{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7099078774205676,
  "eval_steps": 200.0,
  "global_step": 1800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001504042113179169,
      "grad_norm": 12.573603630065918,
      "learning_rate": 2.0000000000000002e-07,
      "loss": 0.8482376337051392,
      "memory(GiB)": 79.65,
      "step": 1,
      "token_acc": 0.7770833333333333,
      "train_speed(iter/s)": 0.0217
    },
    {
      "epoch": 0.007520210565895845,
      "grad_norm": 13.392790794372559,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.8456712961196899,
      "memory(GiB)": 79.65,
      "step": 5,
      "token_acc": 0.770600414078675,
      "train_speed(iter/s)": 0.05049
    },
    {
      "epoch": 0.01504042113179169,
      "grad_norm": 10.737227439880371,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.7528789520263672,
      "memory(GiB)": 79.65,
      "step": 10,
      "token_acc": 0.7820947342316277,
      "train_speed(iter/s)": 0.06053
    },
    {
      "epoch": 0.022560631697687534,
      "grad_norm": 4.092273712158203,
      "learning_rate": 3e-06,
      "loss": 0.5242659091949463,
      "memory(GiB)": 79.65,
      "step": 15,
      "token_acc": 0.8211612044584927,
      "train_speed(iter/s)": 0.064858
    },
    {
      "epoch": 0.03008084226358338,
      "grad_norm": 3.8328819274902344,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.4107938766479492,
      "memory(GiB)": 79.65,
      "step": 20,
      "token_acc": 0.8372093023255814,
      "train_speed(iter/s)": 0.06725
    },
    {
      "epoch": 0.037601052829479224,
      "grad_norm": 1.1977204084396362,
      "learning_rate": 5e-06,
      "loss": 0.3767481565475464,
      "memory(GiB)": 79.65,
      "step": 25,
      "token_acc": 0.856261464065366,
      "train_speed(iter/s)": 0.068784
    },
    {
      "epoch": 0.04512126339537507,
      "grad_norm": 1.3346306085586548,
      "learning_rate": 6e-06,
      "loss": 0.3551741600036621,
      "memory(GiB)": 79.65,
      "step": 30,
      "token_acc": 0.8516994486050531,
      "train_speed(iter/s)": 0.069846
    },
    {
      "epoch": 0.05264147396127092,
      "grad_norm": 1.0302420854568481,
      "learning_rate": 7e-06,
      "loss": 0.341081428527832,
      "memory(GiB)": 79.65,
      "step": 35,
      "token_acc": 0.8633242326466452,
      "train_speed(iter/s)": 0.070632
    },
    {
      "epoch": 0.06016168452716676,
      "grad_norm": 1.1837648153305054,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.3382600784301758,
      "memory(GiB)": 79.65,
      "step": 40,
      "token_acc": 0.8694029850746269,
      "train_speed(iter/s)": 0.071224
    },
    {
      "epoch": 0.0676818950930626,
      "grad_norm": 0.8697211146354675,
      "learning_rate": 9e-06,
      "loss": 0.33454647064208987,
      "memory(GiB)": 79.65,
      "step": 45,
      "token_acc": 0.8626541932279163,
      "train_speed(iter/s)": 0.0717
    },
    {
      "epoch": 0.07520210565895845,
      "grad_norm": 1.3390811681747437,
      "learning_rate": 1e-05,
      "loss": 0.3305459499359131,
      "memory(GiB)": 79.65,
      "step": 50,
      "token_acc": 0.8764549384768873,
      "train_speed(iter/s)": 0.072077
    },
    {
      "epoch": 0.08272231622485429,
      "grad_norm": 0.7468055486679077,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.33402838706970217,
      "memory(GiB)": 79.65,
      "step": 55,
      "token_acc": 0.8601444517399869,
      "train_speed(iter/s)": 0.072349
    },
    {
      "epoch": 0.09024252679075014,
      "grad_norm": 1.2021645307540894,
      "learning_rate": 1.2e-05,
      "loss": 0.32740762233734133,
      "memory(GiB)": 79.65,
      "step": 60,
      "token_acc": 0.8739544513457557,
      "train_speed(iter/s)": 0.0726
    },
    {
      "epoch": 0.09776273735664599,
      "grad_norm": 1.15927255153656,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.3289224624633789,
      "memory(GiB)": 79.65,
      "step": 65,
      "token_acc": 0.874283340257582,
      "train_speed(iter/s)": 0.072826
    },
    {
      "epoch": 0.10528294792254184,
      "grad_norm": 1.1537578105926514,
      "learning_rate": 1.4e-05,
      "loss": 0.31473894119262696,
      "memory(GiB)": 79.65,
      "step": 70,
      "token_acc": 0.8649830816208632,
      "train_speed(iter/s)": 0.073026
    },
    {
      "epoch": 0.11280315848843768,
      "grad_norm": 0.8681713342666626,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.3286761283874512,
      "memory(GiB)": 79.65,
      "step": 75,
      "token_acc": 0.8661670235546038,
      "train_speed(iter/s)": 0.073199
    },
    {
      "epoch": 0.12032336905433352,
      "grad_norm": 0.5740350484848022,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.326131272315979,
      "memory(GiB)": 79.65,
      "step": 80,
      "token_acc": 0.8724307231422151,
      "train_speed(iter/s)": 0.073353
    },
    {
      "epoch": 0.12784357962022935,
      "grad_norm": 1.2273070812225342,
      "learning_rate": 1.7e-05,
      "loss": 0.32001395225524903,
      "memory(GiB)": 79.65,
      "step": 85,
      "token_acc": 0.8689963564094071,
      "train_speed(iter/s)": 0.073484
    },
    {
      "epoch": 0.1353637901861252,
      "grad_norm": 1.3030885457992554,
      "learning_rate": 1.8e-05,
      "loss": 0.3294826984405518,
      "memory(GiB)": 79.65,
      "step": 90,
      "token_acc": 0.8662824207492795,
      "train_speed(iter/s)": 0.073606
    },
    {
      "epoch": 0.14288400075202107,
      "grad_norm": 0.7490714192390442,
      "learning_rate": 1.9e-05,
      "loss": 0.3211155652999878,
      "memory(GiB)": 79.65,
      "step": 95,
      "token_acc": 0.869461373744029,
      "train_speed(iter/s)": 0.073709
    },
    {
      "epoch": 0.1504042113179169,
      "grad_norm": 0.7038081288337708,
      "learning_rate": 2e-05,
      "loss": 0.3172138214111328,
      "memory(GiB)": 79.65,
      "step": 100,
      "token_acc": 0.876581225033289,
      "train_speed(iter/s)": 0.073806
    },
    {
      "epoch": 0.15792442188381275,
      "grad_norm": 0.7102833986282349,
      "learning_rate": 2.1000000000000002e-05,
      "loss": 0.3214402198791504,
      "memory(GiB)": 79.65,
      "step": 105,
      "token_acc": 0.8711681855840928,
      "train_speed(iter/s)": 0.073894
    },
    {
      "epoch": 0.16544463244970858,
      "grad_norm": 0.715777575969696,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.3172021389007568,
      "memory(GiB)": 79.65,
      "step": 110,
      "token_acc": 0.8741798853915788,
      "train_speed(iter/s)": 0.073973
    },
    {
      "epoch": 0.17296484301560444,
      "grad_norm": 1.0129011869430542,
      "learning_rate": 2.3e-05,
      "loss": 0.31590800285339354,
      "memory(GiB)": 79.65,
      "step": 115,
      "token_acc": 0.8736597124096085,
      "train_speed(iter/s)": 0.074045
    },
    {
      "epoch": 0.18048505358150027,
      "grad_norm": 1.2848035097122192,
      "learning_rate": 2.4e-05,
      "loss": 0.3219757080078125,
      "memory(GiB)": 79.65,
      "step": 120,
      "token_acc": 0.8769676884838442,
      "train_speed(iter/s)": 0.074115
    },
    {
      "epoch": 0.18800526414739613,
      "grad_norm": 0.8289978504180908,
      "learning_rate": 2.5e-05,
      "loss": 0.31909029483795165,
      "memory(GiB)": 79.65,
      "step": 125,
      "token_acc": 0.8719134528036997,
      "train_speed(iter/s)": 0.074176
    },
    {
      "epoch": 0.19552547471329199,
      "grad_norm": 0.8131099939346313,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.3183478832244873,
      "memory(GiB)": 79.65,
      "step": 130,
      "token_acc": 0.8736510044828159,
      "train_speed(iter/s)": 0.074235
    },
    {
      "epoch": 0.20304568527918782,
      "grad_norm": 0.8034164905548096,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.315186882019043,
      "memory(GiB)": 79.65,
      "step": 135,
      "token_acc": 0.8744402056725825,
      "train_speed(iter/s)": 0.074286
    },
    {
      "epoch": 0.21056589584508367,
      "grad_norm": 0.9958615899085999,
      "learning_rate": 2.8e-05,
      "loss": 0.3196675777435303,
      "memory(GiB)": 79.65,
      "step": 140,
      "token_acc": 0.8743993371996686,
      "train_speed(iter/s)": 0.074335
    },
    {
      "epoch": 0.2180861064109795,
      "grad_norm": 0.7339183688163757,
      "learning_rate": 2.9e-05,
      "loss": 0.3180043935775757,
      "memory(GiB)": 79.65,
      "step": 145,
      "token_acc": 0.8774578943001742,
      "train_speed(iter/s)": 0.07438
    },
    {
      "epoch": 0.22560631697687536,
      "grad_norm": 0.7186743021011353,
      "learning_rate": 3.0000000000000004e-05,
      "loss": 0.31993443965911866,
      "memory(GiB)": 79.65,
      "step": 150,
      "token_acc": 0.8700420549187763,
      "train_speed(iter/s)": 0.074422
    },
    {
      "epoch": 0.2331265275427712,
      "grad_norm": 0.6232396960258484,
      "learning_rate": 3.1e-05,
      "loss": 0.3148055076599121,
      "memory(GiB)": 79.65,
      "step": 155,
      "token_acc": 0.8787347688198965,
      "train_speed(iter/s)": 0.074461
    },
    {
      "epoch": 0.24064673810866705,
      "grad_norm": 0.7416806221008301,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.31689014434814455,
      "memory(GiB)": 79.65,
      "step": 160,
      "token_acc": 0.871843853820598,
      "train_speed(iter/s)": 0.074498
    },
    {
      "epoch": 0.24816694867456288,
      "grad_norm": 2.383039951324463,
      "learning_rate": 3.3e-05,
      "loss": 0.34125614166259766,
      "memory(GiB)": 79.65,
      "step": 165,
      "token_acc": 0.8648180242634316,
      "train_speed(iter/s)": 0.074533
    },
    {
      "epoch": 0.2556871592404587,
      "grad_norm": 0.8704357743263245,
      "learning_rate": 3.4e-05,
      "loss": 0.32818725109100344,
      "memory(GiB)": 79.65,
      "step": 170,
      "token_acc": 0.8705250103348491,
      "train_speed(iter/s)": 0.074565
    },
    {
      "epoch": 0.26320736980635456,
      "grad_norm": 0.8643216490745544,
      "learning_rate": 3.5000000000000004e-05,
      "loss": 0.3227179527282715,
      "memory(GiB)": 79.65,
      "step": 175,
      "token_acc": 0.8669157326691573,
      "train_speed(iter/s)": 0.074582
    },
    {
      "epoch": 0.2707275803722504,
      "grad_norm": 3.8876283168792725,
      "learning_rate": 3.6e-05,
      "loss": 0.3369649887084961,
      "memory(GiB)": 79.65,
      "step": 180,
      "token_acc": 0.866315008654084,
      "train_speed(iter/s)": 0.074607
    },
    {
      "epoch": 0.2782477909381463,
      "grad_norm": 0.5515977144241333,
      "learning_rate": 3.7000000000000005e-05,
      "loss": 0.34408109188079833,
      "memory(GiB)": 79.65,
      "step": 185,
      "token_acc": 0.8688796680497926,
      "train_speed(iter/s)": 0.074635
    },
    {
      "epoch": 0.28576800150404214,
      "grad_norm": 0.44839155673980713,
      "learning_rate": 3.8e-05,
      "loss": 0.3227226257324219,
      "memory(GiB)": 79.65,
      "step": 190,
      "token_acc": 0.8672931091328717,
      "train_speed(iter/s)": 0.074663
    },
    {
      "epoch": 0.29328821206993794,
      "grad_norm": 3.1341962814331055,
      "learning_rate": 3.9e-05,
      "loss": 0.31741573810577395,
      "memory(GiB)": 79.65,
      "step": 195,
      "token_acc": 0.874813834188317,
      "train_speed(iter/s)": 0.074691
    },
    {
      "epoch": 0.3008084226358338,
      "grad_norm": 0.9332788586616516,
      "learning_rate": 4e-05,
      "loss": 0.3259200811386108,
      "memory(GiB)": 79.65,
      "step": 200,
      "token_acc": 0.8639647869778257,
      "train_speed(iter/s)": 0.074717
    },
    {
      "epoch": 0.30832863320172965,
      "grad_norm": 0.5933415293693542,
      "learning_rate": 3.999923164669689e-05,
      "loss": 0.31851494312286377,
      "memory(GiB)": 79.65,
      "step": 205,
      "token_acc": 0.8654701136176519,
      "train_speed(iter/s)": 0.073044
    },
    {
      "epoch": 0.3158488437676255,
      "grad_norm": 0.5378974080085754,
      "learning_rate": 3.999692664582422e-05,
      "loss": 0.315838098526001,
      "memory(GiB)": 79.65,
      "step": 210,
      "token_acc": 0.8722272991609205,
      "train_speed(iter/s)": 0.073106
    },
    {
      "epoch": 0.32336905433352137,
      "grad_norm": 0.7262965440750122,
      "learning_rate": 3.9993085174487494e-05,
      "loss": 0.30927584171295164,
      "memory(GiB)": 79.65,
      "step": 215,
      "token_acc": 0.8718796495288478,
      "train_speed(iter/s)": 0.073164
    },
    {
      "epoch": 0.33088926489941717,
      "grad_norm": 0.5802333354949951,
      "learning_rate": 3.998770752784745e-05,
      "loss": 0.31653785705566406,
      "memory(GiB)": 79.65,
      "step": 220,
      "token_acc": 0.8635952283011107,
      "train_speed(iter/s)": 0.07322
    },
    {
      "epoch": 0.338409475465313,
      "grad_norm": 0.402685284614563,
      "learning_rate": 3.998079411909731e-05,
      "loss": 0.31244258880615233,
      "memory(GiB)": 79.65,
      "step": 225,
      "token_acc": 0.8728385869115579,
      "train_speed(iter/s)": 0.073273
    },
    {
      "epoch": 0.3459296860312089,
      "grad_norm": 0.4992331266403198,
