_samples_per_second": 35.384,
      "eval_steps_per_second": 4.437,
      "step": 700
    },
    {
      "epoch": 2.1321321321321323,
      "grad_norm": 1.994698405265808,
      "learning_rate": 9.33933933933934e-05,
      "loss": 0.2813,
      "step": 710
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 1.6023216247558594,
      "learning_rate": 9.18918918918919e-05,
      "loss": 0.3322,
      "step": 720
    },
    {
      "epoch": 2.1921921921921923,
      "grad_norm": 6.890353679656982,
      "learning_rate": 9.039039039039039e-05,
      "loss": 0.2649,
      "step": 730
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 4.888972282409668,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.2021,
      "step": 740
    },
    {
      "epoch": 2.2522522522522523,
      "grad_norm": 4.375626564025879,
      "learning_rate": 8.738738738738738e-05,
      "loss": 0.3112,
      "step": 750
    },
    {
      "epoch": 2.2822822822822824,
      "grad_norm": 8.04282283782959,
      "learning_rate": 8.588588588588588e-05,
      "loss": 0.2771,
      "step": 760
    },
    {
      "epoch": 2.3123123123123124,
      "grad_norm": 3.198418140411377,
      "learning_rate": 8.438438438438439e-05,
      "loss": 0.291,
      "step": 770
    },
    {
      "epoch": 2.3423423423423424,
      "grad_norm": 4.040826320648193,
      "learning_rate": 8.288288288288289e-05,
      "loss": 0.2622,
      "step": 780
    },
    {
      "epoch": 2.3723723723723724,
      "grad_norm": 4.993142127990723,
      "learning_rate": 8.138138138138138e-05,
      "loss": 0.3394,
      "step": 790
    },
    {
      "epoch": 2.4024024024024024,
      "grad_norm": 3.0009608268737793,
      "learning_rate": 7.987987987987988e-05,
      "loss": 0.2321,
      "step": 800
    },
    {
      "epoch": 2.4024024024024024,
      "eval_accuracy": 0.9181721572794899,
      "eval_loss": 0.31457969546318054,
      "eval_runtime": 26.9039,
      "eval_samples_per_second": 34.976,
      "eval_steps_per_second": 4.386,
      "step": 800
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 1.5650168657302856,
      "learning_rate": 7.837837837837838e-05,
      "loss": 0.1994,
      "step": 810
    },
    {
      "epoch": 2.4624624624624625,
      "grad_norm": 2.945512533187866,
      "learning_rate": 7.687687687687688e-05,
      "loss": 0.2863,
      "step": 820
    },
    {
      "epoch": 2.4924924924924925,
      "grad_norm": 6.051520824432373,
      "learning_rate": 7.537537537537538e-05,
      "loss": 0.3662,
      "step": 830
    },
    {
      "epoch": 2.5225225225225225,
      "grad_norm": 2.5240094661712646,
      "learning_rate": 7.387387387387387e-05,
      "loss": 0.2587,
      "step": 840
    },
    {
      "epoch": 2.5525525525525525,
      "grad_norm": 2.640320062637329,
      "learning_rate": 7.237237237237238e-05,
      "loss": 0.1814,
      "step": 850
    },
    {
      "epoch": 2.5825825825825826,
      "grad_norm": 2.48514723777771,
      "learning_rate": 7.087087087087088e-05,
      "loss": 0.2494,
      "step": 860
    },
    {
      "epoch": 2.6126126126126126,
      "grad_norm": 4.28499698638916,
      "learning_rate": 6.936936936936938e-05,
      "loss": 0.1975,
      "step": 870
    },
    {
      "epoch": 2.6426426426426426,
      "grad_norm": 2.4091761112213135,
      "learning_rate": 6.786786786786787e-05,
      "loss": 0.1894,
      "step": 880
    },
    {
      "epoch": 2.6726726726726726,
      "grad_norm": 1.1737655401229858,
      "learning_rate": 6.636636636636637e-05,
      "loss": 0.3093,
      "step": 890
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 1.0858086347579956,
      "learning_rate": 6.486486486486487e-05,
      "loss": 0.1816,
      "step": 900
    },
    {
      "epoch": 2.7027027027027026,
      "eval_accuracy": 0.9213602550478215,
      "eval_loss": 0.30446675419807434,
      "eval_runtime": 26.7771,
      "eval_samples_per_second": 35.142,
      "eval_steps_per_second": 4.407,
      "step": 900
    },
    {
      "epoch": 2.7327327327327327,
      "grad_norm": 6.119840145111084,
      "learning_rate": 6.336336336336337e-05,
      "loss": 0.3395,
      "step": 910
    },
    {
      "epoch": 2.7627627627627627,
      "grad_norm": 1.7285120487213135,
      "learning_rate": 6.186186186186186e-05,