      "learning_rate": 3.997234547943116e-05,
      "loss": 0.3138291358947754,
      "memory(GiB)": 79.65,
      "step": 230,
      "token_acc": 0.8737102765167148,
      "train_speed(iter/s)": 0.073325
    },
    {
      "epoch": 0.35344989659710474,
      "grad_norm": 0.3645344376564026,
      "learning_rate": 3.996236225800298e-05,
      "loss": 0.31688103675842283,
      "memory(GiB)": 79.65,
      "step": 235,
      "token_acc": 0.8744286545333666,
      "train_speed(iter/s)": 0.073374
    },
    {
      "epoch": 0.36097010716300054,
      "grad_norm": 0.40172266960144043,
      "learning_rate": 3.9950845221876926e-05,
      "loss": 0.31302390098571775,
      "memory(GiB)": 79.65,
      "step": 240,
      "token_acc": 0.8703243958953989,
      "train_speed(iter/s)": 0.073422
    },
    {
      "epoch": 0.3684903177288964,
      "grad_norm": 0.6376217007637024,
      "learning_rate": 3.993779525596824e-05,
      "loss": 0.3126035690307617,
      "memory(GiB)": 79.65,
      "step": 245,
      "token_acc": 0.8742861872051643,
      "train_speed(iter/s)": 0.073467
    },
    {
      "epoch": 0.37601052829479226,
      "grad_norm": 0.34382951259613037,
      "learning_rate": 3.992321336297537e-05,
      "loss": 0.31143746376037595,
      "memory(GiB)": 79.65,
      "step": 250,
      "token_acc": 0.8750206100577081,
      "train_speed(iter/s)": 0.073508
    },
    {
      "epoch": 0.3835307388606881,
      "grad_norm": 0.33769020438194275,
      "learning_rate": 3.99071006633029e-05,
      "loss": 0.3125121831893921,
      "memory(GiB)": 79.65,
      "step": 255,
      "token_acc": 0.8788231382978723,
      "train_speed(iter/s)": 0.07355
    },
    {
      "epoch": 0.39105094942658397,
      "grad_norm": 0.5663004517555237,
      "learning_rate": 3.9889458394975404e-05,
      "loss": 0.309099817276001,
      "memory(GiB)": 79.65,
      "step": 260,
      "token_acc": 0.8711839166046166,
      "train_speed(iter/s)": 0.073591
    },
    {
      "epoch": 0.3985711599924798,
      "grad_norm": 0.49101531505584717,
      "learning_rate": 3.9870287913542416e-05,
      "loss": 0.3081879854202271,
      "memory(GiB)": 79.65,
      "step": 265,
      "token_acc": 0.8748866727107888,
      "train_speed(iter/s)": 0.07363
    },
    {
      "epoch": 0.40609137055837563,
      "grad_norm": 0.43004897236824036,
      "learning_rate": 3.9849590691974206e-05,
      "loss": 0.30682179927825926,
      "memory(GiB)": 79.65,
      "step": 270,
      "token_acc": 0.8798197295943916,
      "train_speed(iter/s)": 0.073668
    },
    {
      "epoch": 0.4136115811242715,
      "grad_norm": 0.4323391616344452,
      "learning_rate": 3.982736832054862e-05,
      "loss": 0.30874035358428953,
      "memory(GiB)": 79.65,
      "step": 275,
      "token_acc": 0.8773107850451795,
      "train_speed(iter/s)": 0.073704
    },
    {
      "epoch": 0.42113179169016735,
      "grad_norm": 0.3533615469932556,
      "learning_rate": 3.9803622506728916e-05,
      "loss": 0.3092989444732666,
      "memory(GiB)": 79.65,
      "step": 280,
      "token_acc": 0.8765718067504964,
      "train_speed(iter/s)": 0.073739
    },
    {
      "epoch": 0.42865200225606315,
      "grad_norm": 0.59963059425354,
      "learning_rate": 3.977835507503254e-05,
      "loss": 0.3071634292602539,
      "memory(GiB)": 79.65,
      "step": 285,
      "token_acc": 0.8794739470617613,
      "train_speed(iter/s)": 0.073771
    },
    {
      "epoch": 0.436172212821959,
      "grad_norm": 0.34241795539855957,
      "learning_rate": 3.9751567966890946e-05,
      "loss": 0.31339426040649415,
      "memory(GiB)": 79.65,
      "step": 290,
      "token_acc": 0.8726638425118365,
      "train_speed(iter/s)": 0.073801
    },
    {
      "epoch": 0.44369242338785486,
      "grad_norm": 0.4361538589000702,
      "learning_rate": 3.972326324050045e-05,
      "loss": 0.31136701107025144,
      "memory(GiB)": 79.65,
      "step": 295,
      "token_acc": 0.8795550759525192,
      "train_speed(iter/s)": 0.073826
    },
    {
      "epoch": 0.4512126339537507,
      "grad_norm": 0.3572317659854889,
      "learning_rate": 3.969344307066404e-05,
      "loss": 0.3079061985015869,
      "memory(GiB)": 79.65,
      "step": 300,
      "token_acc": 0.8773857721226143,
      "train_speed(iter/s)": 0.073857
    },
    {
      "epoch": 0.4587328445196466,
      "grad_norm": 0.47381100058555603,
      "learning_rate": 3.966210974862433e-05,
      "loss": 0.3093868255615234,
      "memory(GiB)": 79.65,
      "step": 305,
      "token_acc": 0.8700378850271784,
      "train_speed(iter/s)": 0.073887
    },
    {
      "epoch": 0.4662530550855424,
      "grad_norm": 0.3873850107192993,
      "learning_rate": 3.9629265681887456e-05,
      "loss": 0.3090707302093506,
      "memory(GiB)": 79.65,
      "step": 310,
      "token_acc": 0.8732185517752697,
      "train_speed(iter/s)": 0.073916
    },
    {
      "epoch": 0.47377326565143824,
      "grad_norm": 0.46002689003944397,
      "learning_rate": 3.959491339403814e-05,
      "loss": 0.3073902130126953,
      "memory(GiB)": 79.65,
      "step": 315,
      "token_acc": 0.8739212624311663,
      "train_speed(iter/s)": 0.073944
    },
    {
      "epoch": 0.4812934762173341,
      "grad_norm": 0.38025927543640137,
      "learning_rate": 3.9559055524545755e-05,
      "loss": 0.31463534832000734,
      "memory(GiB)": 79.65,
      "step": 320,
      "token_acc": 0.8749272833042466,
      "train_speed(iter/s)": 0.073972
    },
    {
      "epoch": 0.48881368678322995,
      "grad_norm": 0.5703704953193665,
      "learning_rate": 3.9521694828561566e-05,
      "loss": 0.3123070240020752,
      "memory(GiB)": 79.65,
      "step": 325,
      "token_acc": 0.8728939544103073,
      "train_speed(iter/s)": 0.074
    },
    {
      "epoch": 0.49633389734912575,
      "grad_norm": 0.38686856627464294,
      "learning_rate": 3.948283417670698e-05,
      "loss": 0.30756487846374514,
      "memory(GiB)": 79.65,
      "step": 330,
      "token_acc": 0.8759166666666667,
      "train_speed(iter/s)": 0.074025
    },
    {
      "epoch": 0.5038541079150216,
      "grad_norm": 0.4500496983528137,
      "learning_rate": 3.944247655485302e-05,
      "loss": 0.30926506519317626,
      "memory(GiB)": 79.65,
      "step": 335,
      "token_acc": 0.8763895802223328,
      "train_speed(iter/s)": 0.074049
    },
    {
      "epoch": 0.5113743184809174,
      "grad_norm": 0.3525928556919098,
      "learning_rate": 3.940062506389089e-05,
      "loss": 0.31139240264892576,
      "memory(GiB)": 79.65,
      "step": 340,
      "token_acc": 0.8722513948145717,
      "train_speed(iter/s)": 0.074074
    },
    {
      "epoch": 0.5188945290468133,
      "grad_norm": 0.41270679235458374,
      "learning_rate": 3.9357282919493717e-05,
      "loss": 0.30599441528320315,
      "memory(GiB)": 79.65,
      "step": 345,
      "token_acc": 0.8755566551212272,
      "train_speed(iter/s)": 0.074098
    },
    {
      "epoch": 0.5264147396127091,
      "grad_norm": 0.342576265335083,
      "learning_rate": 3.931245345186949e-05,
      "loss": 0.3061314582824707,
      "memory(GiB)": 79.65,
      "step": 350,
      "token_acc": 0.8713652045342534,
      "train_speed(iter/s)": 0.074121
    },
    {
      "epoch": 0.533934950178605,
      "grad_norm": 0.6053279638290405,
      "learning_rate": 3.926614010550516e-05,
      "loss": 0.3078542470932007,
      "memory(GiB)": 79.65,
      "step": 355,
      "token_acc": 0.8771696702931651,
      "train_speed(iter/s)": 0.074143
    },
    {
      "epoch": 0.5414551607445008,
      "grad_norm": 0.5558514595031738,
      "learning_rate": 3.9218346438901996e-05,
      "loss": 0.3026005268096924,
      "memory(GiB)": 79.65,
      "step": 360,
      "token_acc": 0.8785559807020462,
      "train_speed(iter/s)": 0.074165
    },
    {
      "epoch": 0.5489753713103966,
      "grad_norm": 0.44175437092781067,
      "learning_rate": 3.916907612430215e-05,
      "loss": 0.30403385162353513,
      "memory(GiB)": 79.65,
      "step": 365,
      "token_acc": 0.8722986247544204,
      "train_speed(iter/s)": 0.074185
    },
    {
      "epoch": 0.5564955818762926,
      "grad_norm": 0.317410409450531,
      "learning_rate": 3.911833294740653e-05,
      "loss": 0.30560731887817383,
      "memory(GiB)": 79.65,
      "step": 370,
      "token_acc": 0.8801862476095452,
      "train_speed(iter/s)": 0.074205
    },
    {
      "epoch": 0.5640157924421884,
      "grad_norm": 0.3975081741809845,
      "learning_rate": 3.9066120807083875e-05,
      "loss": 0.3044612407684326,
      "memory(GiB)": 79.65,
      "step": 375,
      "token_acc": 0.8787953389135073,
      "train_speed(iter/s)": 0.074225
    },
    {
      "epoch": 0.5715360030080843,
      "grad_norm": 0.3742714524269104,
      "learning_rate": 3.901244371507126e-05,
      "loss": 0.30440506935119627,
      "memory(GiB)": 79.65,
      "step": 380,
      "token_acc": 0.8755063238819542,
      "train_speed(iter/s)": 0.074245
    },
    {
      "epoch": 0.5790562135739801,
      "grad_norm": 0.3854736089706421,
      "learning_rate": 3.895730579566576e-05,
      "loss": 0.3050675868988037,
      "memory(GiB)": 79.65,
      "step": 385,
      "token_acc": 0.8782351491249691,
      "train_speed(iter/s)": 0.074264
    },
    {
      "epoch": 0.5865764241398759,
      "grad_norm": 0.3571319580078125,
      "learning_rate": 3.8900711285407625e-05,
      "loss": 0.3036262273788452,
      "memory(GiB)": 79.65,
      "step": 390,
      "token_acc": 0.8770505385252693,
      "train_speed(iter/s)": 0.074283
    },
    {
      "epoch": 0.5940966347057718,
      "grad_norm": 0.4355558156967163,
      "learning_rate": 3.884266453275476e-05,
      "loss": 0.30002660751342775,
      "memory(GiB)": 79.65,
      "step": 395,
      "token_acc": 0.8763663464723418,
      "train_speed(iter/s)": 0.074301
    },
    {
      "epoch": 0.6016168452716676,
      "grad_norm": 0.31444212794303894,
      "learning_rate": 3.878316999774856e-05,
      "loss": 0.29926114082336425,
      "memory(GiB)": 79.65,
      "step": 400,
      "token_acc": 0.8858474082702388,
      "train_speed(iter/s)": 0.074319
    },
    {
      "epoch": 0.6091370558375635,
      "grad_norm": 0.5459496974945068,
      "learning_rate": 3.872223225167129e-05,
      "loss": 0.2966423034667969,
      "memory(GiB)": 79.65,
      "step": 405,
      "token_acc": 0.8739509763190694,
      "train_speed(iter/s)": 0.073503
    },
    {
      "epoch": 0.6166572664034593,
      "grad_norm": 0.4573298990726471,
      "learning_rate": 3.865985597669478e-05,
      "loss": 0.30276243686676024,
      "memory(GiB)": 79.65,
      "step": 410,
      "token_acc": 0.8737153662747678,
      "train_speed(iter/s)": 0.073525
    },
    {
      "epoch": 0.6241774769693551,
      "grad_norm": 0.41213834285736084,
      "learning_rate": 3.859604596552074e-05,
      "loss": 0.30186285972595217,
      "memory(GiB)": 79.65,
      "step": 415,
      "token_acc": 0.8772422914772258,
      "train_speed(iter/s)": 0.073549
    },
    {
      "epoch": 0.631697687535251,
      "grad_norm": 0.326226145029068,
      "learning_rate": 3.8530807121012446e-05,
      "loss": 0.29910123348236084,
      "memory(GiB)": 79.65,
      "step": 420,
      "token_acc": 0.8792648755562881,
      "train_speed(iter/s)": 0.073573
    },
    {
      "epoch": 0.6392178981011468,
      "grad_norm": 0.3732740879058838,
      "learning_rate": 3.8464144455818065e-05,
      "loss": 0.2988157272338867,
      "memory(GiB)": 79.65,
      "step": 425,
      "token_acc": 0.8769676884838442,
      "train_speed(iter/s)": 0.073598
    },
    {
      "epoch": 0.6467381086670427,
      "grad_norm": 0.3957410454750061,
      "learning_rate": 3.83960630919855e-05,
      "loss": 0.2993508815765381,
      "memory(GiB)": 79.65,
      "step": 430,
      "token_acc": 0.8844455552778472,
      "train_speed(iter/s)": 0.073622
    },
    {
      "epoch": 0.6542583192329385,
      "grad_norm": 0.30964985489845276,
      "learning_rate": 3.8326568260568835e-05,
      "loss": 0.30010457038879396,
      "memory(GiB)": 79.65,
      "step": 435,
      "token_acc": 0.8757528256744493,
      "train_speed(iter/s)": 0.073646
    },
    {
      "epoch": 0.6617785297988343,
      "grad_norm": 0.3214603066444397,
      "learning_rate": 3.825566530122638e-05,
      "loss": 0.30272817611694336,
      "memory(GiB)": 79.65,
      "step": 440,
      "token_acc": 0.8800861854644899,
      "train_speed(iter/s)": 0.073669
    },
    {
      "epoch": 0.6692987403647302,
      "grad_norm": 0.3300005793571472,
      "learning_rate": 3.818335966181045e-05,
      "loss": 0.3003435850143433,
      "memory(GiB)": 79.65,
      "step": 445,
      "token_acc": 0.8815854665565648,
      "train_speed(iter/s)": 0.073691
    },
    {
      "epoch": 0.676818950930626,
      "grad_norm": 0.3879680633544922,
      "learning_rate": 3.8109656897948726e-05,
      "loss": 0.2971020221710205,
      "memory(GiB)": 79.65,
      "step": 450,
      "token_acc": 0.8817848817848818,
      "train_speed(iter/s)": 0.073712
    },
    {
      "epoch": 0.6843391614965219,
      "grad_norm": 0.4819749593734741,
      "learning_rate": 3.803456267261742e-05,
      "loss": 0.30127911567687987,
      "memory(GiB)": 79.65,
      "step": 455,
      "token_acc": 0.8825873951673171,
      "train_speed(iter/s)": 0.073733
    },
    {
      "epoch": 0.6918593720624178,
      "grad_norm": 0.3645648956298828,
      "learning_rate": 3.7958082755706135e-05,
      "loss": 0.2984990119934082,
      "memory(GiB)": 79.65,
      "step": 460,
      "token_acc": 0.8842769384027893,
      "train_speed(iter/s)": 0.073754
    },
    {
      "epoch": 0.6993795826283136,
      "grad_norm": 0.3707161545753479,
      "learning_rate": 3.788022302357455e-05,
      "loss": 0.2968101978302002,
      "memory(GiB)": 79.65,
      "step": 465,
      "token_acc": 0.8793986452998513,
      "train_speed(iter/s)": 0.073775
    },
    {
      "epoch": 0.7068997931942095,
      "grad_norm": 0.2869170904159546,
      "learning_rate": 3.7800989458600906e-05,
      "loss": 0.2998335361480713,
      "memory(GiB)": 79.65,
      "step": 470,
      "token_acc": 0.8837846818408374,
      "train_speed(iter/s)": 0.073795
    },
    {
      "epoch": 0.7144200037601053,
      "grad_norm": 0.37760087847709656,
      "learning_rate": 3.7720388148722336e-05,
      "loss": 0.29495997428894044,
      "memory(GiB)": 79.65,
      "step": 475,
      "token_acc": 0.8836920529801324,
      "train_speed(iter/s)": 0.073816
    },
    {
      "epoch": 0.7219402143260011,
      "grad_norm": 0.3697941303253174,
      "learning_rate": 3.76384252869671e-05,
      "loss": 0.30079140663146975,
      "memory(GiB)": 79.65,
      "step": 480,
      "token_acc": 0.8848289004888558,
      "train_speed(iter/s)": 0.073835
    },
    {
      "epoch": 0.729460424891897,
      "grad_norm": 0.48478394746780396,
      "learning_rate": 3.755510717097877e-05,
      "loss": 0.29771325588226316,
      "memory(GiB)": 79.65,
      "step": 485,
      "token_acc": 0.8831837819873712,
      "train_speed(iter/s)": 0.073855
    },
    {
      "epoch": 0.7369806354577928,
      "grad_norm": 0.42335063219070435,
      "learning_rate": 3.74704402025323e-05,
      "loss": 0.29221677780151367,
      "memory(GiB)": 79.65,
      "step": 490,
      "token_acc": 0.8746511246100804,
      "train_speed(iter/s)": 0.073874
    },
    {
      "epoch": 0.7445008460236887,
      "grad_norm": 0.31420886516571045,
      "learning_rate": 3.738443088704217e-05,
      "loss": 0.29865102767944335,
      "memory(GiB)": 79.65,
      "step": 495,
      "token_acc": 0.883808095952024,
      "train_speed(iter/s)": 0.073892
    },
    {
      "epoch": 0.7520210565895845,
      "grad_norm": 0.39928799867630005,
      "learning_rate": 3.7297085833062567e-05,
      "loss": 0.29097743034362794,
      "memory(GiB)": 79.65,
      "step": 500,
      "token_acc": 0.8790389395194698,
      "train_speed(iter/s)": 0.07391
    },
    {
      "epoch": 0.7595412671554803,
      "grad_norm": 0.44979527592658997,
      "learning_rate": 3.720841175177954e-05,
      "loss": 0.29760844707489015,
      "memory(GiB)": 79.65,
      "step": 505,
      "token_acc": 0.8762030106111706,
      "train_speed(iter/s)": 0.073928
    },
    {
      "epoch": 0.7670614777213762,
      "grad_norm": 0.43242138624191284,
      "learning_rate": 3.711841545649543e-05,
      "loss": 0.29836549758911135,
      "memory(GiB)": 79.65,
      "step": 510,
      "token_acc": 0.8798494025208708,
      "train_speed(iter/s)": 0.073945
    },
    {
      "epoch": 0.774581688287272,
      "grad_norm": 0.507692813873291,
      "learning_rate": 3.702710386210531e-05,
      "loss": 0.2955024242401123,
      "memory(GiB)": 79.65,
      "step": 515,
      "token_acc": 0.8794502401059778,
      "train_speed(iter/s)": 0.073962
    },
    {
      "epoch": 0.7821018988531679,
      "grad_norm": 0.3479640483856201,
      "learning_rate": 3.6934483984565684e-05,
      "loss": 0.2968432426452637,
      "memory(GiB)": 79.65,
      "step": 520,
      "token_acc": 0.8824407229315204,
      "train_speed(iter/s)": 0.073974
    },
    {
      "epoch": 0.7896221094190637,
      "grad_norm": 0.39740270376205444,
      "learning_rate": 3.6840562940355456e-05,
      "loss": 0.29419846534729005,
      "memory(GiB)": 79.65,
      "step": 525,
      "token_acc": 0.8842913155918912,
      "train_speed(iter/s)": 0.07399
    },
    {
      "epoch": 0.7971423199849595,
      "grad_norm": 0.5830900073051453,
      "learning_rate": 3.674534794592907e-05,
      "loss": 0.29729673862457273,
      "memory(GiB)": 79.65,
      "step": 530,
      "token_acc": 0.8755889890055386,
      "train_speed(iter/s)": 0.074001
    },
    {
      "epoch": 0.8046625305508555,
      "grad_norm": 0.37347549200057983,
      "learning_rate": 3.664884631716207e-05,
      "loss": 0.29659602642059324,
      "memory(GiB)": 79.65,
      "step": 535,
      "token_acc": 0.8784908147293846,
      "train_speed(iter/s)": 0.074017
    },
    {
      "epoch": 0.8121827411167513,
      "grad_norm": 0.28800126910209656,
      "learning_rate": 3.655106546878898e-05,
      "loss": 0.29447226524353026,
      "memory(GiB)": 79.65,
      "step": 540,
      "token_acc": 0.879452506596306,
      "train_speed(iter/s)": 0.074033
    },
    {
      "epoch": 0.8197029516826471,
      "grad_norm": 0.34387221932411194,
      "learning_rate": 3.645201291383359e-05,
      "loss": 0.2936647891998291,
      "memory(GiB)": 79.65,
      "step": 545,
      "token_acc": 0.8770972807670055,
      "train_speed(iter/s)": 0.074048
    },
    {
      "epoch": 0.827223162248543,
      "grad_norm": 0.36597007513046265,
      "learning_rate": 3.635169626303168e-05,
      "loss": 0.29578680992126466,
      "memory(GiB)": 79.65,
      "step": 550,
      "token_acc": 0.8803475382705833,
      "train_speed(iter/s)": 0.074063
    },
    {
      "epoch": 0.8347433728144388,
      "grad_norm": 0.5027373433113098,
      "learning_rate": 3.625012322424623e-05,
      "loss": 0.29583015441894533,
      "memory(GiB)": 79.65,
      "step": 555,
      "token_acc": 0.8814833168153287,
      "train_speed(iter/s)": 0.074077
    },
    {
      "epoch": 0.8422635833803347,
      "grad_norm": 0.44290775060653687,
      "learning_rate": 3.6147301601875254e-05,
      "loss": 0.2953122615814209,
      "memory(GiB)": 79.65,
      "step": 560,
      "token_acc": 0.8778594433892146,
      "train_speed(iter/s)": 0.074091
    },
    {
      "epoch": 0.8497837939462305,
      "grad_norm": 0.38042429089546204,
      "learning_rate": 3.6043239296252044e-05,
      "loss": 0.29462323188781736,
      "memory(GiB)": 79.65,
      "step": 565,
      "token_acc": 0.8821047406304294,
      "train_speed(iter/s)": 0.074105
    },
    {
      "epoch": 0.8573040045121263,
      "grad_norm": 0.3541141450405121,
      "learning_rate": 3.593794430303824e-05,
      "loss": 0.29390745162963866,
      "memory(GiB)": 79.65,
      "step": 570,
      "token_acc": 0.8788178150983782,
      "train_speed(iter/s)": 0.07412
    },
    {
      "epoch": 0.8648242150780222,
      "grad_norm": 0.3094688355922699,
      "learning_rate": 3.583142471260943e-05,
      "loss": 0.2885639190673828,
      "memory(GiB)": 79.65,
      "step": 575,
      "token_acc": 0.8854853518070278,
      "train_speed(iter/s)": 0.074133
    },
    {
      "epoch": 0.872344425643918,
      "grad_norm": 0.29323917627334595,
      "learning_rate": 3.5723688709433514e-05,
      "loss": 0.29495530128479003,
      "memory(GiB)": 79.65,
      "step": 580,
      "token_acc": 0.8780931887776214,
      "train_speed(iter/s)": 0.074147
    },
    {
      "epoch": 0.8798646362098139,
      "grad_norm": 0.32218337059020996,
      "learning_rate": 3.561474457144189e-05,
      "loss": 0.29918394088745115,
      "memory(GiB)": 79.65,
      "step": 585,
      "token_acc": 0.8818481848184818,
      "train_speed(iter/s)": 0.07416
    },
    {
      "epoch": 0.8873848467757097,
      "grad_norm": 0.3082771599292755,
      "learning_rate": 3.550460066939339e-05,
      "loss": 0.2934062719345093,
      "memory(GiB)": 79.65,
      "step": 590,
      "token_acc": 0.8800066203243959,
      "train_speed(iter/s)": 0.074173
    },
    {
      "epoch": 0.8949050573416055,
      "grad_norm": 0.3421003520488739,
      "learning_rate": 3.539326546623111e-05,
      "loss": 0.2875396490097046,
      "memory(GiB)": 79.65,
      "step": 595,
      "token_acc": 0.8823674475955611,
      "train_speed(iter/s)": 0.074186
    },
    {
      "epoch": 0.9024252679075014,
      "grad_norm": 0.3828553855419159,
      "learning_rate": 3.528074751643215e-05,
      "loss": 0.2884918212890625,
      "memory(GiB)": 79.65,
      "step": 600,
      "token_acc": 0.8880746169220519,
      "train_speed(iter/s)": 0.074199
    },
    {
      "epoch": 0.9099454784733972,
      "grad_norm": 0.3442157208919525,
      "learning_rate": 3.516705546535036e-05,
      "loss": 0.2889647245407104,
      "memory(GiB)": 79.65,
      "step": 605,
      "token_acc": 0.8830477457993678,
      "train_speed(iter/s)": 0.073663
    },
    {
      "epoch": 0.9174656890392932,
      "grad_norm": 0.3103463351726532,
      "learning_rate": 3.5052198048552034e-05,
      "loss": 0.2901792526245117,
      "memory(GiB)": 79.65,
      "step": 610,
      "token_acc": 0.8818009712733559,
      "train_speed(iter/s)": 0.073679
    },
    {
      "epoch": 0.924985899605189,
      "grad_norm": 0.40294262766838074,
      "learning_rate": 3.493618409114473e-05,
      "loss": 0.28948040008544923,
      "memory(GiB)": 79.65,
      "step": 615,
      "token_acc": 0.8835411471321696,
      "train_speed(iter/s)": 0.073696
    },
    {
      "epoch": 0.9325061101710848,
      "grad_norm": 0.31962162256240845,
      "learning_rate": 3.4819022507099184e-05,
      "loss": 0.2874648094177246,
      "memory(GiB)": 79.65,
      "step": 620,
      "token_acc": 0.8883991683991684,
      "train_speed(iter/s)": 0.073712
    },
    {
      "epoch": 0.9400263207369807,
      "grad_norm": 0.4923133850097656,
      "learning_rate": 3.470072229856441e-05,
      "loss": 0.29647414684295653,
      "memory(GiB)": 79.65,
      "step": 625,
      "token_acc": 0.8791515948701085,
      "train_speed(iter/s)": 0.073728
    },
    {
      "epoch": 0.9475465313028765,
      "grad_norm": 0.3678079843521118,
      "learning_rate": 3.458129255517601e-05,
      "loss": 0.2929518699645996,
      "memory(GiB)": 79.65,
      "step": 630,
      "token_acc": 0.8755544603252834,
      "train_speed(iter/s)": 0.073743
    },
    {
      "epoch": 0.9550667418687723,
      "grad_norm": 0.2857828736305237,
      "learning_rate": 3.446074245335775e-05,
      "loss": 0.291016960144043,
      "memory(GiB)": 79.65,
      "step": 635,
      "token_acc": 0.8845577211394303,
      "train_speed(iter/s)": 0.073758
    },
    {
      "epoch": 0.9625869524346682,
      "grad_norm": 0.3498782813549042,
      "learning_rate": 3.433908125561655e-05,
      "loss": 0.29367480278015134,
      "memory(GiB)": 79.65,
      "step": 640,
      "token_acc": 0.8839152738705941,
      "train_speed(iter/s)": 0.073772
    },
    {
      "epoch": 0.970107163000564,
      "grad_norm": 0.331845760345459,
      "learning_rate": 3.4216318309830695e-05,
      "loss": 0.2898298740386963,
      "memory(GiB)": 79.65,
      "step": 645,
      "token_acc": 0.8792733278282411,
      "train_speed(iter/s)": 0.073785
    },
    {
      "epoch": 0.9776273735664599,
      "grad_norm": 0.30705392360687256,
      "learning_rate": 3.40924630485317e-05,
      "loss": 0.28896470069885255,
      "memory(GiB)": 79.65,
      "step": 650,
      "token_acc": 0.8820786142571619,
      "train_speed(iter/s)": 0.073798
    },
    {
      "epoch": 0.9851475841323557,
      "grad_norm": 0.2710285782814026,
      "learning_rate": 3.3967524988179463e-05,
      "loss": 0.2907625675201416,
      "memory(GiB)": 79.65,
      "step": 655,
      "token_acc": 0.8820266578359136,
      "train_speed(iter/s)": 0.073812
    },
    {
      "epoch": 0.9926677946982515,
      "grad_norm": 0.3244026303291321,
      "learning_rate": 3.3841513728431116e-05,
      "loss": 0.2880063772201538,
      "memory(GiB)": 79.65,
      "step": 660,
      "token_acc": 0.8813882216516193,
      "train_speed(iter/s)": 0.073828
    },
    {
      "epoch": 1.0015040421131791,
      "grad_norm": 0.6774662137031555,
      "learning_rate": 3.371443895140343e-05,
      "loss": 0.3466364622116089,
      "memory(GiB)": 79.65,
      "step": 665,
      "token_acc": 0.8889412134055412,
      "train_speed(iter/s)": 0.073767
    },
    {
      "epoch": 1.009024252679075,
      "grad_norm": 0.29805055260658264,
      "learning_rate": 3.3586310420928884e-05,
      "loss": 0.2910954475402832,
      "memory(GiB)": 79.65,
      "step": 670,
      "token_acc": 0.8801392572944297,
      "train_speed(iter/s)": 0.073781
    },
    {
      "epoch": 1.016544463244971,
      "grad_norm": 0.2707737684249878,
      "learning_rate": 3.3457137981805427e-05,
      "loss": 0.28925237655639646,
      "memory(GiB)": 79.65,
      "step": 675,
      "token_acc": 0.8832292967124428,
      "train_speed(iter/s)": 0.073796
    },
    {
      "epoch": 1.0240646738108667,
      "grad_norm": 0.24032209813594818,
      "learning_rate": 3.3326931559040084e-05,
      "loss": 0.2896096229553223,