      "loss": 0.2643,
      "step": 920
    },
    {
      "epoch": 2.7927927927927927,
      "grad_norm": 4.141608715057373,
      "learning_rate": 6.0360360360360365e-05,
      "loss": 0.1987,
      "step": 930
    },
    {
      "epoch": 2.8228228228228227,
      "grad_norm": 3.5640854835510254,
      "learning_rate": 5.8858858858858854e-05,
      "loss": 0.2566,
      "step": 940
    },
    {
      "epoch": 2.8528528528528527,
      "grad_norm": 0.9691357612609863,
      "learning_rate": 5.7357357357357356e-05,
      "loss": 0.2895,
      "step": 950
    },
    {
      "epoch": 2.8828828828828827,
      "grad_norm": 2.255638360977173,
      "learning_rate": 5.585585585585585e-05,
      "loss": 0.2138,
      "step": 960
    },
    {
      "epoch": 2.9129129129129128,
      "grad_norm": 3.9691970348358154,
      "learning_rate": 5.435435435435435e-05,
      "loss": 0.1973,
      "step": 970
    },
    {
      "epoch": 2.942942942942943,
      "grad_norm": 2.5032782554626465,
      "learning_rate": 5.2852852852852855e-05,
      "loss": 0.2642,
      "step": 980
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 4.827852725982666,
      "learning_rate": 5.135135135135135e-05,
      "loss": 0.2679,
      "step": 990
    },
    {
      "epoch": 3.003003003003003,
      "grad_norm": 0.1294238269329071,
      "learning_rate": 4.984984984984985e-05,
      "loss": 0.2332,
      "step": 1000
    },
    {
      "epoch": 3.003003003003003,
      "eval_accuracy": 0.9043570669500531,
      "eval_loss": 0.3445766568183899,
      "eval_runtime": 26.6419,
      "eval_samples_per_second": 35.32,
      "eval_steps_per_second": 4.429,
      "step": 1000
    },
    {
      "epoch": 3.033033033033033,
      "grad_norm": 3.9579365253448486,
      "learning_rate": 4.834834834834835e-05,
      "loss": 0.196,
      "step": 1010
    },
    {
      "epoch": 3.063063063063063,
      "grad_norm": 3.364076852798462,
      "learning_rate": 4.684684684684685e-05,
      "loss": 0.2462,
      "step": 1020
    },
    {
      "epoch": 3.093093093093093,
      "grad_norm": 2.6246888637542725,
      "learning_rate": 4.5345345345345345e-05,
      "loss": 0.1611,
      "step": 1030
    },
    {
      "epoch": 3.123123123123123,
      "grad_norm": 0.28202980756759644,
      "learning_rate": 4.384384384384385e-05,
      "loss": 0.1105,
      "step": 1040
    },
    {
      "epoch": 3.153153153153153,
      "grad_norm": 0.38636475801467896,
      "learning_rate": 4.234234234234234e-05,
      "loss": 0.1293,
      "step": 1050
    },
    {
      "epoch": 3.1831831831831834,
      "grad_norm": 2.700636625289917,
      "learning_rate": 4.0840840840840845e-05,
      "loss": 0.2852,
      "step": 1060
    },
    {
      "epoch": 3.2132132132132134,
      "grad_norm": 5.883915424346924,
      "learning_rate": 3.933933933933934e-05,
      "loss": 0.1586,
      "step": 1070
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 0.6246801614761353,
      "learning_rate": 3.783783783783784e-05,
      "loss": 0.1551,
      "step": 1080
    },
    {
      "epoch": 3.2732732732732734,
      "grad_norm": 1.7150659561157227,
      "learning_rate": 3.633633633633634e-05,
      "loss": 0.1931,
      "step": 1090
    },
    {
      "epoch": 3.3033033033033035,
      "grad_norm": 0.42575761675834656,
      "learning_rate": 3.483483483483483e-05,
      "loss": 0.1173,
      "step": 1100
    },
    {
      "epoch": 3.3033033033033035,
      "eval_accuracy": 0.9415515409139213,
      "eval_loss": 0.23813442885875702,
      "eval_runtime": 26.6996,
      "eval_samples_per_second": 35.244,
      "eval_steps_per_second": 4.42,
      "step": 1100
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 1.8270289897918701,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.3149,
      "step": 1110
    },
    {
      "epoch": 3.3633633633633635,
      "grad_norm": 0.11395557224750519,
      "learning_rate": 3.183183183183183e-05,
      "loss": 0.2494,
      "step": 1120
    },
    {
      "epoch": 3.3933933933933935,
      "grad_norm": 3.1319284439086914,
      "learning_rate": 3.0330330330330332e-05,
      "loss": 0.1403,
      "step": 1130
    },
    {
      "epoch": 3.4234234234234235,
      "grad_norm": 1.6305768489837646,