      "memory(GiB)": 79.65,
      "step": 680,
      "token_acc": 0.8813700224457561,
      "train_speed(iter/s)": 0.07381
    },
    {
      "epoch": 1.0315848843767625,
      "grad_norm": 0.35086068511009216,
      "learning_rate": 3.3195701157086354e-05,
      "loss": 0.28840909004211424,
      "memory(GiB)": 79.65,
      "step": 685,
      "token_acc": 0.8864882831976068,
      "train_speed(iter/s)": 0.073824
    },
    {
      "epoch": 1.0391050949426583,
      "grad_norm": 0.23582202196121216,
      "learning_rate": 3.306345685907553e-05,
      "loss": 0.2879499912261963,
      "memory(GiB)": 79.65,
      "step": 690,
      "token_acc": 0.8865782932891466,
      "train_speed(iter/s)": 0.073838
    },
    {
      "epoch": 1.0466253055085541,
      "grad_norm": 0.19086644053459167,
      "learning_rate": 3.293020882604194e-05,
      "loss": 0.2861684799194336,
      "memory(GiB)": 79.65,
      "step": 695,
      "token_acc": 0.8828125,
      "train_speed(iter/s)": 0.073851
    },
    {
      "epoch": 1.0541455160744502,
      "grad_norm": 0.2418331801891327,
      "learning_rate": 3.2795967296142185e-05,
      "loss": 0.2888832092285156,
      "memory(GiB)": 79.65,
      "step": 700,
      "token_acc": 0.881951299769661,
      "train_speed(iter/s)": 0.073864
    },
    {
      "epoch": 1.061665726640346,
      "grad_norm": 0.33799418807029724,
      "learning_rate": 3.2660742583868585e-05,
      "loss": 0.28574295043945314,
      "memory(GiB)": 79.65,
      "step": 705,
      "token_acc": 0.8785093167701863,
      "train_speed(iter/s)": 0.073878
    },
    {
      "epoch": 1.0691859372062418,
      "grad_norm": 0.246758371591568,
      "learning_rate": 3.2524545079256554e-05,
      "loss": 0.2918140411376953,
      "memory(GiB)": 79.65,
      "step": 710,
      "token_acc": 0.8831544178364987,
      "train_speed(iter/s)": 0.073891
    },
    {
      "epoch": 1.0767061477721376,
      "grad_norm": 0.2471436709165573,
      "learning_rate": 3.238738524708636e-05,
      "loss": 0.2883563280105591,
      "memory(GiB)": 79.65,
      "step": 715,
      "token_acc": 0.8862906545903664,
      "train_speed(iter/s)": 0.073903
    },
    {
      "epoch": 1.0842263583380334,
      "grad_norm": 0.20913682878017426,
      "learning_rate": 3.2249273626079005e-05,
      "loss": 0.28980541229248047,
      "memory(GiB)": 79.65,
      "step": 720,
      "token_acc": 0.8819993412384717,
      "train_speed(iter/s)": 0.073916
    },
    {
      "epoch": 1.0917465689039294,
      "grad_norm": 0.2716258764266968,
      "learning_rate": 3.211022082808652e-05,
      "loss": 0.28816471099853513,
      "memory(GiB)": 79.65,
      "step": 725,
      "token_acc": 0.8852621986579405,
      "train_speed(iter/s)": 0.073928
    },
    {
      "epoch": 1.0992667794698252,
      "grad_norm": 0.2366819679737091,
      "learning_rate": 3.197023753727655e-05,
      "loss": 0.2890036106109619,
      "memory(GiB)": 79.65,
      "step": 730,
      "token_acc": 0.8808816245666171,
      "train_speed(iter/s)": 0.073941
    },
    {
      "epoch": 1.106786990035721,
      "grad_norm": 0.30629608035087585,
      "learning_rate": 3.1829334509311496e-05,
      "loss": 0.286466646194458,
      "memory(GiB)": 79.65,
      "step": 735,
      "token_acc": 0.8861997513468711,
      "train_speed(iter/s)": 0.073953
    },
    {
      "epoch": 1.1143072006016168,
      "grad_norm": 0.24446699023246765,
      "learning_rate": 3.1687522570522055e-05,
      "loss": 0.28462073802947996,
      "memory(GiB)": 79.65,
      "step": 740,
      "token_acc": 0.880608052588332,
      "train_speed(iter/s)": 0.073965
    },
    {
      "epoch": 1.1218274111675126,
      "grad_norm": 0.22261127829551697,
      "learning_rate": 3.1544812617075384e-05,
      "loss": 0.2871187686920166,
      "memory(GiB)": 79.65,
      "step": 745,
      "token_acc": 0.8859590750291133,
      "train_speed(iter/s)": 0.073977
    },
    {
      "epoch": 1.1293476217334084,
      "grad_norm": 0.25827860832214355,
      "learning_rate": 3.140121561413789e-05,
      "loss": 0.2903823614120483,
      "memory(GiB)": 79.65,
      "step": 750,
      "token_acc": 0.8780008193363376,
      "train_speed(iter/s)": 0.073989
    },
    {
      "epoch": 1.1368678322993044,
      "grad_norm": 0.19528216123580933,
      "learning_rate": 3.125674259503273e-05,
      "loss": 0.2866347789764404,
      "memory(GiB)": 79.65,
      "step": 755,
      "token_acc": 0.885481592287875,
      "train_speed(iter/s)": 0.073999
    },
    {
      "epoch": 1.1443880428652002,
      "grad_norm": 0.25423479080200195,
      "learning_rate": 3.111140466039205e-05,
      "loss": 0.28595871925354005,
      "memory(GiB)": 79.65,
      "step": 760,
      "token_acc": 0.8885020310038962,
      "train_speed(iter/s)": 0.074011
    },
    {
      "epoch": 1.151908253431096,
      "grad_norm": 0.25926610827445984,
      "learning_rate": 3.096521297730406e-05,
      "loss": 0.2877946138381958,
      "memory(GiB)": 79.65,
      "step": 765,
      "token_acc": 0.8828176979374585,
      "train_speed(iter/s)": 0.074016
    },
    {
      "epoch": 1.1594284639969918,
      "grad_norm": 0.2528335154056549,
      "learning_rate": 3.081817877845502e-05,
      "loss": 0.2850656509399414,
      "memory(GiB)": 79.65,
      "step": 770,
      "token_acc": 0.8857425090549885,
      "train_speed(iter/s)": 0.074026
    },
    {
      "epoch": 1.1669486745628879,
      "grad_norm": 0.31983044743537903,
      "learning_rate": 3.0670313361266165e-05,
      "loss": 0.28856089115142824,
      "memory(GiB)": 79.65,
      "step": 775,
      "token_acc": 0.8833430063842136,
      "train_speed(iter/s)": 0.074037
    },
    {
      "epoch": 1.1744688851287837,
      "grad_norm": 0.22499898076057434,
      "learning_rate": 3.0521628087025675e-05,
      "loss": 0.28852667808532717,
      "memory(GiB)": 79.65,
      "step": 780,
      "token_acc": 0.8872593950504125,
      "train_speed(iter/s)": 0.074048
    },
    {
      "epoch": 1.1819890956946795,
      "grad_norm": 0.2510216534137726,
      "learning_rate": 3.037213438001569e-05,
      "loss": 0.28934569358825685,
      "memory(GiB)": 79.65,
      "step": 785,
      "token_acc": 0.8825226710634789,
      "train_speed(iter/s)": 0.074059
    },
    {
      "epoch": 1.1895093062605753,
      "grad_norm": 0.18021002411842346,
      "learning_rate": 3.0221843726634587e-05,
      "loss": 0.28499023914337157,
      "memory(GiB)": 79.65,
      "step": 790,
      "token_acc": 0.8781498673740054,
      "train_speed(iter/s)": 0.07407
    },
    {
      "epoch": 1.197029516826471,
      "grad_norm": 0.24111808836460114,
      "learning_rate": 3.0070767674514355e-05,
      "loss": 0.2870931148529053,
      "memory(GiB)": 79.65,
      "step": 795,
      "token_acc": 0.8837460528502576,
      "train_speed(iter/s)": 0.07408
    },
    {
      "epoch": 1.2045497273923669,
      "grad_norm": 0.33728495240211487,
      "learning_rate": 2.9918917831633362e-05,
      "loss": 0.288606071472168,
      "memory(GiB)": 79.65,
      "step": 800,
      "token_acc": 0.8792605430387059,
      "train_speed(iter/s)": 0.07409
    },
    {
      "epoch": 1.212069937958263,
      "grad_norm": 0.22697192430496216,
      "learning_rate": 2.9766305865424435e-05,
      "loss": 0.28791687488555906,
      "memory(GiB)": 79.65,
      "step": 805,
      "token_acc": 0.8871413034384673,
      "train_speed(iter/s)": 0.073659
    },
    {
      "epoch": 1.2195901485241587,
      "grad_norm": 0.2359168827533722,
      "learning_rate": 2.9612943501878412e-05,
      "loss": 0.29138979911804197,
      "memory(GiB)": 79.65,
      "step": 810,
      "token_acc": 0.8794761273209549,
      "train_speed(iter/s)": 0.073672
    },
    {
      "epoch": 1.2271103590900545,
      "grad_norm": 0.31455326080322266,
      "learning_rate": 2.945884252464316e-05,
      "loss": 0.29040348529815674,
      "memory(GiB)": 79.65,
      "step": 815,
      "token_acc": 0.882221671212497,
      "train_speed(iter/s)": 0.073684
    },
    {
      "epoch": 1.2346305696559503,
      "grad_norm": 0.2171451896429062,
      "learning_rate": 2.9304014774118153e-05,
      "loss": 0.28681347370147703,
      "memory(GiB)": 79.65,
      "step": 820,
      "token_acc": 0.8789892526048076,
      "train_speed(iter/s)": 0.073697
    },
    {
      "epoch": 1.2421507802218463,
      "grad_norm": 0.26972663402557373,
      "learning_rate": 2.9148472146544757e-05,
      "loss": 0.2849359273910522,
      "memory(GiB)": 79.65,
      "step": 825,
      "token_acc": 0.8843992728474632,
      "train_speed(iter/s)": 0.07371
    },
    {
      "epoch": 1.2496709907877421,
      "grad_norm": 0.22206652164459229,
      "learning_rate": 2.8992226593092135e-05,
      "loss": 0.28711600303649903,
      "memory(GiB)": 79.65,
      "step": 830,
      "token_acc": 0.8830157415078708,
      "train_speed(iter/s)": 0.073722
    },
    {
      "epoch": 1.257191201353638,
      "grad_norm": 0.2502540946006775,
      "learning_rate": 2.8835290118938995e-05,
      "loss": 0.28503079414367677,
      "memory(GiB)": 79.65,
      "step": 835,
      "token_acc": 0.8803887366060305,
      "train_speed(iter/s)": 0.073734
    },
    {
      "epoch": 1.2647114119195337,
      "grad_norm": 0.21333985030651093,
      "learning_rate": 2.8677674782351164e-05,
      "loss": 0.28586583137512206,
      "memory(GiB)": 79.65,
      "step": 840,
      "token_acc": 0.8851552179656539,
      "train_speed(iter/s)": 0.073746
    },
    {
      "epoch": 1.2722316224854295,
      "grad_norm": 0.24539323151111603,
      "learning_rate": 2.85193926937551e-05,
      "loss": 0.2865161418914795,
      "memory(GiB)": 79.65,
      "step": 845,
      "token_acc": 0.8874390948880998,
      "train_speed(iter/s)": 0.073758
    },
    {
      "epoch": 1.2797518330513253,
      "grad_norm": 0.29327914118766785,
      "learning_rate": 2.8360456014807353e-05,
      "loss": 0.2898313045501709,
      "memory(GiB)": 79.65,
      "step": 850,
      "token_acc": 0.8840854714957512,
      "train_speed(iter/s)": 0.07377
    },
    {
      "epoch": 1.2872720436172214,
      "grad_norm": 0.26839759945869446,
      "learning_rate": 2.820087695746015e-05,
      "loss": 0.28841476440429686,
      "memory(GiB)": 79.65,
      "step": 855,
      "token_acc": 0.8845802540895126,
      "train_speed(iter/s)": 0.073781
    },
    {
      "epoch": 1.2947922541831172,
      "grad_norm": 0.27348655462265015,
      "learning_rate": 2.804066778302307e-05,
      "loss": 0.2830035209655762,
      "memory(GiB)": 79.65,
      "step": 860,
      "token_acc": 0.8877390512459641,
      "train_speed(iter/s)": 0.073792
    },
    {
      "epoch": 1.302312464749013,
      "grad_norm": 0.2061777263879776,
      "learning_rate": 2.7879840801220967e-05,
      "loss": 0.2841815710067749,
      "memory(GiB)": 79.65,
      "step": 865,
      "token_acc": 0.8776635129576306,
      "train_speed(iter/s)": 0.073803
    },
    {
      "epoch": 1.3098326753149088,
      "grad_norm": 0.1932431012392044,
      "learning_rate": 2.7718408369248093e-05,
      "loss": 0.2798025131225586,
      "memory(GiB)": 79.65,
      "step": 870,
      "token_acc": 0.8887501040886002,
      "train_speed(iter/s)": 0.073814
    },
    {
      "epoch": 1.3173528858808048,
      "grad_norm": 0.29601460695266724,
      "learning_rate": 2.7556382890818682e-05,
      "loss": 0.28728861808776857,
      "memory(GiB)": 79.65,
      "step": 875,
      "token_acc": 0.8841095210521962,
      "train_speed(iter/s)": 0.073825
    },
    {
      "epoch": 1.3248730964467006,
      "grad_norm": 0.29293093085289,
      "learning_rate": 2.7393776815213903e-05,
      "loss": 0.28348283767700194,
      "memory(GiB)": 79.65,
      "step": 880,
      "token_acc": 0.8797128949756621,
      "train_speed(iter/s)": 0.073836
    },
    {
      "epoch": 1.3323933070125964,
      "grad_norm": 0.26017335057258606,
      "learning_rate": 2.723060263632527e-05,
      "loss": 0.2872117519378662,
      "memory(GiB)": 79.65,
      "step": 885,
      "token_acc": 0.8814147591108172,
      "train_speed(iter/s)": 0.073844
    },
    {
      "epoch": 1.3399135175784922,
      "grad_norm": 0.21063359081745148,
      "learning_rate": 2.706687289169472e-05,
      "loss": 0.2855399131774902,
      "memory(GiB)": 79.65,
      "step": 890,
      "token_acc": 0.8832867247902616,
      "train_speed(iter/s)": 0.073854
    },
    {
      "epoch": 1.347433728144388,
      "grad_norm": 0.22006599605083466,
      "learning_rate": 2.6902600161551273e-05,
      "loss": 0.2876161575317383,
      "memory(GiB)": 79.65,
      "step": 895,
      "token_acc": 0.8865525265957447,
      "train_speed(iter/s)": 0.073864
    },
    {
      "epoch": 1.3549539387102838,
      "grad_norm": 0.22089773416519165,
      "learning_rate": 2.6737797067844403e-05,
      "loss": 0.28572721481323243,
      "memory(GiB)": 79.65,
      "step": 900,
      "token_acc": 0.8883084577114427,
      "train_speed(iter/s)": 0.073874
    },
    {
      "epoch": 1.3624741492761796,
      "grad_norm": 0.3232302665710449,
      "learning_rate": 2.6572476273274253e-05,
      "loss": 0.2867557525634766,
      "memory(GiB)": 79.65,
      "step": 905,
      "token_acc": 0.8833759169207945,
      "train_speed(iter/s)": 0.073884
    },
    {
      "epoch": 1.3699943598420756,
      "grad_norm": 0.24157921969890594,
      "learning_rate": 2.6406650480318675e-05,
      "loss": 0.29031193256378174,
      "memory(GiB)": 79.65,
      "step": 910,
      "token_acc": 0.8816269052352551,
      "train_speed(iter/s)": 0.073894
    },
    {
      "epoch": 1.3775145704079714,
      "grad_norm": 0.2636571228504181,
      "learning_rate": 2.6240332430257254e-05,
      "loss": 0.2877202033996582,
      "memory(GiB)": 79.65,
      "step": 915,
      "token_acc": 0.8873626373626373,
      "train_speed(iter/s)": 0.073904
    },
    {
      "epoch": 1.3850347809738672,
      "grad_norm": 0.20964060723781586,
      "learning_rate": 2.6073534902192295e-05,
      "loss": 0.2886173725128174,
      "memory(GiB)": 79.65,
      "step": 920,
      "token_acc": 0.885331347728965,
      "train_speed(iter/s)": 0.073913
    },
    {
      "epoch": 1.392554991539763,
      "grad_norm": 0.23985759913921356,
      "learning_rate": 2.5906270712066972e-05,
      "loss": 0.28805909156799314,
      "memory(GiB)": 79.65,
      "step": 925,
      "token_acc": 0.8773157678056813,
      "train_speed(iter/s)": 0.073923
    },
    {
      "epoch": 1.400075202105659,
      "grad_norm": 0.2342899739742279,
      "learning_rate": 2.5738552711680575e-05,
      "loss": 0.2863925457000732,
      "memory(GiB)": 79.65,
      "step": 930,
      "token_acc": 0.8833250290649394,
      "train_speed(iter/s)": 0.073933
    },
    {
      "epoch": 1.4075954126715549,
      "grad_norm": 0.22644104063510895,
      "learning_rate": 2.5570393787701063e-05,
      "loss": 0.285622239112854,
      "memory(GiB)": 79.65,
      "step": 935,
      "token_acc": 0.8857380457380457,
      "train_speed(iter/s)": 0.073942
    },
    {
      "epoch": 1.4151156232374507,
      "grad_norm": 0.21371813118457794,
      "learning_rate": 2.5401806860674912e-05,
      "loss": 0.28616771697998045,
      "memory(GiB)": 79.65,
      "step": 940,
      "token_acc": 0.8805736421330256,
      "train_speed(iter/s)": 0.073952
    },
    {
      "epoch": 1.4226358338033465,
      "grad_norm": 0.2841584384441376,
      "learning_rate": 2.5232804884034336e-05,
      "loss": 0.2837015390396118,
      "memory(GiB)": 79.65,
      "step": 945,
      "token_acc": 0.8856763044200732,
      "train_speed(iter/s)": 0.073962
    },
    {
      "epoch": 1.4301560443692423,
      "grad_norm": 0.3443674147129059,
      "learning_rate": 2.506340084310204e-05,
      "loss": 0.28619914054870604,
      "memory(GiB)": 79.65,
      "step": 950,
      "token_acc": 0.8829568788501027,
      "train_speed(iter/s)": 0.073971
    },
    {
      "epoch": 1.437676254935138,
      "grad_norm": 0.28178870677948,
      "learning_rate": 2.4893607754093467e-05,
      "loss": 0.283730673789978,
      "memory(GiB)": 79.65,
      "step": 955,
      "token_acc": 0.888511273209549,
      "train_speed(iter/s)": 0.07398
    },
    {
      "epoch": 1.445196465501034,
      "grad_norm": 0.2970364987850189,
      "learning_rate": 2.4723438663116692e-05,
      "loss": 0.28200397491455076,
      "memory(GiB)": 79.65,
      "step": 960,
      "token_acc": 0.884932410154962,
      "train_speed(iter/s)": 0.073989
    },
    {
      "epoch": 1.45271667606693,
      "grad_norm": 0.2049390822649002,
      "learning_rate": 2.4552906645170024e-05,
      "loss": 0.28285489082336424,
      "memory(GiB)": 79.65,
      "step": 965,
      "token_acc": 0.8924766977363515,
      "train_speed(iter/s)": 0.073998
    },
    {
      "epoch": 1.4602368866328257,
      "grad_norm": 0.18950092792510986,
      "learning_rate": 2.4382024803137396e-05,
      "loss": 0.27791404724121094,
      "memory(GiB)": 79.65,
      "step": 970,
      "token_acc": 0.8823285512894933,
      "train_speed(iter/s)": 0.074007
    },
    {
      "epoch": 1.4677570971987215,
      "grad_norm": 0.24535216391086578,
      "learning_rate": 2.4210806266781594e-05,
      "loss": 0.2802283525466919,
      "memory(GiB)": 79.65,
      "step": 975,
      "token_acc": 0.8822616226815916,
      "train_speed(iter/s)": 0.074017
    },
    {
      "epoch": 1.4752773077646175,
      "grad_norm": 0.25220349431037903,
      "learning_rate": 2.4039264191735395e-05,
      "loss": 0.2870103597640991,
      "memory(GiB)": 79.65,
      "step": 980,
      "token_acc": 0.8847685415629667,
      "train_speed(iter/s)": 0.074026
    },
    {
      "epoch": 1.4827975183305133,
      "grad_norm": 0.2624325752258301,
      "learning_rate": 2.3867411758490816e-05,
      "loss": 0.28549647331237793,
      "memory(GiB)": 79.65,
      "step": 985,
      "token_acc": 0.87865833607366,
      "train_speed(iter/s)": 0.074035
    },
    {
      "epoch": 1.4903177288964091,
      "grad_norm": 0.24539606273174286,
      "learning_rate": 2.3695262171386318e-05,
      "loss": 0.279841947555542,
      "memory(GiB)": 79.65,
      "step": 990,
      "token_acc": 0.8891733421422082,
      "train_speed(iter/s)": 0.074043
    },
    {
      "epoch": 1.497837939462305,
      "grad_norm": 0.26320719718933105,
      "learning_rate": 2.3522828657592295e-05,
      "loss": 0.2846808910369873,
      "memory(GiB)": 79.65,
      "step": 995,
      "token_acc": 0.8851200792275316,
      "train_speed(iter/s)": 0.074052
    },
    {
      "epoch": 1.5053581500282007,
      "grad_norm": 0.1951611340045929,
      "learning_rate": 2.335012446609473e-05,
      "loss": 0.2793398380279541,
      "memory(GiB)": 79.65,
      "step": 1000,
      "token_acc": 0.8877635729702806,
      "train_speed(iter/s)": 0.074059
    },
    {
      "epoch": 1.5128783605940965,
      "grad_norm": 0.277101069688797,
      "learning_rate": 2.317716286667723e-05,
      "loss": 0.2791220903396606,
      "memory(GiB)": 79.65,
      "step": 1005,
      "token_acc": 0.8879381957135737,
      "train_speed(iter/s)": 0.073737
    },
    {
      "epoch": 1.5203985711599923,
      "grad_norm": 0.23428307473659515,
      "learning_rate": 2.300395714890142e-05,
      "loss": 0.2815703868865967,
      "memory(GiB)": 79.65,
      "step": 1010,
      "token_acc": 0.8851039260969977,
      "train_speed(iter/s)": 0.073747
    },
    {
      "epoch": 1.5279187817258884,
      "grad_norm": 0.2545008659362793,
      "learning_rate": 2.283052062108582e-05,
      "loss": 0.2828678607940674,
      "memory(GiB)": 79.65,
      "step": 1015,
      "token_acc": 0.8790747448139612,
      "train_speed(iter/s)": 0.073757
    },
    {
      "epoch": 1.5354389922917842,
      "grad_norm": 0.19433118402957916,
      "learning_rate": 2.265686660928336e-05,
      "loss": 0.281368088722229,
      "memory(GiB)": 79.65,
      "step": 1020,
      "token_acc": 0.8873110778940375,
      "train_speed(iter/s)": 0.073767
    },
    {
      "epoch": 1.5429592028576802,
      "grad_norm": 0.21996212005615234,
      "learning_rate": 2.248300845625737e-05,
      "loss": 0.2805771827697754,
      "memory(GiB)": 79.65,
      "step": 1025,
      "token_acc": 0.883298969072165,
      "train_speed(iter/s)": 0.073777
    },
    {
      "epoch": 1.550479413423576,
      "grad_norm": 0.2305791974067688,
      "learning_rate": 2.2308959520456485e-05,
      "loss": 0.2788412570953369,
      "memory(GiB)": 79.65,
      "step": 1030,
      "token_acc": 0.8869371598218704,
      "train_speed(iter/s)": 0.073786
    },
    {
      "epoch": 1.5579996239894718,
      "grad_norm": 0.29013922810554504,
      "learning_rate": 2.2134733174988164e-05,
      "loss": 0.281440806388855,
      "memory(GiB)": 79.65,
      "step": 1035,
      "token_acc": 0.8883009508061183,
      "train_speed(iter/s)": 0.073796
    },
    {
      "epoch": 1.5655198345553676,
      "grad_norm": 0.2971268594264984,
      "learning_rate": 2.196034280659122e-05,
      "loss": 0.2808208703994751,
      "memory(GiB)": 79.65,
      "step": 1040,
      "token_acc": 0.8855316676350958,
      "train_speed(iter/s)": 0.073805
    },
    {
      "epoch": 1.5730400451212634,
      "grad_norm": 0.25820252299308777,
      "learning_rate": 2.17858018146072e-05,
      "loss": 0.27763383388519286,
      "memory(GiB)": 79.65,
      "step": 1045,
      "token_acc": 0.8882445530610554,
      "train_speed(iter/s)": 0.073814
    },
    {
      "epoch": 1.5805602556871592,
      "grad_norm": 0.2027405947446823,
      "learning_rate": 2.1611123609950885e-05,
      "loss": 0.2834439277648926,
      "memory(GiB)": 79.65,
      "step": 1050,
      "token_acc": 0.8827666087532059,
      "train_speed(iter/s)": 0.073823
    },
    {
      "epoch": 1.588080466253055,
      "grad_norm": 0.18557527661323547,
      "learning_rate": 2.1436321614079833e-05,
      "loss": 0.28252296447753905,
      "memory(GiB)": 79.65,
      "step": 1055,
      "token_acc": 0.8841940532081377,
      "train_speed(iter/s)": 0.073832
    },
    {
      "epoch": 1.5956006768189508,
      "grad_norm": 0.2463424950838089,
      "learning_rate": 2.1261409257963122e-05,
      "loss": 0.2786983013153076,
      "memory(GiB)": 79.65,
      "step": 1060,
      "token_acc": 0.8829839907575507,
      "train_speed(iter/s)": 0.073841
    },
    {
      "epoch": 1.6031208873848468,
      "grad_norm": 0.2423444241285324,
      "learning_rate": 2.1086399981049422e-05,
      "loss": 0.2816389799118042,
      "memory(GiB)": 79.65,
      "step": 1065,
      "token_acc": 0.8848049452844374,
      "train_speed(iter/s)": 0.073851
    },
    {
      "epoch": 1.6106410979507426,
      "grad_norm": 0.22321483492851257,
      "learning_rate": 2.0911307230234326e-05,
      "loss": 0.27858695983886717,
      "memory(GiB)": 79.65,
      "step": 1070,
      "token_acc": 0.8847070424850029,
      "train_speed(iter/s)": 0.073859
    },
    {
      "epoch": 1.6181613085166384,
      "grad_norm": 0.21224167943000793,
      "learning_rate": 2.073614445882718e-05,
      "loss": 0.279494571685791,
      "memory(GiB)": 79.65,
      "step": 1075,
      "token_acc": 0.8850906095551895,
      "train_speed(iter/s)": 0.073868
    },
    {
      "epoch": 1.6256815190825344,
      "grad_norm": 0.22231175005435944,
      "learning_rate": 2.056092512551738e-05,
      "loss": 0.2793027400970459,
      "memory(GiB)": 79.65,
      "step": 1080,
      "token_acc": 0.8833002973240832,
      "train_speed(iter/s)": 0.073877
    },
    {
      "epoch": 1.6332017296484302,
      "grad_norm": 0.21154510974884033,
      "learning_rate": 2.0385662693340286e-05,
      "loss": 0.27746267318725587,
      "memory(GiB)": 79.65,
      "step": 1085,
      "token_acc": 0.8888336234767471,
      "train_speed(iter/s)": 0.073886
    },
    {
      "epoch": 1.640721940214326,
      "grad_norm": 0.2765055298805237,
      "learning_rate": 2.0210370628642758e-05,
      "loss": 0.2778115749359131,
      "memory(GiB)": 79.65,
      "step": 1090,
      "token_acc": 0.8852095316351685,
      "train_speed(iter/s)": 0.073894
    },
    {
      "epoch": 1.6482421507802218,
      "grad_norm": 0.17052045464515686,
      "learning_rate": 2.0035062400048487e-05,
      "loss": 0.2814920663833618,
      "memory(GiB)": 79.65,
      "step": 1095,
      "token_acc": 0.8821211623707109,
      "train_speed(iter/s)": 0.073903
    },
    {
      "epoch": 1.6557623613461177,
      "grad_norm": 0.2029465287923813,
      "learning_rate": 1.985975147742313e-05,
      "loss": 0.27739577293395995,
      "memory(GiB)": 79.65,
      "step": 1100,
      "token_acc": 0.8891664584894662,
      "train_speed(iter/s)": 0.073911
    },
    {
      "epoch": 1.6632825719120135,
      "grad_norm": 0.27032309770584106,
      "learning_rate": 1.9684451330839336e-05,
      "loss": 0.2813339948654175,
      "memory(GiB)": 79.65,
      "step": 1105,
      "token_acc": 0.8828234518017647,
      "train_speed(iter/s)": 0.073918
    },
    {
      "epoch": 1.6708027824779093,
      "grad_norm": 0.28053125739097595,
      "learning_rate": 1.950917542954176e-05,
      "loss": 0.2807764053344727,
      "memory(GiB)": 79.65,
      "step": 1110,
      "token_acc": 0.8848240710292666,
      "train_speed(iter/s)": 0.073926
    },
    {
      "epoch": 1.6783229930438053,
      "grad_norm": 0.25592294335365295,
      "learning_rate": 1.933393724091218e-05,
      "loss": 0.28082756996154784,
      "memory(GiB)": 79.65,
      "step": 1115,
      "token_acc": 0.8885392514077509,
      "train_speed(iter/s)": 0.073934
    },
    {
      "epoch": 1.685843203609701,
      "grad_norm": 0.29153063893318176,
      "learning_rate": 1.9158750229434703e-05,
      "loss": 0.2801581859588623,
      "memory(GiB)": 79.65,
      "step": 1120,
      "token_acc": 0.8843475377217709,
      "train_speed(iter/s)": 0.073939
    },
    {
      "epoch": 1.6933634141755969,
      "grad_norm": 0.26778560876846313,
      "learning_rate": 1.898362785566122e-05,
      "loss": 0.2771186351776123,
      "memory(GiB)": 79.65,
      "step": 1125,
      "token_acc": 0.8802682785459965,
      "train_speed(iter/s)": 0.073948
    },
    {
      "epoch": 1.700883624741493,
      "grad_norm": 0.2282964140176773,
      "learning_rate": 1.8808583575177172e-05,
      "loss": 0.279494309425354,
      "memory(GiB)": 79.65,