      "learning_rate": 2.882882882882883e-05,
      "loss": 0.2106,
      "step": 1140
    },
    {
      "epoch": 3.4534534534534536,
      "grad_norm": 2.2145802974700928,
      "learning_rate": 2.732732732732733e-05,
      "loss": 0.1435,
      "step": 1150
    },
    {
      "epoch": 3.4834834834834836,
      "grad_norm": 1.2664133310317993,
      "learning_rate": 2.582582582582583e-05,
      "loss": 0.2026,
      "step": 1160
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 0.11266277730464935,
      "learning_rate": 2.4324324324324327e-05,
      "loss": 0.2054,
      "step": 1170
    },
    {
      "epoch": 3.5435435435435436,
      "grad_norm": 4.44317626953125,
      "learning_rate": 2.2822822822822822e-05,
      "loss": 0.1943,
      "step": 1180
    },
    {
      "epoch": 3.5735735735735736,
      "grad_norm": 2.41487979888916,
      "learning_rate": 2.132132132132132e-05,
      "loss": 0.2326,
      "step": 1190
    },
    {
      "epoch": 3.6036036036036037,
      "grad_norm": 6.012082099914551,
      "learning_rate": 1.981981981981982e-05,
      "loss": 0.2694,
      "step": 1200
    },
    {
      "epoch": 3.6036036036036037,
      "eval_accuracy": 0.9426142401700319,
      "eval_loss": 0.21462436020374298,
      "eval_runtime": 26.5228,
      "eval_samples_per_second": 35.479,
      "eval_steps_per_second": 4.449,
      "step": 1200
    },
    {
      "epoch": 3.6336336336336337,
      "grad_norm": 4.844743251800537,
      "learning_rate": 1.831831831831832e-05,
      "loss": 0.1067,
      "step": 1210
    },
    {
      "epoch": 3.6636636636636637,
      "grad_norm": 0.3227928876876831,
      "learning_rate": 1.6816816816816817e-05,
      "loss": 0.0959,
      "step": 1220
    },
    {
      "epoch": 3.6936936936936937,
      "grad_norm": 1.930338978767395,
      "learning_rate": 1.5315315315315316e-05,
      "loss": 0.2313,
      "step": 1230
    },
    {
      "epoch": 3.7237237237237237,
      "grad_norm": 4.675666332244873,
      "learning_rate": 1.3813813813813815e-05,
      "loss": 0.1652,
      "step": 1240
    },
    {
      "epoch": 3.7537537537537538,
      "grad_norm": 0.07279150187969208,
      "learning_rate": 1.2312312312312313e-05,
      "loss": 0.1007,
      "step": 1250
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 0.06191149726510048,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.1737,
      "step": 1260
    },
    {
      "epoch": 3.813813813813814,
      "grad_norm": 0.23188060522079468,
      "learning_rate": 9.309309309309309e-06,
      "loss": 0.1923,
      "step": 1270
    },
    {
      "epoch": 3.843843843843844,
      "grad_norm": 0.0755883976817131,
      "learning_rate": 7.807807807807808e-06,
      "loss": 0.0793,
      "step": 1280
    },
    {
      "epoch": 3.873873873873874,
      "grad_norm": 3.1278254985809326,
      "learning_rate": 6.306306306306306e-06,
      "loss": 0.1839,
      "step": 1290
    },
    {
      "epoch": 3.903903903903904,
      "grad_norm": 0.0690230205655098,
      "learning_rate": 4.804804804804805e-06,
      "loss": 0.1227,
      "step": 1300
    },
    {
      "epoch": 3.903903903903904,
      "eval_accuracy": 0.9489904357066951,
      "eval_loss": 0.22585047781467438,
      "eval_runtime": 26.5743,
      "eval_samples_per_second": 35.41,
      "eval_steps_per_second": 4.44,
      "step": 1300
    },
    {
      "epoch": 3.933933933933934,
      "grad_norm": 4.087301254272461,
      "learning_rate": 3.3033033033033035e-06,
      "loss": 0.1755,
      "step": 1310
    },
    {
      "epoch": 3.963963963963964,
      "grad_norm": 1.5509567260742188,
      "learning_rate": 1.801801801801802e-06,
      "loss": 0.2882,
      "step": 1320
    },
    {
      "epoch": 3.993993993993994,
      "grad_norm": 0.1417875587940216,
      "learning_rate": 3.003003003003003e-07,
      "loss": 0.1315,
      "step": 1330
    },
    {
      "epoch": 4.0,
      "step": 1332,
      "total_flos": 1.651775717862015e+18,
      "train_loss": 0.4770685946261203,
      "train_runtime": 1116.421,
      "train_samples_per_second": 19.09,
      "train_steps_per_second": 1.193
    }
  ],
  "logging_steps": 10,
  "max_steps": 1332,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.651775717862015e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}