      "step": 1130,
      "token_acc": 0.8869860180359064,
      "train_speed(iter/s)": 0.073955
    },
    {
      "epoch": 1.7084038353073887,
      "grad_norm": 0.2220524549484253,
      "learning_rate": 1.863363083756766e-05,
      "loss": 0.28255314826965333,
      "memory(GiB)": 79.65,
      "step": 1135,
      "token_acc": 0.8846089988377884,
      "train_speed(iter/s)": 0.073963
    },
    {
      "epoch": 1.7159240458732845,
      "grad_norm": 0.21929582953453064,
      "learning_rate": 1.845878308538408e-05,
      "loss": 0.2789367437362671,
      "memory(GiB)": 79.65,
      "step": 1140,
      "token_acc": 0.8845804801584027,
      "train_speed(iter/s)": 0.073971
    },
    {
      "epoch": 1.7234442564391803,
      "grad_norm": 0.22542671859264374,
      "learning_rate": 1.8284053753111205e-05,
      "loss": 0.2753563165664673,
      "memory(GiB)": 79.65,
      "step": 1145,
      "token_acc": 0.8852038706475891,
      "train_speed(iter/s)": 0.073979
    },
    {
      "epoch": 1.7309644670050761,
      "grad_norm": 0.2623978853225708,
      "learning_rate": 1.810945626613501e-05,
      "loss": 0.2741701602935791,
      "memory(GiB)": 79.65,
      "step": 1150,
      "token_acc": 0.8845521774856204,
      "train_speed(iter/s)": 0.073987
    },
    {
      "epoch": 1.738484677570972,
      "grad_norm": 0.20295007526874542,
      "learning_rate": 1.7935004039711078e-05,
      "loss": 0.28223118782043455,
      "memory(GiB)": 79.65,
      "step": 1155,
      "token_acc": 0.8845294263719891,
      "train_speed(iter/s)": 0.073994
    },
    {
      "epoch": 1.7460048881368677,
      "grad_norm": 0.24625885486602783,
      "learning_rate": 1.7760710477933846e-05,
      "loss": 0.27779061794281007,
      "memory(GiB)": 79.65,
      "step": 1160,
      "token_acc": 0.8870673952641166,
      "train_speed(iter/s)": 0.074002
    },
    {
      "epoch": 1.7535250987027635,
      "grad_norm": 0.22748979926109314,
      "learning_rate": 1.7586588972706703e-05,
      "loss": 0.27157177925109866,
      "memory(GiB)": 79.65,
      "step": 1165,
      "token_acc": 0.8863768115942029,
      "train_speed(iter/s)": 0.07401
    },
    {
      "epoch": 1.7610453092686595,
      "grad_norm": 0.29340866208076477,
      "learning_rate": 1.741265290271302e-05,
      "loss": 0.27838060855865476,
      "memory(GiB)": 79.65,
      "step": 1170,
      "token_acc": 0.8869809295797904,
      "train_speed(iter/s)": 0.074017
    },
    {
      "epoch": 1.7685655198345553,
      "grad_norm": 0.2462192326784134,
      "learning_rate": 1.7238915632388198e-05,
      "loss": 0.27757294178009034,
      "memory(GiB)": 79.65,
      "step": 1175,
      "token_acc": 0.8902651924515753,
      "train_speed(iter/s)": 0.074024
    },
    {
      "epoch": 1.7760857304004514,
      "grad_norm": 0.2414316087961197,
      "learning_rate": 1.7065390510892767e-05,
      "loss": 0.27706668376922605,
      "memory(GiB)": 79.65,
      "step": 1180,
      "token_acc": 0.8918581418581418,
      "train_speed(iter/s)": 0.074032
    },
    {
      "epoch": 1.7836059409663472,
      "grad_norm": 0.2678639590740204,
      "learning_rate": 1.6892090871086772e-05,
      "loss": 0.2797917604446411,
      "memory(GiB)": 79.65,
      "step": 1185,
      "token_acc": 0.8833689751603883,
      "train_speed(iter/s)": 0.074039
    },
    {
      "epoch": 1.791126151532243,
      "grad_norm": 0.27251312136650085,
      "learning_rate": 1.671903002850528e-05,
      "loss": 0.2821605920791626,
      "memory(GiB)": 79.65,
      "step": 1190,
      "token_acc": 0.8858474082702388,
      "train_speed(iter/s)": 0.074046
    },
    {
      "epoch": 1.7986463620981388,
      "grad_norm": 0.23939917981624603,
      "learning_rate": 1.6546221280335283e-05,
      "loss": 0.27451205253601074,
      "memory(GiB)": 79.65,
      "step": 1195,
      "token_acc": 0.884056775045387,
      "train_speed(iter/s)": 0.074053
    },
    {
      "epoch": 1.8061665726640346,
      "grad_norm": 0.2269574999809265,
      "learning_rate": 1.6373677904394038e-05,
      "loss": 0.27640342712402344,
      "memory(GiB)": 79.65,
      "step": 1200,
      "token_acc": 0.8850249584026623,
      "train_speed(iter/s)": 0.07406
    },
    {
      "epoch": 1.8136867832299304,
      "grad_norm": 0.19086390733718872,
      "learning_rate": 1.6201413158108823e-05,
      "loss": 0.2743852138519287,
      "memory(GiB)": 79.65,
      "step": 1205,
      "token_acc": 0.8870206245340844,
      "train_speed(iter/s)": 0.073793
    },
    {
      "epoch": 1.8212069937958262,
      "grad_norm": 0.2674992084503174,
      "learning_rate": 1.6029440277498333e-05,
      "loss": 0.27567086219787595,
      "memory(GiB)": 79.65,
      "step": 1210,
      "token_acc": 0.8854235605559232,
      "train_speed(iter/s)": 0.073801
    },
    {
      "epoch": 1.828727204361722,
      "grad_norm": 0.1904677003622055,
      "learning_rate": 1.5857772476155634e-05,
      "loss": 0.2795327425003052,
      "memory(GiB)": 79.65,
      "step": 1215,
      "token_acc": 0.8822468224682247,
      "train_speed(iter/s)": 0.073809
    },
    {
      "epoch": 1.836247414927618,
      "grad_norm": 0.22527740895748138,
      "learning_rate": 1.5686422944232952e-05,
      "loss": 0.27476723194122316,
      "memory(GiB)": 79.65,
      "step": 1220,
      "token_acc": 0.8880664528332922,
      "train_speed(iter/s)": 0.073816
    },
    {
      "epoch": 1.8437676254935138,
      "grad_norm": 0.22377359867095947,
      "learning_rate": 1.5515404847428185e-05,
      "loss": 0.28055691719055176,
      "memory(GiB)": 79.65,
      "step": 1225,
      "token_acc": 0.8886405033946018,
      "train_speed(iter/s)": 0.073824
    },
    {
      "epoch": 1.8512878360594096,
      "grad_norm": 0.2622053027153015,
      "learning_rate": 1.534473132597327e-05,
      "loss": 0.2793696880340576,
      "memory(GiB)": 79.65,
      "step": 1230,
      "token_acc": 0.8865800149018959,
      "train_speed(iter/s)": 0.073832
    },
    {
      "epoch": 1.8588080466253056,
      "grad_norm": 0.22427786886692047,
      "learning_rate": 1.5174415493624621e-05,
      "loss": 0.27779622077941896,
      "memory(GiB)": 79.65,
      "step": 1235,
      "token_acc": 0.8877466423478694,
      "train_speed(iter/s)": 0.073839
    },
    {
      "epoch": 1.8663282571912014,
      "grad_norm": 0.19943411648273468,
      "learning_rate": 1.500447043665546e-05,
      "loss": 0.2749001979827881,
      "memory(GiB)": 79.65,
      "step": 1240,
      "token_acc": 0.8862586605080831,
      "train_speed(iter/s)": 0.073846
    },
    {
      "epoch": 1.8738484677570972,
      "grad_norm": 0.23258428275585175,
      "learning_rate": 1.4834909212850393e-05,
      "loss": 0.27734966278076173,
      "memory(GiB)": 79.65,
      "step": 1245,
      "token_acc": 0.8847229074165434,
      "train_speed(iter/s)": 0.073854
    },
    {
      "epoch": 1.881368678322993,
      "grad_norm": 0.20784735679626465,
      "learning_rate": 1.4665744850502035e-05,
      "loss": 0.2754035949707031,
      "memory(GiB)": 79.65,
      "step": 1250,
      "token_acc": 0.8883543463835435,
      "train_speed(iter/s)": 0.073861
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.18474063277244568,
      "learning_rate": 1.4496990347410056e-05,
      "loss": 0.2744397401809692,
      "memory(GiB)": 79.65,
      "step": 1255,
      "token_acc": 0.8891737656107849,
      "train_speed(iter/s)": 0.073869
    },
    {
      "epoch": 1.8964090994547846,
      "grad_norm": 0.234056293964386,
      "learning_rate": 1.432865866988245e-05,
      "loss": 0.27312302589416504,
      "memory(GiB)": 79.65,
      "step": 1260,
      "token_acc": 0.8841318103377435,
      "train_speed(iter/s)": 0.073876
    },
    {
      "epoch": 1.9039293100206804,
      "grad_norm": 0.23731455206871033,
      "learning_rate": 1.4160762751739245e-05,
      "loss": 0.2745504379272461,
      "memory(GiB)": 79.65,
      "step": 1265,
      "token_acc": 0.8865546218487395,
      "train_speed(iter/s)": 0.073884
    },
    {
      "epoch": 1.9114495205865765,
      "grad_norm": 0.24919211864471436,
      "learning_rate": 1.399331549331878e-05,
      "loss": 0.27961087226867676,
      "memory(GiB)": 79.65,
      "step": 1270,
      "token_acc": 0.8842087875784606,
      "train_speed(iter/s)": 0.073891
    },
    {
      "epoch": 1.9189697311524723,
      "grad_norm": 0.24301506578922272,
      "learning_rate": 1.3826329760486462e-05,
      "loss": 0.2711665630340576,
      "memory(GiB)": 79.65,
      "step": 1275,
      "token_acc": 0.8934685058999502,
      "train_speed(iter/s)": 0.073898
    },
    {
      "epoch": 1.926489941718368,
      "grad_norm": 0.2331959307193756,
      "learning_rate": 1.3659818383646242e-05,
      "loss": 0.27373428344726564,
      "memory(GiB)": 79.65,
      "step": 1280,
      "token_acc": 0.8936294078509648,
      "train_speed(iter/s)": 0.073905
    },
    {
      "epoch": 1.934010152284264,
      "grad_norm": 0.2018069177865982,
      "learning_rate": 1.3493794156754744e-05,
      "loss": 0.27259933948516846,
      "memory(GiB)": 79.65,
      "step": 1285,
      "token_acc": 0.8890082644628099,
      "train_speed(iter/s)": 0.073912
    },
    {
      "epoch": 1.94153036285016,
      "grad_norm": 0.24646151065826416,
      "learning_rate": 1.3328269836338285e-05,
      "loss": 0.27399535179138185,
      "memory(GiB)": 79.65,
      "step": 1290,
      "token_acc": 0.8858642177763201,
      "train_speed(iter/s)": 0.073919
    },
    {
      "epoch": 1.9490505734160557,
      "grad_norm": 0.23400379717350006,
      "learning_rate": 1.316325814051271e-05,
      "loss": 0.2743144273757935,
      "memory(GiB)": 79.65,
      "step": 1295,
      "token_acc": 0.8924302788844621,
      "train_speed(iter/s)": 0.073926
    },
    {
      "epoch": 1.9565707839819515,
      "grad_norm": 0.19613775610923767,
      "learning_rate": 1.2998771748006153e-05,
      "loss": 0.2734565734863281,
      "memory(GiB)": 79.65,
      "step": 1300,
      "token_acc": 0.8851278655962923,
      "train_speed(iter/s)": 0.073933
    },
    {
      "epoch": 1.9640909945478473,
      "grad_norm": 0.19771267473697662,
      "learning_rate": 1.2834823297184932e-05,
      "loss": 0.27503745555877684,
      "memory(GiB)": 79.65,
      "step": 1305,
      "token_acc": 0.8902489799317179,
      "train_speed(iter/s)": 0.07394
    },
    {
      "epoch": 1.971611205113743,
      "grad_norm": 0.2214132398366928,
      "learning_rate": 1.267142538508241e-05,
      "loss": 0.27499420642852784,
      "memory(GiB)": 79.65,
      "step": 1310,
      "token_acc": 0.888298755186722,
      "train_speed(iter/s)": 0.073947
    },
    {
      "epoch": 1.979131415679639,
      "grad_norm": 0.2264689952135086,
      "learning_rate": 1.2508590566431149e-05,
      "loss": 0.2694978713989258,
      "memory(GiB)": 79.65,
      "step": 1315,
      "token_acc": 0.8929438352724534,
      "train_speed(iter/s)": 0.073953
    },
    {
      "epoch": 1.986651626245535,
      "grad_norm": 0.20919960737228394,
      "learning_rate": 1.2346331352698206e-05,
      "loss": 0.2705239295959473,
      "memory(GiB)": 79.65,
      "step": 1320,
      "token_acc": 0.8880026586905949,
      "train_speed(iter/s)": 0.07396
    },
    {
      "epoch": 1.9941718368114307,
      "grad_norm": 0.21414001286029816,
      "learning_rate": 1.2184660211123878e-05,
      "loss": 0.27195770740509034,
      "memory(GiB)": 79.65,
      "step": 1325,
      "token_acc": 0.8900908340214698,
      "train_speed(iter/s)": 0.073968
    },
    {
      "epoch": 2.0030080842263582,
      "grad_norm": 0.11679200083017349,
      "learning_rate": 1.2023589563763731e-05,
      "loss": 0.3252572059631348,
      "memory(GiB)": 79.65,
      "step": 1330,
      "token_acc": 0.8873622108368265,
      "train_speed(iter/s)": 0.073933
    },
    {
      "epoch": 2.010528294792254,
      "grad_norm": 0.15688148140907288,
      "learning_rate": 1.1863131786534146e-05,
      "loss": 0.26926493644714355,
      "memory(GiB)": 79.65,
      "step": 1335,
      "token_acc": 0.8880975769482645,
      "train_speed(iter/s)": 0.073939
    },
    {
      "epoch": 2.01804850535815,
      "grad_norm": 0.18938730657100677,
      "learning_rate": 1.1703299208261447e-05,
      "loss": 0.2670731782913208,
      "memory(GiB)": 79.65,
      "step": 1340,
      "token_acc": 0.8912466843501327,
      "train_speed(iter/s)": 0.073945
    },
    {
      "epoch": 2.025568715924046,
      "grad_norm": 0.16086824238300323,
      "learning_rate": 1.1544104109734578e-05,
      "loss": 0.2682967185974121,
      "memory(GiB)": 79.65,
      "step": 1345,
      "token_acc": 0.889979381443299,
      "train_speed(iter/s)": 0.073952
    },
    {
      "epoch": 2.033088926489942,
      "grad_norm": 0.18687959015369415,
      "learning_rate": 1.1385558722761525e-05,
      "loss": 0.2669731378555298,
      "memory(GiB)": 79.65,
      "step": 1350,
      "token_acc": 0.8855869242199108,
      "train_speed(iter/s)": 0.073958
    },
    {
      "epoch": 2.0406091370558377,
      "grad_norm": 0.16507050395011902,
      "learning_rate": 1.1227675229229453e-05,
      "loss": 0.2657780170440674,
      "memory(GiB)": 79.65,
      "step": 1355,
      "token_acc": 0.8913457556935818,
      "train_speed(iter/s)": 0.073962
    },
    {
      "epoch": 2.0481293476217335,
      "grad_norm": 0.1394694298505783,
      "learning_rate": 1.1070465760168746e-05,
      "loss": 0.2665443181991577,
      "memory(GiB)": 79.65,
      "step": 1360,
      "token_acc": 0.8893882646691635,
      "train_speed(iter/s)": 0.073967
    },
    {
      "epoch": 2.0556495581876293,
      "grad_norm": 0.16859273612499237,
      "learning_rate": 1.0913942394820882e-05,
      "loss": 0.2695469856262207,
      "memory(GiB)": 79.65,
      "step": 1365,
      "token_acc": 0.8895492818226844,
      "train_speed(iter/s)": 0.073973
    },
    {
      "epoch": 2.063169768753525,
      "grad_norm": 0.14011797308921814,
      "learning_rate": 1.0758117159710343e-05,
      "loss": 0.26610255241394043,
      "memory(GiB)": 79.65,
      "step": 1370,
      "token_acc": 0.8866694180767644,
      "train_speed(iter/s)": 0.07398
    },
    {
      "epoch": 2.070689979319421,
      "grad_norm": 0.15233328938484192,
      "learning_rate": 1.0603002027720544e-05,
      "loss": 0.26400294303894045,
      "memory(GiB)": 79.65,
      "step": 1375,
      "token_acc": 0.8919953403228491,
      "train_speed(iter/s)": 0.073986
    },
    {
      "epoch": 2.0782101898853167,
      "grad_norm": 0.12170787155628204,
      "learning_rate": 1.044860891717388e-05,
      "loss": 0.26834468841552733,
      "memory(GiB)": 79.65,
      "step": 1380,
      "token_acc": 0.8918291165491498,
      "train_speed(iter/s)": 0.073992
    },
    {
      "epoch": 2.0857304004512125,
      "grad_norm": 0.15005825459957123,
      "learning_rate": 1.029494969091601e-05,
      "loss": 0.26338939666748046,
      "memory(GiB)": 79.65,
      "step": 1385,
      "token_acc": 0.8917873601053324,
      "train_speed(iter/s)": 0.073998
    },
    {
      "epoch": 2.0932506110171083,
      "grad_norm": 0.15118637681007385,
      "learning_rate": 1.0142036155404322e-05,
      "loss": 0.2662912607192993,
      "memory(GiB)": 79.65,
      "step": 1390,
      "token_acc": 0.8956326801727001,
      "train_speed(iter/s)": 0.074004
    },
    {
      "epoch": 2.100770821583004,
      "grad_norm": 0.16032150387763977,
      "learning_rate": 9.989880059800832e-06,
      "loss": 0.26346986293792723,
      "memory(GiB)": 79.65,
      "step": 1395,
      "token_acc": 0.888176813737573,
      "train_speed(iter/s)": 0.074011
    },
    {
      "epoch": 2.1082910321489003,
      "grad_norm": 0.17994718253612518,
      "learning_rate": 9.838493095069418e-06,
      "loss": 0.26875948905944824,
      "memory(GiB)": 79.65,
      "step": 1400,
      "token_acc": 0.8856085526315789,
      "train_speed(iter/s)": 0.074016
    },
    {
      "epoch": 2.115811242714796,
      "grad_norm": 0.1501842737197876,
      "learning_rate": 9.687886893077498e-06,
      "loss": 0.26648333072662356,
      "memory(GiB)": 79.65,
      "step": 1405,
      "token_acc": 0.8911022576361222,
      "train_speed(iter/s)": 0.073772
    },
    {
      "epoch": 2.123331453280692,
      "grad_norm": 0.15215978026390076,
      "learning_rate": 9.538073025702361e-06,
      "loss": 0.26583385467529297,
      "memory(GiB)": 79.65,
      "step": 1410,
      "token_acc": 0.8889896159551673,
      "train_speed(iter/s)": 0.073779
    },
    {
      "epoch": 2.1308516638465878,
      "grad_norm": 0.164270281791687,
      "learning_rate": 9.389063003941991e-06,
      "loss": 0.2647554397583008,
      "memory(GiB)": 79.65,
      "step": 1415,
      "token_acc": 0.8924286305337195,
      "train_speed(iter/s)": 0.073785
    },
    {
      "epoch": 2.1383718744124836,
      "grad_norm": 0.1713220626115799,
      "learning_rate": 9.240868277030637e-06,
      "loss": 0.268373966217041,
      "memory(GiB)": 79.65,
      "step": 1420,
      "token_acc": 0.8879880873593646,
      "train_speed(iter/s)": 0.07379
    },
    {
      "epoch": 2.1458920849783794,
      "grad_norm": 0.15909770131111145,
      "learning_rate": 9.093500231559076e-06,
      "loss": 0.26570265293121337,
      "memory(GiB)": 79.65,
      "step": 1425,
      "token_acc": 0.8910612211084418,
      "train_speed(iter/s)": 0.073797
    },
    {
      "epoch": 2.153412295544275,
      "grad_norm": 0.16741392016410828,
      "learning_rate": 8.94697019059977e-06,
      "loss": 0.26313204765319825,
      "memory(GiB)": 79.65,
      "step": 1430,
      "token_acc": 0.8918027379185222,
      "train_speed(iter/s)": 0.073803
    },
    {
      "epoch": 2.160932506110171,
      "grad_norm": 0.1670864224433899,
      "learning_rate": 8.801289412836811e-06,
      "loss": 0.26590843200683595,
      "memory(GiB)": 79.65,
      "step": 1435,
      "token_acc": 0.8919275123558484,
      "train_speed(iter/s)": 0.07381
    },
    {
      "epoch": 2.1684527166760668,
      "grad_norm": 0.1373482048511505,
      "learning_rate": 8.656469091700882e-06,
      "loss": 0.2620683670043945,
      "memory(GiB)": 79.65,
      "step": 1440,
      "token_acc": 0.8905747505566092,
      "train_speed(iter/s)": 0.073816
    },
    {
      "epoch": 2.175972927241963,
      "grad_norm": 0.14378611743450165,
      "learning_rate": 8.512520354509196e-06,
      "loss": 0.26408073902130125,
      "memory(GiB)": 79.65,
      "step": 1445,
      "token_acc": 0.8888155724183437,
      "train_speed(iter/s)": 0.073823
    },
    {
      "epoch": 2.183493137807859,
      "grad_norm": 0.13253071904182434,
      "learning_rate": 8.369454261610516e-06,
      "loss": 0.26999673843383787,
      "memory(GiB)": 79.65,
      "step": 1450,
      "token_acc": 0.8936382536382537,
      "train_speed(iter/s)": 0.073829
    },
    {
      "epoch": 2.1910133483737546,
      "grad_norm": 0.14312045276165009,
      "learning_rate": 8.227281805535361e-06,
      "loss": 0.2614304542541504,
      "memory(GiB)": 79.65,
      "step": 1455,
      "token_acc": 0.8964057441686727,
      "train_speed(iter/s)": 0.073835
    },
    {
      "epoch": 2.1985335589396504,
      "grad_norm": 0.14327174425125122,
      "learning_rate": 8.086013910151334e-06,
      "loss": 0.2622921705245972,
      "memory(GiB)": 79.65,
      "step": 1460,
      "token_acc": 0.8909995852343426,
      "train_speed(iter/s)": 0.073841
    },
    {
      "epoch": 2.206053769505546,
      "grad_norm": 0.18833936750888824,
      "learning_rate": 7.94566142982385e-06,
      "loss": 0.264266037940979,
      "memory(GiB)": 79.65,
      "step": 1465,
      "token_acc": 0.8897156084656085,
      "train_speed(iter/s)": 0.073848
    },
    {
      "epoch": 2.213573980071442,
      "grad_norm": 0.15815502405166626,
      "learning_rate": 7.806235148582102e-06,
      "loss": 0.26674180030822753,
      "memory(GiB)": 79.65,
      "step": 1470,
      "token_acc": 0.8898137464974452,
      "train_speed(iter/s)": 0.073854
    },
    {
      "epoch": 2.221094190637338,
      "grad_norm": 0.12422753125429153,
      "learning_rate": 7.66774577929045e-06,
      "loss": 0.26320085525512693,
      "memory(GiB)": 79.65,
      "step": 1475,
      "token_acc": 0.8924223602484472,
      "train_speed(iter/s)": 0.073859
    },
    {
      "epoch": 2.2286144012032336,
      "grad_norm": 0.1535126268863678,
      "learning_rate": 7.530203962825331e-06,
      "loss": 0.26366519927978516,
      "memory(GiB)": 79.65,
      "step": 1480,
      "token_acc": 0.8965289256198347,
      "train_speed(iter/s)": 0.073865
    },
    {
      "epoch": 2.2361346117691294,
      "grad_norm": 0.16536694765090942,
      "learning_rate": 7.3936202672576485e-06,
      "loss": 0.2615305662155151,
      "memory(GiB)": 79.65,
      "step": 1485,
      "token_acc": 0.8907158279355589,
      "train_speed(iter/s)": 0.073871
    },
    {
      "epoch": 2.2436548223350252,
      "grad_norm": 0.18761932849884033,
      "learning_rate": 7.258005187040773e-06,
      "loss": 0.2668294906616211,
      "memory(GiB)": 79.65,
      "step": 1490,
      "token_acc": 0.8895537272500623,
      "train_speed(iter/s)": 0.073877
    },
    {
      "epoch": 2.251175032900921,
      "grad_norm": 0.16170427203178406,
      "learning_rate": 7.123369142204175e-06,
      "loss": 0.26272242069244384,
      "memory(GiB)": 79.65,
      "step": 1495,
      "token_acc": 0.8976952412535235,
      "train_speed(iter/s)": 0.073883
    },
    {
      "epoch": 2.258695243466817,
      "grad_norm": 0.152368426322937,
      "learning_rate": 6.989722477552836e-06,
      "loss": 0.2644169807434082,
      "memory(GiB)": 79.65,
      "step": 1500,
      "token_acc": 0.8905431256757881,
      "train_speed(iter/s)": 0.073889
    },
    {
      "epoch": 2.266215454032713,
      "grad_norm": 0.17155104875564575,
      "learning_rate": 6.857075461872382e-06,
      "loss": 0.2669036865234375,
      "memory(GiB)": 79.65,
      "step": 1505,
      "token_acc": 0.894046333969941,
      "train_speed(iter/s)": 0.073895
    },
    {
      "epoch": 2.273735664598609,
      "grad_norm": 0.13482429087162018,
      "learning_rate": 6.725438287140076e-06,
      "loss": 0.2606958389282227,
      "memory(GiB)": 79.65,
      "step": 1510,
      "token_acc": 0.8868266930627733,
      "train_speed(iter/s)": 0.073901
    },
    {
      "epoch": 2.2812558751645047,
      "grad_norm": 0.1460316777229309,
      "learning_rate": 6.594821067741719e-06,
      "loss": 0.26482319831848145,
      "memory(GiB)": 79.65,
      "step": 1515,
      "token_acc": 0.8907180675049636,
      "train_speed(iter/s)": 0.073907
    },
    {
      "epoch": 2.2887760857304005,
      "grad_norm": 0.16570395231246948,
      "learning_rate": 6.465233839694511e-06,
      "loss": 0.26358795166015625,
      "memory(GiB)": 79.65,
      "step": 1520,
      "token_acc": 0.8908941079522044,
      "train_speed(iter/s)": 0.073913
    },
    {
      "epoch": 2.2962962962962963,
      "grad_norm": 0.17693059146404266,
      "learning_rate": 6.3366865598759285e-06,
      "loss": 0.26563291549682616,
      "memory(GiB)": 79.65,
      "step": 1525,
      "token_acc": 0.8956694543346858,
      "train_speed(iter/s)": 0.073919
    },
    {
      "epoch": 2.303816506862192,
      "grad_norm": 0.17319715023040771,
      "learning_rate": 6.209189105258661e-06,
      "loss": 0.2657719373703003,
      "memory(GiB)": 79.65,
      "step": 1530,
      "token_acc": 0.8914614121510673,
      "train_speed(iter/s)": 0.073924
    },
    {
      "epoch": 2.311336717428088,
      "grad_norm": 0.15345337986946106,
      "learning_rate": 6.082751272151759e-06,
      "loss": 0.2649385452270508,
      "memory(GiB)": 79.65,
      "step": 1535,
      "token_acc": 0.8873251028806585,
      "train_speed(iter/s)": 0.07393
    },
    {
      "epoch": 2.3188569279939837,
      "grad_norm": 0.16597911715507507,
      "learning_rate": 5.957382775447897e-06,
      "loss": 0.2605454683303833,
      "memory(GiB)": 79.65,
      "step": 1540,
      "token_acc": 0.8902025630425796,
      "train_speed(iter/s)": 0.073936
    },
    {
      "epoch": 2.3263771385598795,
      "grad_norm": 0.13824202120304108,
      "learning_rate": 5.833093247876918e-06,
      "loss": 0.2636253356933594,
      "memory(GiB)": 79.65,
      "step": 1545,
      "token_acc": 0.8926768721555648,
      "train_speed(iter/s)": 0.073942
    },
    {
      "epoch": 2.3338973491257757,
      "grad_norm": 0.13943323493003845,
      "learning_rate": 5.709892239265733e-06,
      "loss": 0.26460967063903806,
      "memory(GiB)": 79.65,
      "step": 1550,
      "token_acc": 0.8934351398456304,
      "train_speed(iter/s)": 0.073948
    },
    {
      "epoch": 2.3414175596916715,
      "grad_norm": 0.12790897488594055,
      "learning_rate": 5.5877892158045335e-06,
      "loss": 0.26376543045043943,
      "memory(GiB)": 79.65,
      "step": 1555,
      "token_acc": 0.8898423962373133,
      "train_speed(iter/s)": 0.073953
    },
    {
      "epoch": 2.3489377702575673,
      "grad_norm": 0.1576036959886551,
      "learning_rate": 5.4667935593194635e-06,
      "loss": 0.26642694473266604,
      "memory(GiB)": 79.65,
      "step": 1560,
      "token_acc": 0.8944946452476573,
      "train_speed(iter/s)": 0.073959
    },
    {
      "epoch": 2.356457980823463,
      "grad_norm": 0.15614335238933563,
      "learning_rate": 5.346914566551746e-06,
      "loss": 0.25992136001586913,
      "memory(GiB)": 79.65,
      "step": 1565,
      "token_acc": 0.8944877967790138,
      "train_speed(iter/s)": 0.073964
    },
    {
      "epoch": 2.363978191389359,
      "grad_norm": 0.15744738280773163,
      "learning_rate": 5.228161448443394e-06,
      "loss": 0.26295444965362547,
      "memory(GiB)": 79.65,
      "step": 1570,
      "token_acc": 0.8909524200838196,
      "train_speed(iter/s)": 0.07397
    },
    {
      "epoch": 2.3714984019552547,
      "grad_norm": 0.16500192880630493,
      "learning_rate": 5.1105433294294605e-06,
      "loss": 0.26115808486938474,
      "memory(GiB)": 79.65,
      "step": 1575,
      "token_acc": 0.8928423665701283,
      "train_speed(iter/s)": 0.073975
    },
    {
      "epoch": 2.3790186125211505,
      "grad_norm": 0.1452684849500656,
      "learning_rate": 4.994069246736973e-06,
      "loss": 0.2584149122238159,
      "memory(GiB)": 79.65,
      "step": 1580,
      "token_acc": 0.897588864031817,
      "train_speed(iter/s)": 0.073981
    },
    {
      "epoch": 2.3865388230870463,
      "grad_norm": 0.15923888981342316,
      "learning_rate": 4.878748149690544e-06,
      "loss": 0.25870580673217775,
      "memory(GiB)": 79.65,
      "step": 1585,
      "token_acc": 0.8959456098167647,
      "train_speed(iter/s)": 0.073985
    },
    {
      "epoch": 2.394059033652942,
      "grad_norm": 0.19006413221359253,
      "learning_rate": 4.764588899024763e-06,
      "loss": 0.25927095413208007,
      "memory(GiB)": 79.65,
      "step": 1590,
      "token_acc": 0.8954526491447643,
      "train_speed(iter/s)": 0.073991
    },
    {
      "epoch": 2.401579244218838,
      "grad_norm": 0.18122485280036926,
      "learning_rate": 4.6516002662033645e-06,
      "loss": 0.2600280523300171,
      "memory(GiB)": 79.65,
      "step": 1595,
      "token_acc": 0.8928895114113866,
      "train_speed(iter/s)": 0.073995
    },
    {
      "epoch": 2.4090994547847338,
      "grad_norm": 0.13251742720603943,
      "learning_rate": 4.53979093274526e-06,
      "loss": 0.2628070592880249,
      "memory(GiB)": 79.65,
      "step": 1600,
      "token_acc": 0.8908462298320711,
      "train_speed(iter/s)": 0.074
    },
    {
      "epoch": 2.41661966535063,
      "grad_norm": 0.1347128301858902,
      "learning_rate": 4.429169489557528e-06,
      "loss": 0.2637340545654297,
      "memory(GiB)": 79.65,
      "step": 1605,
      "token_acc": 0.8926235615531087,
      "train_speed(iter/s)": 0.073799
    },
    {
      "epoch": 2.424139875916526,
      "grad_norm": 0.14901773631572723,
      "learning_rate": 4.3197444362752994e-06,
      "loss": 0.2581412553787231,
      "memory(GiB)": 79.65,
      "step": 1610,
      "token_acc": 0.8969149112622326,
      "train_speed(iter/s)": 0.073805
    },
    {
      "epoch": 2.4316600864824216,
      "grad_norm": 0.17272411286830902,
      "learning_rate": 4.211524180608672e-06,
      "loss": 0.2594130039215088,
      "memory(GiB)": 79.65,
      "step": 1615,
      "token_acc": 0.8946096809131051,
      "train_speed(iter/s)": 0.073811
    },
    {
      "epoch": 2.4391802970483174,
      "grad_norm": 0.14816541969776154,
      "learning_rate": 4.104517037696749e-06,
      "loss": 0.2576144695281982,
      "memory(GiB)": 79.65,
      "step": 1620,
      "token_acc": 0.8968891118916207,
      "train_speed(iter/s)": 0.073817
    },
    {
      "epoch": 2.446700507614213,
      "grad_norm": 0.14304767549037933,
      "learning_rate": 3.998731229468697e-06,
      "loss": 0.2579361915588379,
      "memory(GiB)": 79.65,
      "step": 1625,
      "token_acc": 0.895538448803907,
      "train_speed(iter/s)": 0.073823
    },
    {
      "epoch": 2.454220718180109,
      "grad_norm": 0.16488561034202576,
      "learning_rate": 3.8941748840120404e-06,
      "loss": 0.25913052558898925,
      "memory(GiB)": 79.65,
      "step": 1630,
      "token_acc": 0.8949146954586664,
      "train_speed(iter/s)": 0.073828
    },
    {
      "epoch": 2.461740928746005,
      "grad_norm": 0.1475977599620819,
      "learning_rate": 3.7908560349481072e-06,
      "loss": 0.2617921829223633,
      "memory(GiB)": 79.65,
      "step": 1635,
      "token_acc": 0.8957818421486241,
      "train_speed(iter/s)": 0.073834
    },
    {
      "epoch": 2.4692611393119006,
      "grad_norm": 0.1820904016494751,
      "learning_rate": 3.6887826208147968e-06,
      "loss": 0.262072229385376,
      "memory(GiB)": 79.65,
      "step": 1640,
      "token_acc": 0.8941467436108821,
      "train_speed(iter/s)": 0.07384
    },
    {
      "epoch": 2.4767813498777964,
      "grad_norm": 0.145757794380188,
      "learning_rate": 3.587962484456611e-06,
      "loss": 0.26173441410064696,
      "memory(GiB)": 79.65,
      "step": 1645,
      "token_acc": 0.8936417143330846,
      "train_speed(iter/s)": 0.073845
    },
    {
      "epoch": 2.4843015604436927,
      "grad_norm": 0.15350989997386932,
      "learning_rate": 3.488403372422011e-06,
      "loss": 0.2630826473236084,
      "memory(GiB)": 79.65,
      "step": 1650,
      "token_acc": 0.8907908814089376,
      "train_speed(iter/s)": 0.073851
    },
    {
      "epoch": 2.4918217710095885,
      "grad_norm": 0.13873544335365295,
      "learning_rate": 3.390112934368266e-06,
      "loss": 0.2567378759384155,
      "memory(GiB)": 79.65,
      "step": 1655,
      "token_acc": 0.8962406015037594,
      "train_speed(iter/s)": 0.073857
    },
    {
      "epoch": 2.4993419815754843,
      "grad_norm": 0.12694109976291656,
      "learning_rate": 3.2930987224736465e-06,
      "loss": 0.26199419498443605,
      "memory(GiB)": 79.65,
      "step": 1660,
      "token_acc": 0.8919701986754967,
      "train_speed(iter/s)": 0.073863
    },
    {
      "epoch": 2.50686219214138,
      "grad_norm": 0.16447237133979797,
      "learning_rate": 3.197368190857173e-06,
      "loss": 0.2622325897216797,
      "memory(GiB)": 79.65,
      "step": 1665,
      "token_acc": 0.8919255690314005,
      "train_speed(iter/s)": 0.073868
    },
    {
      "epoch": 2.514382402707276,
      "grad_norm": 0.1433202177286148,
      "learning_rate": 3.102928695005858e-06,
      "loss": 0.2609424591064453,
      "memory(GiB)": 79.65,
      "step": 1670,
      "token_acc": 0.89528060170262,
      "train_speed(iter/s)": 0.073873
    },
    {
      "epoch": 2.5219026132731717,
      "grad_norm": 0.16500116884708405,
      "learning_rate": 3.0097874912095636e-06,
      "loss": 0.2639314651489258,
      "memory(GiB)": 79.65,
      "step": 1675,
      "token_acc": 0.8930817610062893,
      "train_speed(iter/s)": 0.073879
    },
    {
      "epoch": 2.5294228238390675,
      "grad_norm": 0.16409310698509216,
      "learning_rate": 2.9179517360034526e-06,
      "loss": 0.25854926109313964,
      "memory(GiB)": 79.65,
      "step": 1680,
      "token_acc": 0.8943049400951912,
      "train_speed(iter/s)": 0.073884
    },
    {
      "epoch": 2.5369430344049633,
      "grad_norm": 0.14108499884605408,
      "learning_rate": 2.827428485618102e-06,
      "loss": 0.25849237442016604,
      "memory(GiB)": 79.65,
      "step": 1685,
      "token_acc": 0.8958523056544416,
      "train_speed(iter/s)": 0.07389
    },
    {
      "epoch": 2.544463244970859,
      "grad_norm": 0.16303665935993195,
      "learning_rate": 2.738224695437357e-06,
      "loss": 0.259888219833374,
      "memory(GiB)": 79.65,
      "step": 1690,
      "token_acc": 0.892691415313225,
      "train_speed(iter/s)": 0.073895
    },
    {
      "epoch": 2.551983455536755,
      "grad_norm": 0.15851441025733948,
      "learning_rate": 2.650347219463907e-06,
      "loss": 0.2603473663330078,
      "memory(GiB)": 79.65,
      "step": 1695,
      "token_acc": 0.8976240100041684,
      "train_speed(iter/s)": 0.0739
    },
    {
      "epoch": 2.5595036661026507,
      "grad_norm": 0.16855500638484955,
      "learning_rate": 2.563802809792646e-06,
      "loss": 0.26097941398620605,
      "memory(GiB)": 79.65,
      "step": 1700,
      "token_acc": 0.8949639094001494,
      "train_speed(iter/s)": 0.073905
    },
    {
      "epoch": 2.5670238766685465,
      "grad_norm": 0.16529929637908936,
      "learning_rate": 2.4785981160918703e-06,
      "loss": 0.2607457637786865,
      "memory(GiB)": 79.65,
      "step": 1705,
      "token_acc": 0.8880922950144211,
      "train_speed(iter/s)": 0.07391
    },
    {
      "epoch": 2.5745440872344427,
      "grad_norm": 0.18323875963687897,
      "learning_rate": 2.3947396850923664e-06,
      "loss": 0.2571398735046387,
      "memory(GiB)": 79.65,
      "step": 1710,
      "token_acc": 0.8893852899809713,
      "train_speed(iter/s)": 0.073915
    },
    {
      "epoch": 2.5820642978003385,
      "grad_norm": 0.13609886169433594,
      "learning_rate": 2.312233960084389e-06,
      "loss": 0.26127750873565675,
      "memory(GiB)": 79.65,
      "step": 1715,
      "token_acc": 0.8872211286089239,
      "train_speed(iter/s)": 0.07392
    },
    {
      "epoch": 2.5895845083662343,
      "grad_norm": 0.1616513431072235,
      "learning_rate": 2.231087280422557e-06,
      "loss": 0.26447813510894774,
      "memory(GiB)": 79.65,
      "step": 1720,
      "token_acc": 0.8936523762047192,
      "train_speed(iter/s)": 0.073925
    },
    {
      "epoch": 2.59710471893213,
      "grad_norm": 0.13646076619625092,
      "learning_rate": 2.151305881038814e-06,
      "loss": 0.25830256938934326,
      "memory(GiB)": 79.65,
      "step": 1725,
      "token_acc": 0.8933750823994726,
      "train_speed(iter/s)": 0.07393
    },
    {
      "epoch": 2.604624929498026,
      "grad_norm": 0.16205361485481262,
      "learning_rate": 2.0728958919633337e-06,
      "loss": 0.2566520690917969,
      "memory(GiB)": 79.65,
      "step": 1730,
      "token_acc": 0.8940786749482401,
      "train_speed(iter/s)": 0.073936
    },
    {
      "epoch": 2.6121451400639217,
      "grad_norm": 0.1583632230758667,
      "learning_rate": 1.9958633378535277e-06,
      "loss": 0.2593334197998047,
      "memory(GiB)": 79.65,
      "step": 1735,
      "token_acc": 0.8968471840945013,
      "train_speed(iter/s)": 0.073941
    },
    {
      "epoch": 2.6196653506298175,
      "grad_norm": 0.16150888800621033,
      "learning_rate": 1.9202141375311335e-06,
      "loss": 0.25293402671813964,
      "memory(GiB)": 79.65,
      "step": 1740,
      "token_acc": 0.8940237719225335,
      "train_speed(iter/s)": 0.073946
    },
    {
      "epoch": 2.6271855611957133,
      "grad_norm": 0.15309233963489532,
      "learning_rate": 1.8459541035274453e-06,
      "loss": 0.2575088024139404,
      "memory(GiB)": 79.65,
      "step": 1745,
      "token_acc": 0.8925674557192518,
      "train_speed(iter/s)": 0.073951
    },
    {
      "epoch": 2.6347057717616096,
      "grad_norm": 0.15103407204151154,
      "learning_rate": 1.7730889416367115e-06,
      "loss": 0.2573860645294189,
      "memory(GiB)": 79.65,
      "step": 1750,
      "token_acc": 0.8916528925619834,
      "train_speed(iter/s)": 0.073956
    },
    {
      "epoch": 2.6422259823275054,
      "grad_norm": 0.17293822765350342,
      "learning_rate": 1.7016242504777048e-06,
      "loss": 0.2626305103302002,
      "memory(GiB)": 79.65,
      "step": 1755,
      "token_acc": 0.8934236615994713,
      "train_speed(iter/s)": 0.073961
    },
    {
      "epoch": 2.649746192893401,
      "grad_norm": 0.17097805440425873,
      "learning_rate": 1.6315655210635784e-06,
      "loss": 0.26284041404724123,
      "memory(GiB)": 79.65,
      "step": 1760,
      "token_acc": 0.8951279933938894,
      "train_speed(iter/s)": 0.073966
    },
    {
      "epoch": 2.657266403459297,
      "grad_norm": 0.139834925532341,
      "learning_rate": 1.5629181363799517e-06,
      "loss": 0.2600309610366821,
      "memory(GiB)": 79.65,
      "step": 1765,
      "token_acc": 0.8978144811478145,
      "train_speed(iter/s)": 0.07397
    },
    {
      "epoch": 2.664786614025193,
      "grad_norm": 0.1551225632429123,
      "learning_rate": 1.4956873709713016e-06,
      "loss": 0.2573253870010376,
      "memory(GiB)": 79.65,
      "step": 1770,
      "token_acc": 0.8932351968243467,
      "train_speed(iter/s)": 0.073975
    },
    {
      "epoch": 2.6723068245910886,
      "grad_norm": 0.15004055202007294,
      "learning_rate": 1.4298783905356906e-06,
      "loss": 0.2537161588668823,
      "memory(GiB)": 79.65,
      "step": 1775,
      "token_acc": 0.8961599071079042,
      "train_speed(iter/s)": 0.07398
    },
    {
      "epoch": 2.6798270351569844,
      "grad_norm": 0.16113132238388062,
      "learning_rate": 1.3654962515278692e-06,
      "loss": 0.2536770343780518,
      "memory(GiB)": 79.65,
      "step": 1780,
      "token_acc": 0.8981034769589087,
      "train_speed(iter/s)": 0.073985
    },
    {
      "epoch": 2.68734724572288,
      "grad_norm": 0.16579777002334595,
      "learning_rate": 1.302545900770762e-06,
      "loss": 0.2598405361175537,
      "memory(GiB)": 79.65,
      "step": 1785,
      "token_acc": 0.8957157090667552,
      "train_speed(iter/s)": 0.07399
    },
    {
      "epoch": 2.694867456288776,
      "grad_norm": 0.19123908877372742,
      "learning_rate": 1.2410321750753518e-06,
      "loss": 0.2594654321670532,
      "memory(GiB)": 79.65,
      "step": 1790,
      "token_acc": 0.891751728679618,
      "train_speed(iter/s)": 0.073995
    },
    {
      "epoch": 2.702387666854672,
      "grad_norm": 0.1508457064628601,
      "learning_rate": 1.180959800869077e-06,
      "loss": 0.2609714508056641,
      "memory(GiB)": 79.65,
      "step": 1795,
      "token_acc": 0.889967637540453,
      "train_speed(iter/s)": 0.073999
    },
    {
      "epoch": 2.7099078774205676,
      "grad_norm": 0.1396402269601822,
      "learning_rate": 1.1223333938326486e-06,
      "loss": 0.2563958168029785,
      "memory(GiB)": 79.65,
      "step": 1800,
      "token_acc": 0.8946888723175077,
      "train_speed(iter/s)": 0.074004
    }
  ],
  "logging_steps": 5,
  "max_steps": 1992,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.4505754714759496e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}